Hypercholesterolemia and tissue-specific differential mRNA expression of type-1 5 ’-iodothyronine deiodinase under different selenium status in rats

Type-1 5’-iodothyronine deiodinase (5’-DI) is responsible for conversion of T4 to T3. Selenium (Se) is an integral part of this enzyme. Keeping in view the strong association between atherosclerosis and hypothyroidism, the present study examined the behavior of 5’-DI in liver, aorta and thyroid during hypercholesterolemia following different Se status, i.e., Se deficiency (0.02ppm), adequate (0.2ppm) and excess dose (1ppm) in SD male rats. Animals were fed a control or high-cholesterol diet (2%) for 1 and 2 months. 5’-DI activity and mRNA expression was measured by RIA and RT-PCR respectively. In liver and aorta, 5’-DI expression significantly decreased with the Se-deficient and the high-cholesterol diet. The trend was opposite in thyroid, i.e., mRNA expression increased significantly during selenium deficiency and with a high-cholesterol feeding. But with 1ppm Se supplementation, the 5’-DI expression increased in all the three tissues. The present study indicates that hypercholesterolemia along with selenium deficiency is coresponsible for differential regulation of 5’-DI enzyme in thyroidal vs. extrathyroidal tissues. Distinct regulation of 5’-DI in the thyroid reflects the clinical importance of this selenoprotein during hypercholesterolemia as this enzyme is essential for T3 production, which further has a vital role in the maintenance of lipid metabolism. Key terms: selenium, hypercholesterolemia, 5’-DI, cholesterol, RT-PCR. Corresponding author: Mohinder P Bansal, Department of Biophysics, Panjab University, Chandigarh-160014, India, Tel.: (91-172) 253-4120, Fax: (91-172) 253-4118, E-mail: mpbansal@pu.ac.in Received: October 10, 2005. In revised form: March 10, 2006. Accepted: April 19, 2006 INTRODUCTION Type-1 5’-iodothyronine deiodinase (5’-DI) has a pivotal role in controlling the supply of T3 (the biologically active form of thyroid hormone) to the tissues. Under normal circumstances, more than 80% of plasma T3 is derived from 5’-monodeiodination of T4 (Vander Geyten et al., 2005), mainly in the liver and kidney, where this reaction is catalyzed by 5’-DI (Alvarez et al., 2005). Thyroid also express 5’-DI activity, and studies have shown that it is identical to the hepatic 5’-DI (Green, 1978). The activity of 5’-DI has been shown to be decreased in a hypothyroid state and elevated during hyperthyroidism (Wassen et al., 2004). Selenium (Se), an essential element and potent antioxidant, has an important role in the control of thyroid hormone metabolism. Evidence from animal, clinical and in vitro cell culture studies suggested a clear Sedependent expression of 5’-DI varying with Se availability (Korhle, 1999). As selenium is an integral part of this enzyme (Behne et al., 1990; Berry et al., 1991), 5’-DI activity rapidly decreases with Se deficiency (Backett et al., 1987). Hypercholesterolemia has been shown to induce hypothyroidism (Wojcicki et al., 1991; Frank et al., 2004). Thyroid hormones (T3 and T4) have a vital role in the maintenance of normal lipid metabolism, and abnormal thyroid status can lead to biochemical and pathological changes that are potentially injurious to the cardiovascular system (Pingitore et al., 2005). Studies have demonstrated that almost 10% of asymptomatic hypercholesterolemic patients have subclinical hypothyroidism DHINGRA & BANSAL Biol Res 39, 2006, 307-319 308 (Michalopoulou et al., 1998). Thyroid hormone replacement therapy improved the cardiac function in several patients (Siegmund et al., 2004). Several studies indicate that Se deficiency is associated with hypercholesterolemia (Kang et al., 2000; Lee et al., 2003). Keeping in view the important role of Se in hypercholesterolemia and its association with 5’-DI, the purpose of present study was to explore the response of 5’-DI enzyme in liver, aorta and thyroid under different Se status during experimental hypercholesterolemia. MATERIALS AND METHODS


