Skip to main content
NCBI home page
Endocrinology logoLink to Endocrinology
. 2008 Nov 26;150(4):1976–1983. doi: 10.1210/en.2008-0938

Type 2 Deiodinase Expression Is Induced by Peroxisomal Proliferator-Activated Receptor-γ Agonists in Skeletal Myocytes

Renata Grozovsky 1, Scott Ribich 1, Matthew L Rosene 1, Michelle A Mulcahey 1, Stephen A Huang 1, Mary Elizabeth Patti 1, Antonio C Bianco 1, Brian W Kim 1
PMCID: PMC2659265  PMID: 19036883

Abstract

The thyroid hormone activating type 2 deiodinase (D2) is known to play a role in brown adipose tissue-mediated adaptive thermogenesis in rodents, but the finding of D2 in skeletal muscle raises the possibility of a broader metabolic role. In the current study, we examined the regulation of the D2 pathway in primary skeletal muscle myoblasts taken from both humans and mice. We found that pioglitazone treatment led to a 1.6- to 1.9-fold increase in primary human skeletal myocyte D2 activity; this effect was seen with other peroxisomal proliferator-activated receptor-γ agonists. D2 activity in primary murine skeletal myotubes increased 2.8-fold in response to 5 μm pioglitazone and 1.6-fold in response to 5 nm insulin and increased in a dose-dependent manner in response to lithocholic acid (maximum response at 25 μm was ∼3.8-fold). We compared Akt phosphorylation in primary myotubes derived from wild-type and D2 knockout (D2KO) mice: phospho-Akt was reduced by 50% in the D2KO muscle after 1 nm insulin exposure. Expression of T3-responsive muscle genes via quantitative RT-PCR suggests that D2KO cells have decreased thyroid hormone signaling, which could contribute to the abnormalities in insulin signaling. D2 activity in skeletal muscle fragments from both murine and human sources was low, on the order of about 0.01 fmol/min · mg of muscle protein. The phenotypic changes seen with D2KO cells support a metabolic role for D2 in muscle, hinting at a D2-mediated linkage between thyroid hormone and insulin signaling, but the low activity calls into question whether skeletal muscle D2 is a major source of plasma T3.

The thyroid hormone activating type 2 deiodinase is up-regulated in skeletal myocytes by PPAR-γ agonists, but is not likely a major source of plasma tri-iodothyronine.

Whereas a small amount of thyroid hormone is secreted in the active form by the thyroid gland, most active hormone is both generated and ultimately metabolized in peripheral tissues by the actions of the deiodinases. The type 2 deiodinase (D2) coverts the major circulating form, T4, into active thyroid hormone (T3) in the anterior pituitary, brain, and brown fat (1,2,3). Type 3 deiodinase (D3) inactivates T3 in the skin (4,5), uteroplacental unit (6,7), and many developing structures (8), whereas type 1 deiodinase is expressed in liver and kidney. D2 and D3 play a role in the maintenance of thyroid hormone homeostasis; for example, in the face of iodine deficiency, a generalized increase in D2 activity and a reciprocal fall in D3 activity helps to maintain normal serum T3 levels. At the same time, these enzymes can have tissue-specific effects, regulating thyroid hormone signaling without necessarily changing circulating levels of thyroid hormone (9).

Several lines of evidence suggest that D2 plays a role in the regulation of energy expenditure. An early clue came from studies of adaptive thermogenesis in brown adipose tissue (BAT); in response to cold or high-caloric diet, sympathetic stimulation of BAT triggers a potent induction of D2 activity (10,11,12,13). This increase in D2 activity in turn raises the intranuclear T3 concentration, thus inducing the expression of T3-responsive genes important for thermogenesis. In this setting, the effect is tissue specific because circulating T3 levels do not change. The next clue was the discovery that bile acids can increase energy expenditure via binding to the G protein-coupled receptor TGR5, with downstream activation of D2 in BAT being critical for the energetic response (14).

Whereas BAT is present in limited quantities in adult humans, both D2 (15,16,17) and TGR5 (14) are expressed in skeletal muscle, a tissue that is a major site of energy expenditure and one that is also highly sensitive to thyroid hormone (18). Based on these clues, it is reasonable to hypothesize that D2 in skeletal muscle might play a thermogenic role analogous to the role it plays in BAT adaptive thermogenesis, potentially altering the behavior of metabolic processes in muscle and thus affecting overall energy expenditure.

The current studies were undertaken to characterize the regulation of D2 activity in muscle, as a step toward further understanding the metabolic role of the enzyme in this tissue. A number of readily available immortalized or primary muscle-derived cell lines have been shown to express D2, including RMS-13 (19), SkMC primary human muscle satellite cells (Lonza) (20), and human skeletal muscle myoblasts (HSMM) primary human skeletal myoblasts (Lonza, Walkersville, MD) (14). We initiated studies of the metabolic regulation of D2 in HSMM primary cells because these are readily expanded and can be fused to myotubes in culture as needed. HSMM D2 activity (∼0.01 fmol/min · mg in sonicates under basal conditions) is responsive to forskolin and bile acids (14). Moreover, treatment with 1 μm MG132, a proteasomal inhibitor, increases D2 activity in these cells by about 2-fold compared with dimethylsulfoxide (DMSO)-controls (our unpublished data).