INTRODUCTION
Type-1 5'-iodothyronine deiodinase (5'-DI) has a pivotal role in controlling the supply of T 3 (the biologically active form of thyroid hormone) to the tissues.Under normal circumstances, more than 80% of plasma T 3 is derived from 5'-monodeiodination of T 4 (Vander Geyten et al., 2005), mainly in the liver and kidney, where this reaction is catalyzed by 5'-DI (Alvarez et al., 2005).Thyroid also express 5'-DI activity, and studies have shown that it is identical to the hepatic 5'-DI (Green, 1978).The activity of 5'-DI has been shown to be decreased in a hypothyroid state and elevated during hyperthyroidism (Wassen et al., 2004).
Selenium (Se), an essential element and potent antioxidant, has an important role in the control of thyroid hormone metabolism.
Evidence from animal, clinical and in vitro cell culture studies suggested a clear Sedependent expression of 5'-DI varying with Se availability (Korhle, 1999).As selenium is an integral part of this enzyme (Behne et al., 1990;Berry et al., 1991), 5'-DI activity rapidly decreases with Se deficiency (Backett et al., 1987).
Hypercholesterolemia has been shown to induce hypothyroidism (Wojcicki et al., 1991;Frank et al., 2004).Thyroid hormones (T 3 and T 4 ) have a vital role in the maintenance of normal lipid metabolism, and abnormal thyroid status can lead to biochemical and pathological changes that are potentially injurious to the cardiovascular system (Pingitore et al., 2005).Studies have demonstrated that almost 10% of asymptomatic hypercholesterolemic patients have subclinical hypothyroidism (Michalopoulou et al., 1998).Thyroid hormone replacement therapy improved the cardiac function in several patients (Siegmund et al., 2004).Several studies indicate that Se deficiency is associated with hypercholesterolemia (Kang et al., 2000;Lee et al., 2003).Keeping in view the important role of Se in hypercholesterolemia and its association with 5'-DI, the purpose of present study was to explore the response of 5'-DI enzyme in liver, aorta and thyroid under different Se status during experimental hypercholesterolemia.

Animals
Male Sprauge-Dawley rats (100g-body weight) were used in the present study.Animals were obtained from the Central Animal House, Panjab University, Chandigarh.

Treatment Protocol
Animals were acclimatized to the laboratory animal room and initially divided into three groups: group I (Se-deficient diet fed); group II (Se-adequate diet fed); and group III (Seexcess diet fed).Feed and water were given ad libitum.This Se diet was given to the animals for 10 days in order to achieve the required Se status.The animals in these three groups were further divided into two each, viz.: group Ia (Se-deficient control), group Ib (Se-deficient + high-cholesterol diet fed); group IIa (Se-adequate control), group IIb (Se-adequate + high-cholesterol diet fed); group IIIa (Se-excess control), group IIIb (Seexcess + high-cholesterol diet fed).Treatment protocol was for 1 and 2 months.
Se-supplemented diets: Se-adequate and Se-excess diets were prepared by supplementing the Se-deficient diet with 0.2 ppm and 1ppm of Se as sodium selenite (Sigma Chemicals).
High-cholesterol diet (HCD): 2% of cholesterol (Loba-Chemie, India) was added to the respective diets of the HCD groups.
After completion of the diet feeding schedule, the rats were kept on fasting for 10hrs, anesthetized and exsanguinated.Serum and tissue (liver, aorta and thyroid) samples were collected from each animal.Tissues were snap frozen in liquid nitrogen.Serum total cholesterol and triglycerides levels were estimated by enzymatic colorimetric kits obtained from HUMAN Gmbh (Germany).Various parameters were evaluated as detailed below.

Selenium estimation
Selenium level was estimated by fluorimetric method (Hasunuma et al., 1982), based on the principle that Se content in serum or tissue on acid digestion is converted to selenous acid.The reaction between selenous acid and aromatic-odiamines, such as 2,3-diaminonaphthalene (DAN), leads to the formation of 4,5benzopiazselenol, which displays brilliant lime-green fluorescence when excited at 366nm in cyclohexane.Fluorescence emission in extracted cyclohexane was read on a fluorescence spectrophotometer using 366nm as the excitation wavelength and 520nm as the emission wavelength.

Se-dependent glutathione peroxidase activity
Glutathione peroxidase (GSH-Px) activity was assayed by the coupled enzyme procedure with glutathione reductase, using H 2 O 2 as substrate (Paglia and Valentine, 1967).The assay was carried out in the postmitochondrial fraction (PMF) of liver as already published by the authors (Dhingra et al., 2003) and the activity expressed as µmoles of NADPH oxidized/min/mg protein.
Total protein estimation was done in all the samples (Lowry et al., 1951).