For some experiments, we used a previously characterized mouse line with targeted disruption of the D2 gene (Dio2), the Dio2 knockout (D2KO) mouse (21). The D2KO mouse exhibits signs of thyroid hormone deficiency in tissues in which D2 is known to play an important role. These include inefficient heat generation from BAT (22,23), deafness (24), and relative pituitary resistance to T4 as evidenced by high TSH, high T4, and normal serum T3 levels (21). The current studies provide evidence for novel regulatory mechanisms for D2 in skeletal muscle and hint at a linkage between thyroid signaling and insulin signaling in muscle.

Materials and Methods

Reagents and materials

Unless otherwise specified, all reagents were purchased from Sigma (St. Louis, MO). Fungizone (amphotericin B), PBS, and Dispase were purchased from Invitrogen (Carlsbad, CA). Streptomycin was purchased from Fisher Scientific (Pittsburgh, PA). Pioglitazone was obtained from Discovery Fine Chemicals (Dorset, UK). Forskolin, ciglitazone, GW7647, and L-165,041 were from EMD Chemicals (San Diego, CA). Outer ring-labeled T4 (specific activity 4400 Ci/mmol) and T3 (specific activity 2200 Ci/mmol) were purchased from PerkinElmer (Boston, MA) and purified on LH-20 columns before use. HSMM cells and subculture reagents were purchased from Lonza.

Animals

All studies were performed under a protocol approved by the Standing Committee on Animal Research. C57BL/6J mice were purchased from The Jackson Laboratories (Bar Harbor, ME). D2KO/C57BL/6J mice were maintained in our laboratory as previously described (23).

Culture of murine primary skeletal muscle cells

Murine primary skeletal muscle cells were prepared as previously described (25) with minor modifications. Quadriceps, gastrocnemius, and soleus muscles were collected in ice-cold PBS supplemented with 1% PenStrep (100 U/ml penicillin, 100 μg/ml streptomycin) and 0.5% Fungizone (250 μg/ml amphotericin B). Skeletal muscle specimens were trimmed free from visible connective tissue and fat, minced finely, transferred to a digestion solution (1 mg/ml collagenase type IA, 5 mg/ml Dispase, 100 mm CaCl2) and incubated with agitation at 37 C for 50–55 min. Cell suspensions were filtered via 70 μm strainer, and the supernatant was centrifuged at 2500 rpm for 10 min at 4 C. The resulting pellet was resuspended in HAM's/F10 media containing 20% fetal bovine serum (FBS), 25 mg/liter tetracycline, 25 mg/liter streptomycin, 25 mg/liter ampicillin, 1 mg/liter Fungizone, 100 nm sodium selenite, and 25 μg/ml basic fibroblast growth factor. Myoblasts were plated in 75-cm2 collagen-coated flasks. Medium was changed every 2–3 d. At confluence (80%), myoblasts were transferred to new flasks using only PBS (no trypsin) to enrich the culture with myoblasts and cultivated in HAM's F10/DMEM media with supplements as above. After three passages, cells were harvested using trypsin and plated at density of 20,000 cells/cm2 and cultured in differentiation media (DMEM supplemented with 5% FBS, 25 mg/liter tetracycline, 25 mg/liter streptomycin, 25 mg/liter ampicillin, 1 mg/liter Fungizone, 100 nm sodium selenite) for 1 d (myoblasts) or 7 ds (myotubes).

Immunocytochemistry of myoblasts and myotubes

Immunocytochemistry was performed according to the antibody manufacturer's protocol with minor changes (Abcam, Cambridge, MA). Murine primary skeletal muscle cells were cultured on LabTek (Nunc, Glostrup, Denmark) culture chamber slides. Cells were washed with PBS and fixed with 4% paraformaldehyde for 20 min. Cells were permeabilized with Triton X-100 (0.2%), nonspecific binding sites were blocked with 10% FBS, and the cells were then incubated with polyclonal anti-Desmin (1:200; Abcam) and monoclonal anti-α-tubulin (1:1000; Sigma). Secondary antibodies were goat antirabbit (1:500, Alexa Fluor 488; Invitrogen) and goat antimouse (1:500 Alexa Fluor 594), respectively. Nuclei were visualized by inclusion of 4′, 6-diamino-2-phenylindole (Invitrogen) in the mounting medium.