RNA isolation
Total RNA from tissues was extracted using TRI REAGENT (Molecular Research Center, Inc., Ohio, USA).The integrity and size distribution (quality) of RNA was examined by formaldehyde agarose gel electrophoresis.

RT-PCR
2µg of total RNA template from the different groups after treatment with DNase I (Ambion) was used in RT-PCR reaction.To the reaction mixture, we added 10µl of 5X QIAGEN One Step RT-PCR buffer (2.5mM MgCl2 as final concentration), 2µl of dNTP mix (10mM of each dNTP), 5µl of each coding (+) and noncoding (-) gene-specific primers (from 10µM stock), 2µl QIAGEN One Step RT-PCR Enzyme Mix, 1µl RNase inhibitor (1U/µl) and, finally, 25µl of PCRgrade RNase-free water (provided in the kit) to reach a total volume of 50µl, which was mixed gently by vortex and centrifuged.The PCR reaction was performed in the thermal cycler (Techne Ltd., England) under the following conditions: the RT reaction was performed at 50ºC for 50 min; initial PCR activation was done at 95ºC for 15 min, followed by 35 cycles of 94ºC (denaturation) for 45 sec, 58.8ºC (annealing) for 45 sec, 72ºC (extension) for 1 min.Finally, the reaction mixture was incubated at 72ºC for 10 min to extend any incomplete single strands.
The optimal oligonucleotide primer pair for 5'-DI was selected with the aid of software Gene Runner (Informax Ltd., USA) and primer sequences for ß-actin were taken from the literature (Kimura et al., 1998).The primer sequence (5' to 3') for rat 5'-DI gene coding (+) strand was TCTGGGATTTCATTCAAGGC, the noncoding strand was TAGAGCCTCTCAGGCAGAGC.For rat ß-actin, the gene coding (+) strand was AGAGCTATGAGCTGCCTGAC, and the noncoding (-) strand was CTGCATCCTGTCAGCCTACG.
The lengths of the RT-PCR products were 346bp and 236bp respectively for 5'-DI and ß-actin.
Final PCR products were analyzed on 1.5% agarose gel electrophoresis using 10 mM TE buffer.5µl of PCR product was used from each tube.Densitometric analysis of the bands was done by UviBandMap software (Uvitech, England).
The above-described RT-PCR reactions for 5'-DI and ß-actin were determined on the basis of a series of experiments to test the reaction conditions, such as amount of RNA linearity, cycle linearity, primer concentrations etc.The final RT-PCR conditions were shown to be in the linear range of amplification for both the genes.

Statistical analysis
Data is expressed as mean ± SD.Difference between different groups was tested using student's t-test for unpaired values.

Selenium levels
Selenium levels in the serum and liver decreased significantly (p<0.001) in Sedeficient groups (Ia and Ib) and increased in 1ppm Se-supplemented diet fed groups (IIIa and IIIb) in comparison to the respective adequate groups (IIa and IIb).Significant decrease (p<0.001) in the Se level was observed in HCD-fed groups as compared to respective controls in all the three Se status, i.e., deficient, adequate and excess groups.In deficient groups (Ia and Ib) and in HCD fed adequate group, Se level decreased significantly (p<0.001),whereas in 1ppm Se supplemented groups, the level increased significantly (p<0.001) after 2 months in comparison to 1-month data (Fig. 1).

GSH-Px activity
GSH-Px activity in liver decreased significantly (p<0.001) in Se deficiency (Ia and Ib), and it increased with 1ppm Se supplementation (IIIa and IIIb) in comparison to the respective adequate groups (IIa and IIb).With HCD feeding, significant increase (p<0.001) was observed in all the groups in comparison to respective controls.In the Se-deficient control group, GSH-Px level decreased, whereas in HCD-supplemented Se-deficient and Se-adequate as well as Se-excess fed groups, the level increased significantly (p<0.001) after 2 months as compared to 1month data (Fig. 1).

Lipid profile
In all the three Se status groups on HCD feeding, significant increase (p<0.001) in cholesterol and triglycerides concentration was observed in comparison to the respective control groups.In Se-deficient groups (Ia and Ib), lipid level increased, and with 1ppm Se supplementation, it decreased significantly (p<0.001) in comparison to respective adequate groups (IIa and IIb).In both the Se-deficient groups (Ia and Ib) and in HCD-fed adequate group, lipid levels increased significantly (p<0.001),whereas they decreased with Se supplementation after 2 months in comparison to 1 month (Table 1).