Insulin signaling analysis

For in vitro insulin signaling studies, myoblasts were differentiated into myotubes, then serum starved for 20 h. Cells were subsequently stimulated with 1 or 10 nm insulin for 5 min, lysed at 4 C in 1× cell lysis buffer (Cell Signaling, Beverly, MA). Protein levels were quantified using Bio-Rad protein assay (Bio-Rad Laboratories, Hercules, CA). Total lysates were separated on a polyacrylamide precast gel (Bio-Rad), transferred to Immobilon-P transfer membranes (Millipore, Bedford, MA) and blotted as directed by manufacturer. Western blots were stripped using either Restore PLUS Western blot stripping buffer (Thermo Scientific, Rockford, IL) or 62.5 mm Tris-HCl (pH 6.7), 2% sodium dodecyl sulfate, and 100 mm β-mercaptoethanol at 50 C for 10 min.

D2 and D3 activity assays

Muscle tissue D2 assays

Animals were euthanized with CO2 inhalation; gastrocnemius, quadriceps, soleus, and extensor digitorum longus muscles were dissected, frozen immediately in liquid nitrogen, and stored at −80 C until assay. For homogenization, frozen tissue fragments were cryopulverized (BioPulverizer; BioSpec Products Inc., Bartlesville, OK), and the resulting powder was resuspended in a buffer containing 0.25 m sucrose with 10 mm dithiothreitol, 0.01% Brij-56 (Sigma), and complete protease inhibitor cocktail from Roche (Basel, Switzerland). Homogenization was accomplished using a Tissue-Tearor (BioSpec Products). The resulting homogenate was centrifuged at 10,000 × g at 4 C for 5 min to remove any remaining particulate matter, and the supernatant protein concentration was measured via Bradford assay (Bio-Rad). Homogenate D2 activity was measured following methods previously described (17), with 0.1 nm T4 compared with 100 nm T4 (saturating)-treated samples processed in parallel and 80–150,000 cpm of tracer 125I-labeled T4. D2KO muscle samples were also used to establish the assay background (as negative controls). Unlabeled T3 (10−7 m final) was added to the reaction mixture to saturate any D3 activity. For cell-sonicate D2 assays, cells were harvested in the homogenization buffer and processed as above. For D3 assays, sonicated samples were harvested as above and processed for HPLC as previously described (18).

For live-cell deiodination assays, cells were incubated for 24 h with radiolabeled substrate and drugs as indicated in DMEM medium supplemented with 0.1% BSA, 2 mmol/liter glutamine, 50 μg/ml ampicillin, 50 μg/ml gentamicin, 100 nm sodium selenite, and 10 pm free T4. Deiodination products in the media (T3, rT3, T2, T1, iodide) were resolved via HPLC and T4 outer-ring deiodination was calculated by multiplying the fractional conversion of T4 to T3 by the total T4 concentration in the media. The fractional conversion itself was calculated as the quotient of the iodide peak times 2 divided by the total of all peaks (26). When indicated, production of iodide was determined in the remaining media and cell pellets via trichloroacetic acid precipitation as previously described (26). Total protein was quantified via Bradford assay. Assay background was determined via comparison with cells treated with 20 μm iopanoic acid.

mRNA quantitation

Abundance of specific mRNA molecules was analyzed via quantitative SYBR-green based real-time PCR as previously described (23). Primer sequences were as follows: peroxisome proliferator-activated receptor (PPAR)-γ sense primer (5′-TGGAGCTCGATGACAGTGAC-3′), antisense primer (5′-GTACTGGCTGTCAG-GGTGGT-3′); myosin heavy chain (MHC)-I sense primer (5′-CTTCAACCACCACATGTTCG-3′), antisense primer (5′-AGGTTTGGGCTTT-TGGAAGT-3′), medium-chain acyl CoA dehydrogenase sense primer (5′-AGGTTTCAAGATCGCAATGG-3′), antisense primer (5′-CTCCTTGGTGCTCCACTAGC-3′); troponin I sense primer (5′-TCATGCTGAAGAGC-3′), antisense primer (5′-GGAGGCATTTGGCT-3′).

Results

Regulation of D2 activity in human myocytes by PPAR-γ agonists

We screened a number of drugs relevant for metabolism for effects on D2 in the primary human skeletal muscle cells. Interestingly, pioglitazone increased D2 activity by 1.7- to 1.9-fold (5 or10 μm, respectively) after overnight incubation (Fig. 1A). The effect was measurable after 8 h incubation (not shown). Doses higher than 10 μm produced no further increase in D2 activity nor did prolonged exposures beyond 16 h (data not shown). Treatment with insulin alone (5 nm) had no significant effect and did not alter response to pioglitazone (10 μm). Similar effects were seen in differentiated myotubes (basal activity ∼0.04 fmol/min · mg, ∼2-fold stimulation by 10 μm pioglitazone, n = 3, P < 0.05).

Figure 1.