T3 and T4 levels
Levels of T 3 decreased and T 4 increased significantly (p<0.001) with HCD feeding in comparison to respective controls in all the three Se status groups.In Se deficiency (Ia and Ib), T 3 decreased and T 4 level increased in comparison to respective adequate groups, whereas the reverse trend was observed with 1ppm Se supplementation, i.e., the T 3 level increased and the T 4 level decreased significantly.In both the Se-deficient groups and in the HCD-fed adequate group (IIb) T 3 decreased and T 4 increased significantly, and in the Se-supplemented groups, T 3 increased and T 4 decreased after 2 months in comparison to 1 month (Table 1).

Type-I iodothyronine deiodinase (5'-DI) activity
5'-DI activity was estimated in liver, aorta, and thyroid.The activity decreased significantly (p<0.001) in HCD-fed animals in comparison to controls in liver and aorta.In Se deficiency, there was a significant (p<0.001)decrease in the activity in comparison to respective adequate groups.The trend was just opposite in thyroid in Se deficiency as well with HCD feeding: the activity increased significantly (p<0.001).With 1ppm Se supplementation, the activity increased in all the tissues.The 5'-DI activity decreased in Se-deficient groups (Ia and Ib), as well as in HCD-fed, adequate group, whereas it increased in Sesupplemented groups after 2 months in comparison to 1-month data in liver and aorta.In thyroid, the activity increased in all the groups (except IIa) after 2 months in comparison to 1-month data (Table 2).