Figure 1

Open in a new tab

Regulation of D2 in HSMM myoblasts. A, D2 activity in cell-sonicates was measured as iodide released (trichloroacetic acid precipitation) after drug treatments as indicated expressed as fold change vs. vehicle (DMSO)-treated cells. B, As in A, with additional PPAR agonist drugs. For A and B, vehicle-treated myoblasts typically had activities about 0.01 fmol/min · mg protein; all treatments lasted 16 h. *, P < 0.001; **, P < 0.05 compared with DMSO-treated cells; n = 3, except for 10 μm pioglitazone for which n = 10. Bars, mean ± sd. PIO, Pioglitazone, INS, insulin, CIG, ciglitazone, TROG, troglitazone.

The stimulatory effect on D2 activity in HSMM cells was also seen with other PPAR-γ agonists, including ciglitazone and troglitazone (∼1.6-fold increase for both) (Fig. 1B). The highly specific PPAR-γ agonist GW7647 also increased D2 activity by about 1.5-fold (at 6.2 μm, 50% higher than the ED50 for PPAR-γ). By contrast, the PPAR-δ agonist L-165,041 did not stimulate D2 (Fig. 1B).

D2 activity in tissue homogenates

To gain further insight into the regulation of D2 in skeletal muscle, we sought to measure D2 activity in murine quadriceps samples. We used methods previously established for measuring D2 activity in various tissue homogenates (17). We validated the method using Hanaford minipigs, whose hind-limb muscles had activities ranging from 0.01 to 0.08 fmol/min · mg (similar activity levels are seen for Sinclair and Yucatan minipigs). However, whereas iodide release (the index for outer-ring deiodination) was reproducibly higher than background in murine muscle sonicates, the net amount was quite low, such that in most experiments we could not confirm linearity with respect to protein used or time of incubation. Altering the tissue preparation conditions or buffer composition (different homogenization devices, detergents, protease inhibitors) did not significantly improve the sensitivity of the standard assay. Similar results were obtained using rat muscle, human strap muscle (discarded tissue from thyroidectomy patients), and human rectus abdominis muscle.

D2 activity in isolated primary murine myoblasts and myotubes

Given that the D2 activity in murine skeletal muscle was below the sensitivity threshold of the tissue homogenate assay, we pursued an alternate approach to study D2 in this tissue via isolation of myocytes from D2KO and wild-type mice. These primary cells can be differentiated into myotubes in vitro under controlled conditions (Fig. 2A). The studies were also optimized in that the deiodinase assays were performed in live cells during their incubation with various drugs, i.e. live-cell deiodinase (27). This method reflects not only the expression of deiodinases, but also the transport of thyroid hormone and the endogenous levels of deiodinase reducing cofactor and thus may be more physiologically relevant than the cell-sonicate method.

Figure 2.

Figure 2

Open in a new tab

A, Myogenic differentiation of mouse primary cell cultures. Primary skeletal muscle cells from wild-type (WT) mice were plated at confluence and cultures were maintained in differentiation medium for 7 d. Cells were then paraformaldehyde fixed at d 1 (myoblasts) and d 7 (myotubes). Myogenic differentiation was confirmed by immunofluorescent labeling with antidesmin antibody (green). 4′, 6-Diamino-2-phenylindole staining was used to visualize nuclei (blue), and anti-α-tubulin antibody (red) was used as background. The image shows the differentiation of myoblasts and the formation of myotubes at a magnification of ×10 and ×60. B, D2 activity in myoblasts vs. myotubes. D2 activity measured as iodide released (HPLC) in media samples from mouse primary muscle myoblasts and myotubes after treatment with vehicle (DMSO) or 100 μm forskolin (FSK). Bars, mean ± sem n = 4. *, P < 0.001 vs. other groups. C, HPLC chromatograms for identification of D3- or D2-mediated deiodination products. Representative chromatograms of HPLC data in myotubes. On the left, D3 assay conditions were 6 h incubation with 125I-T3 tracer and drugs as indicated. On the right, D2 assay conditions were 12 h incubation with 125I-T4 tracer and drugs as indicated. The oval highlights the T3 peak. ORD, outer ring deiodination.

Myoblasts exhibited little or no outer ring deiodination, with the differences in net iodide release and T3 accumulation failing to reach significance even after treatment with forskolin (Fig. 2B). However, after differentiation to myotubes, substantial outer-ring deiodination was detected based on iodide release. T3 accumulation could not be used as an index of D2 activity in this setting due to the presence of inner-ring deiodination catalyzed by D3. This reaction converted the T3 into T2 and T1, whereas also converting T4 into reverse T3 as detected via HPLC (Fig. 2C, left column). Forskolin treatment increased D2 activity in myotubes substantially. In this case, iodide release was increased about 10-fold, and in addition to T2 accumulation, T3 accumulation was also detectable via HPLC (Fig. 2C, right column). To establish the assay background, the experiment was performed in parallel using cells prepared from D2KO animals. A small amount of non-D2-mediated outer-ring deiodination was identified when comparing vehicle vs. forskolin-treated D2KO myotubes, the mechanism for which is unclear (Fig. 2B).

Regulation of D2 in isolated primary murine myotubes

Having established that isolated murine myotubes express D2 and that their D2 activity could be measured using the more physiological live-cell assay, we used this system to extend our knowledge of D2 regulation.

We first tested lithocholic acid (LCA), the most potent naturally occurring bile-acid ligand for the G protein-coupled receptor TGR5. TGR5 is known to be expressed at low to moderate levels in human skeletal muscle (14,28); similarly, TGR5 mRNA is present in murine myotubes (data not shown). LCA stimulated D2 in a dose-dependent fashion between 10 and 25 μm (Fig. 3A). When pretreated with the protein kinase A inhibitor H89 for 1 h, the LCA effect was lost. These data are consistent with a TGR5-protein kinase A-mediated pathway for D2 induction (14). Next, we compared isolated myotubes prepared from different muscle groups for basal and LCA-stimulated D2 activity, finding that the leg muscles had higher activity than back or neck and the leg muscles were also the most responsive to LCA, indicating that D2 expression in skeletal muscle is heterogeneous with respect to location (Fig. 3B).

Figure 3.

Figure 3

Open in a new tab

Measurement of D2 activity by live-cell assay in murine primary myotubes. A, Myotubes were incubated with concentrations of LCA as indicated for 24 h. Some cells were pretreated with H89 for1 h before treatment with 20 μm LCA. B, Myotubes were prepared from quadriceps, intercostals or trapezius and treated with 20 μm LCA for 24 h. C, Same as A, except myotubes were incubated with INS and/or PIO as indicated. D, Quantitative real-time PCR for D2 mRNA using cyclophilin B as the housekeeping gene in wild-type myotubes treated for 24 h with 5 nm INS, 5 μm PIO, 25 μm LCA, or 0.5 mm DIB, expressed as fold change vs. DMSO. Bars (A–D), mean ± sem; n = 5. *, P < 0.05; **, P < 0.01; ***, P < 0.001 compared with DMSO. PIO, Pioglitazone, INS, insulin; DIB, dibutyryl cAMP.

Pioglitazone also stimulated D2 activity in myotubes (2.7-fold at 5 μm and 1.7-fold at 10 μm) after 24 h treatment (Fig. 3C). Whereas D2 activity was significantly increased after insulin stimulation (1.6-fold at 5 nm), no further stimulation was observed when insulin and pioglitazone were combined. In parallel with activity, Dio2 mRNA was increased by both pioglitazone and LCA; however, there was no significant induction by insulin (Fig. 3D). Small increases in D3 activity were also observed after treatment with insulin (∼10%, n = 4, P < 0.05) and LCA (∼14%, n = 4, P < 0.01), whereas there was a more dramatic fall in D3 activity (∼46%, n = 4, P < 0.001) after treatment with 5 μm pioglitazone for 6 h (not shown).

Effect of Dio2 knockout on insulin signaling

Having validated that the isolated myotube approach was suitable for studies of D2, we next investigated the potential metabolic role of D2 in myotubes comparing control and D2KO mice.

Insulin signaling was tested in myotubes via measurement of Akt phosphorylation by Western blot (Fig. 4, A and B). We observed no difference in serum-starved (basal) Akt phosphorylation between wild-type and D2KO cells. However, a 50% decrease in insulin-stimulated Akt phosphorylation was observed in the D2KO cells (1 nm insulin, 5 min). Stimulation with higher insulin concentration (10 nm) resulted in similar Akt phosphorylation signal in both groups (Fig. 4, A and B).

Figure 4.

Figure 4

Open in a new tab

Akt activation before and after insulin stimulation. A, Akt phosphorylation was quantified in murine primary myotubes after serum starvation for 20 h and treatment with vehicle or 1 or 10 nm insulin for 5 min. A representative blot is shown. B, Pooled densitometry results for three experiments. Bars, mean ± sem. *, P < 0.05 vs. wild type (WT).

We hypothesized that the insulin signaling abnormalities seen in D2KO myotubes could stem from a relative deficiency of active thyroid hormone that would normally be generated from T4 in wild-type myotubes. If so, one might also expect D2KO myotubes to exhibit other signs of hypothyroidism, such as having a gene expression pattern consistent with type I muscle fibers (because T3 promotes the development of type II fibers) (18). Therefore, we prepared RNA from myotubes cultured in the presence of 5% FBS and determined expression of genes known to be T3 responsive and/or relevant for fiber type determination. Troponin-I and MHC-I, which are known to be down-regulated by T3 and up-regulated in type I fibers, were both increased in D2KO myotubes. PGC-1b, another gene known to be up-regulated in type I fibers, was increased in the D2KO myotubes as well (18). Lastly, PPAR-γ itself was reduced in the D2KO myotubes, also consistent with a hypothyroid phenotype (Fig. 5) (29).