Type-I iodothyronine deiodinase (5'-DI) mRNA expression
RT-PCR products of expected size, i.e., 346bp and 236bp, were obtained for 5'-DI and ß-actin, respectively.mRNA expression in liver and aorta decreased (p<0.001) in Se-deficient groups (Figs. 2  and 3) in comparison to adequate groups (IIa and IIb).With HCD feeding, the mRNA expression decreased (p<0.001) in liver and aorta in all the selenium status groups (Figs. 2 and 3).However, as with enzyme activity, in thyroid tissue the trend was the opposite, i.e., mRNA expression increased in selenium deficiency, as well as with HCD feeding (p<0.001) (Fig. 4).With selenium supplementation, the mRNA expression increased significantly (p<0.001) in all the tissues.The 5'-DI mRNA expression decreased in Sedeficient groups (Ia and Ib) as well as in HCD-fed adequate group, whereas it increased in Se-supplemented groups after 2 months in comparison to 1-month data in liver and aorta.In thyroid, the expression increased in all the groups (except IIa), after 2 months in comparison to 1 month (Figs. 2, 3 and 4).Type-I 5'-iodothyronine deiodinase (5'-DI) activity in liver, aorta, and thyroid after 1 and 2 months of control and high-cholesterol diet (HCD) feeding schedule.
(Data is represented as mean ± SD from 6 observations)  DISCUSSION Present data demonstrate that HCD feeding leads to Se depletion in rats (Lee et al., 2003;Dhingra et al., 2005).In Se deficiency, significant increase in cholesterol and triglyceride level was observed in comparison to the animals fed adequate Se after 1 and 2 months (Table 1).
Low Se levels are associated with increased platelet aggregation and thromboxane A2 production along with decreased prostacyclin production, all of which may be linked with cardiovascular disease (Huang et al., 2002).With 1ppm Se supplementation, lipid levels decreased significantly in comparison to groups fed adequate Se (Kang et al., 2000).The Se supplementation leads to an increase in HDL cholesterol fraction (Wojcicki et al., 1991).HDL cholesterol may downregulate the total cholesterol via reverse cholesterol transport to the liver, i.e., HDL fraction increases the cholesterol elimination from tissues including smooth muscle cells in the aorta wall and facilitates cholesterol transport to the liver, thus preventing its deposition and the formation of atheromatous plaque (Pelton et al., 2005).
The present results demonstrated that in Sedeficient groups, hepatic glutathione peroxidase (GSH-Px) activity decreased significantly in comparison to adequate groups (Fig. 1).In the Se-deficient control group, after 2 months the GSH-Px activity decreased in comparison to 1 month, so these observations  confirm the Se deficiency (Arthur et al., 1993), which is associated with decreased GSH-Px activity.With HCD feeding, GSH-Px activity increased in all the groups.Further, it can be interpreted from our results that with cholesterol-rich diet feeding in all the Se status groups, GSH-Px activity increased after 2 months in comparison to 1-month data.This finding can be attributed to the increased lipoperoxidative stress associated with the HCD feeding, which concurs with the literature reporting elevation of GSH-Px activity as associated with a minor increase in oxidative stress (Kang et al., 2000).This increase in the Se-dependent GSH-Px with HCD feeding explains the decrease in the Se levels during hypercholesterolemia, as observed in the present study.The levels of T 3 decreased and T4 increased significantly in Se deficiency as well as with HCD feeding (Table 1).This finding is probably due to the decreased conversion of T 4 to T 3 in the liver due to decreased 5'-DI expression during Se deficiency (Arthur el al., 1991;Dhingra et al., 2004).As observed in Se deficiency, increased plasma T 4 levels usually upregulate the hepatic 5'-DI expression (Berry et al., 1990).In the current study, although plasma T 4 levels are increased in Se deficiency, hepatic 5'-DI expression is diminished, presumably because the hepatic stores of Se are insufficient to allow the synthesis of 5'-DI.However, the trend was the opposite with Se supplementation up to 1ppm, i.e., T 3 levels increased and T 4 levels decreased; probably, with Se supplementation, increase in 5'-DI expression upregulated the T 4 to T 3 conversion.
With the selenium-deficient diet, 5'-DI activity as well as mRNA expression in liver and aorta decreased (Bermano et al., 1995) in comparison to the Se-adequate groups (Table 2; Figs. 2 and 3).On highcholesterol diet feeding, 5'-DI activity as well as mRNA expression decreased in liver and aorta.This finding could be due to the fact that hypercholesterolemia might have led to depletion in the Se pool, which is needed for normal 5'-DI expression.Another reason for this finding could be that hypercholesterolemia might be leading to hypothyroidism, which in turn downregulates the 5'-DI levels in liver and other peripheral tissues (Verhoelst et al., 2004).Furthermore, we have found a significant increase in 5'-DI activity as well as mRNA expression in liver and aorta with 1ppm Se supplementation.In cell culture experiments, Gross et al. (1995) have shown the clear dependence of 5'-DI expression on Se supply: enzyme activity as well as mRNA levels rapidly increased with the increase in Se concentration.
In contrast to liver and aorta, thyroidal 5'-DI activity and mRNA expression increased significantly with the highcholesterol diet feeding as well as in Se deficiency (Table 2; Fig. 4).This increased expression of thyroidal 5'-DI in Se deficiency confirms that thyroid is a higher priority tissue than liver and aorta for Se when intake of the element is very low, which is consistent with the fact that thyroid has the ability to retain a significant pool of trace element in Se deficiency (Behne et al., 1988).With 1ppm Se supplementation, 5'-DI activity and expression were further increased in thyroid.Beech et al. (1995) demonstrated that human thyrocytes grown in primary culture in a Se-free medium were able to retain the trace element to maintain the 5'-DI enzyme activity (the expression was further increased by the addition of Se to the diet).The increased 5'-DI expression in thyroid during a hypercholesterolemic state suggests that there is some compensatory mechanism that might get activated when 5'-DI expression is downregulated in other peripheral tissues (liver and aorta).So, this tissue-specific differential behavior of 5'-DI might be maintaining the T 3 level in the body during hypercholesterolemia. T 3 is involved at the transcriptional level in the cardiovascular system.T 3 enters into the cardiomyocytes through T 3 -binding nuclear receptors and interacts with specific transcriptional activators to modify the transcription rate of specific target genes (Brent, 1994).Thyroid hormones also decrease peripheral vascular resistance by promoting relaxation in vascular smoothmuscle cells (Park et al., 1997).Napoli et al. (2001) reported that thyroid hormones exert profound effects on the vascular system by improving both endotheliumdependent and -independent mechanisms.
The present data indicate that hypercholesterolemia along with the Se deficiency is co-responsible for tissuespecific differential expression of 5'-DI enzyme, i.e., the expression decreased in liver and aorta whereas it increased in thyroid.Until now, no direct association between 5'-DI behavior and hypercholesterolemia has been observed.Thus, longitudinal studies to clarify the potential association between 5'-DI expression and hypercholesterolemia under different Se status must be performed.
Further studies must be undertaken to explore the therapeutic role of Se supplementation in hypercholesterolemia through its dependent enzyme 5'-DI.

TABLE 1 Total
A p<0.05, AA p<0.01,BBB p<0.001 comparison between groups IIb and IIIb.