Figure 5.

Figure 5

Open in a new tab

Gene expression data from primary murine myotubes. Quantitative real-time PCR of genes as indicated was performed using cyclophilin B as the housekeeping gene. Results are expressed as fold change compared with wild type (WT). Bars, mean ± sem, n = 4. *, P < 0.05. PGC-1α and -β, PPAR-γ coactivator 1α and -β; MCAD, medium-chain acyl CoA dehydrogenase. PGC, peroxisome proliferator-activated receptor-gamma coactivator.

Discussion

The concept that D2 has a metabolic role beyond its homeostastic function to maintain circulating thyroid hormone levels has had a long gestation. Indirect support for a D2-dependent energy expenditure mechanism in human subjects arose from a study of patients on thyroid hormone replacement who underwent incremental changes in T4 dose either upward or downward. In this study resting energy expenditure, about 22% of which occurs in skeletal muscle (29), correlated directly with T4 replacement dose, even though serum T3 levels did not change (30). Additional indirect support comes from genetic analysis of Dio2 polymorphisms; in some populations, Dio2 single nucleotide polymorphisms are linked to insulin sensitivity (17,31,32,33,34). The exact nature of the metabolic events downstream from an increase in D2-generated T3 remain to be determined, but intracellular T3 has been linked to increased calcium cycling, mitochondrial uncoupling, and other energetic processes in skeletal muscle (35,36,37,38,39,40).

The most significant finding of the current studies is that D2 is up-regulated in skeletal myocytes by PPAR-γ ligands (Figs. 1 and 3, C and D). That this class of insulin-sensitizing drugs can regulate D2 in muscle is striking, from both the clinical and biological perspective, raising the possibility that endogenous or exogenous PPAR-γ ligands could exert their metabolic effects in part by altering thyroid hormone signaling. Given that D2 activity was increased by multiple members of the thiazolidinedione class as well as the PPAR-γ-specific agonist GW7647, it is reasonable to postulate that Dio2 is functionally downstream from PPAR-γ. In fact, a PPAR element in the Dio2 promoter was recently identified via in silico analysis and gel shift assay (34). It remains to be determined whether the pioglitazone effect observed in the current study is purely transcriptional (Fig. 3D) or whether D2 activity is altered via a posttranscriptional mechanism, e.g. an effect on D2 ubiquitination.

Given that insulin increased D2 activity in murine myotubes (Fig. 3C) and has been reported to stimulate D2 activity in freshly isolated murine brown adipocytes (41), one might speculate that the insulin-sensitizing drugs might stimulate D2 simply via increasing insulin sensitivity. However, the pioglitazone effect on D2 was observed even in the absence of insulin (serum free media) (Fig. 3C). Another striking finding obtained in these studies is that D2 is relevant for the insulin signaling cascade, as indicated by the findings of reduced Akt phosphorylation in myotubes isolated from D2KO mice (Fig. 4, A and B). The findings are consistent with a prior study in which isolated adipocytes and myocytes from rats were found to be insulin resistant after being made hypothyroid (42). The concept that D2-mediated T3 production may promote insulin responsiveness may seem initially contradictory, given the clinical experience that hyperthyroidism worsens glycemic control in established diabetes and may unmask latent diabetes (43). However, it must be kept in mind that systemic hyperthyroidism, i.e. increased circulating thyroid hormone levels, affects not only skeletal muscle but also other tissues. These systemic effects, including increased hepatic glucose output, may well outweigh any local effects of D2 in skeletal muscle.

An unexpected finding of the current studies is that using the current methodology, the tissue-homogenate deiodinase assay was found to lack the sensitivity needed to perform more extensive in vivo experiments. Prior studies have demonstrated D2 activity and mRNA in human skeletal muscle, with reported activities ranging from about 1.0 (15,17) to as low as 0.003 fmol/min · mg protein (16). The data in the present study are more in line with the lower values; the reasons for the wide range in estimates remain unclear, although different ethnic backgrounds or the specific muscle groups being sampled could contribute to variability. If the lower estimate is taken to be correct, then the role of skeletal muscle D2 as the primary source of plasma T3 would need to be revisited (26). One must keep in mind that the current data do not invalidate the concept that the D2 pathway in peripheral tissues is a major source of plasma T3; the question is what is the anatomical site. D2 has been reported to be expressed in bone (44), vascular smooth muscle (45,46), and skin (4), for example.

The low levels of D2 activity in skeletal muscle also beg the question as to how much D2 activity is required for a significant effect on nuclear T3 levels and thus thyroid signaling in a given cell type. The answer is unknown, but the phenotypic differences seen between D2KO cells and wild-type cells suggests that even a low amount of D2 is important, at least if the level is altered in a chronic fashion. By establishing factors such as PPAR-γ agonists and LCA that can increase D2 activity in muscle cells, the current data provide a basis for additional studies investigating the potential metabolic effects of mild, long-term increases in D2 activity in skeletal muscle.

HSMM cells have previously been reported to have D3 activity (47), so it is not surprising that substantial D3 activity was found in primary cultures from murine skeletal muscle. The expression and regulation of D3 has best been studied in developing tissues (8,48), but recently examples of D3 induction in adult tissues have been discovered, e.g. by hypoxia inducible factor and sonic hedgehog, in postnatal heart and skin (18,49). Coexpression of the two deiodinases is consistent with a developmental phenotype because D3 is not thought to be expressed in healthy adult skeletal muscle (Visser, T. J., personal communication).

Whereas insulin and LCA treatments led to minimal increases in D3 activity, the striking decrease in D3 activity induced by pioglitazone treatment is notable in that it provides another example of the coordinated regulation of the two deiodinases by a single pathway, the others being thyroid hormone itself and the hedgehog pathway (9). The fact that Akt phosphorylation is abnormal in D2KO cells suggests that the presence of D2 in wild-type cells is still relevant despite the presence of D3; in other words, the net balance of D2 and D3 under these conditions favors net generation of T3.

The current data extend previous work that have established that D2 is up-regulated in BAT by bile acids (14), insulin (41), and sympathetic innervation (10). Taken together, the results of these studies indicate that the deiodinases may serve as downstream effectors for a wide variety of metabolic factors. Furthermore, if the proposed linkage between D2 activity and insulin responsiveness in skeletal muscle indicated by the current data are confirmed, then one could envision that pharmacological manipulation of the deiodinases could potentially be used to improve muscle metabolic phenotypes.

Acknowledgments

The authors thank Dr. Valerie Galton and Dr. Donald St. Germain for providing the D2KO mouse.

Footnotes

This work was supported by National Institutes of Health Grants DK77148 and K08 DK064643 and an American Thyroid Association research grant.

Disclosure Summary: The authors have nothing to disclose.

First Published Online November 26, 2008

Abbreviations: BAT, Brown adipose tissue; D2, type 2 deiodinase; D3, type 3 deiodinase; D2KO, Dio2 knockout; DMSO, dimethylsulfoxide; FBS, fetal bovine serum; HSMM, human skeletal muscle myoblasts; LCA, lithocholic acid; MHC, myosin heavy chain; PPAR, peroxisome proliferator-activated receptor.

References

  1. Larsen PR, Silva JE, Kaplan MM 1981 Relationships between circulating and intracellular thyroid hormones: physiological and clinical implications. Endocr Rev 2:87–102 [DOI] [PubMed] [Google Scholar]
  2. St. Germain DL, Galton VA 1997 The deiodinase family of selenoproteins. Thyroid 7:655–668 [DOI] [PubMed] [Google Scholar]
  3. Kuiper GG, Kester MH, Peeters RP, Visser TJ 2005 Biochemical mechanisms of thyroid hormone deiodination. Thyroid 15:787–798 [DOI] [PubMed] [Google Scholar]
  4. Kaplan MM, Pan C, Gordon PR, Lee JK, Gilchrest BA 1988 Human epidermal keratinocytes in culture convert thyroxine to 3,5,3′-triiodothyronine by type II iodothyronine deiodination: a novel endocrine function of the skin. J Clin Endocrinol Metab 66:815–822 [DOI] [PubMed] [Google Scholar]
  5. Santini F, Vitti P, Chiovato L, Ceccarini G, Macchia M, Montanelli L, Gatti G, Rosellini V, Mammoli C, Martino E, Chopra IJ, Safer JD, Braverman LE, Pinchera A 2003 Role for inner ring deiodination preventing transcutaneous passage of thyroxine. J Clin Endocrinol Metab 88:2825–2830 [DOI] [PubMed] [Google Scholar]
  6. Huang SA, Dorfman DM, Genest DR, Salvatore D, Larsen PR 2003 Type 3 iodothyronine deiodinase is highly expressed in the human uteroplacental unit and in fetal epithelium. J Clin Endocrinol Metab 88:1384–1388 [DOI] [PubMed] [Google Scholar]
  7. Wasco EC, Martinez E, Grant KS, St. Germain EA, St. Germain DL, Galton VA 2003 Determinants of iodothyronine deiodinase activities in rodent uterus. Endocrinology 144:4253–4261 [DOI] [PubMed] [Google Scholar]
  8. Galton VA 2005 The roles of the iodothyronine deiodinases in mammalian development. Thyroid 15:823–834 [DOI] [PubMed] [Google Scholar]
  9. Bianco AC, Kim BW 2006 Deiodinases: implications of the local control of thyroid hormone action. J Clin Invest 116:2571–2579 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Silva JE, Larsen PR 1983 Adrenergic activation of triiodothyronine production in brown adipose tissue. Nature 305:712–713 [DOI] [PubMed] [Google Scholar]
  11. Bianco AC, Silva JE 1988 Cold exposure rapidly induces virtual saturation of brown adipose tissue nuclear T3 receptors. Am J Physiol 255:E496–E503 [DOI] [PubMed] [Google Scholar]
  12. Bianco AC, Sheng X, Silva JE 1988 Cold exposure rapidly induces virtual saturation of brown adipose tissue nuclear T3 receptors. J Biol Chem 263:18168–18175 [DOI] [PubMed] [Google Scholar]
  13. Bianco AC, Silva JE 1987 Intracellular conversion of thyroxine to triiodothyronine is required for the optimal thermogenic function of brown adipose tissue. J Clin Invest 79:295–300 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Watanabe M, Houten SM, Mataki C, Christoffolete MA, Kim BW, Sato H, Messaddeq N, Harney JW, Ezaki O, Kodama T, Schoonjans K, Bianco AC, Auwerx J 2006Bile acids induce energy expenditure by promoting intracellular thyroid hormone activation. Nature 439:484–489 [DOI] [PubMed] [Google Scholar]
  15. Salvatore D, Bartha T, Harney JW, Larsen PR 1996 Molecular biological and biochemical characterization of the human type 2 selenodeiodinase. Endocrinology 137:3308–3315 [DOI] [PubMed] [Google Scholar]
  16. Mebis L, Langouche L, Visser TJ, Van den Berghe G 2007 The type II iodothyronine deiodinase is up-regulated in skeletal muscle during prolonged critical illness. J Clin Endocrinol Metab 92:3330–3333 [DOI] [PubMed] [Google Scholar]
  17. Canani LH, Capp C, Dora JM, Meyer EL, Wagner MS, Harney JW, Larsen PR, Gross JL, Bianco AC, Maia AL 2005 The type 2 deiodinase A/G (Thr92Ala) polymorphism is associated with decreased enzyme velocity and increased insulin resistance in patients with type 2 diabetes mellitus. J Clin Endocrinol Metab 90:3472–3478 [DOI] [PubMed] [Google Scholar]
  18. Simonides WS, Mulcahey MA, Redout EM, Muller A, Zuidwijk MJ, Visser TJ, Wassen FW, Crescenzi A, da-Silva WS, Harney J, Engel FB, Obregon MJ, Larsen PR, Bianco AC, Huang SA 2008 Hypoxia-inducible factor induces local thyroid hormone inactivation during hypoxic-ischemic disease in rats. J Clin Invest 118:975–983 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. da-Silva WS, Harney JW, Kim BW, Li J, Bianco SD, Crescenzi A, Christoffolete MA, Huang SA, Bianco AC 2007 The small polyphenolic molecule kaempferol increases cellular energy expenditure and thyroid hormone activation. Diabetes 56:767–776 [DOI] [PubMed] [Google Scholar]
  20. Hosoi Y, Murakami M, Mizuma H, Ogiwara T, Imamura M, Mori M 1999. J Clin Endocrinol Metab 84:3293–3300 [DOI] [PubMed] [Google Scholar]
  21. Schneider MJ, Fiering SN, Pallud SE, Parlow AF, St. Germain DL, Galton VA 2001 Targeted disruption of the type 2 selenodeiodinase gene (DIO2) results in a phenotype of pituitary resistance to T4. Mol Endocrinol 15:2137–2148 [DOI] [PubMed] [Google Scholar]
  22. de Jesus LA, Carvalho SD, Ribeiro MO, Schneider M, Kim S-W, Harney JW, Larsen PR, Bianco AC 2001 The type 2 iodothyronine deiodinase is essential for adaptive thermogenesis in brown adipose tissue. J Clin Invest 108:1379–1385 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Christoffolete MA, Linardi CCG, de Jesus LA, Ebina KN, Carvalho SD, Ribeiro MO, Rabelo R, Curcio C, Martins L, Kimura ET, Bianco AC 2004 Mice with targeted disruption of the Dio2 gene have cold-induced overexpression of the uncoupling protein 1 gene but fail to increase brown adipose tissue lipogenesis and adaptive thermogenesis. Diabetes 53:577–584 [DOI] [PubMed] [Google Scholar]
  24. Forrest D, Ng L, Kelley M, Schneider MJ, St. Germain DL, Galton VA Cochlear defects and deafness in mice lacking type II selenodeiodinase. Proc 73rd Annual Meeting of the American Thyroid Association, Washington, DC, 2001 (Abstract P4) [Google Scholar]
  25. Rando TA, Blau HM 1994 Primary mouse myoblast purification, characterization, and transplantation for cell-mediated gene therapy. J Cell Biol 125:1275–1287 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Maia AL, Kim BW, Huang SA, Harney JW, Larsen PR 2005 Type 2 iodothyronine deiodinase is the major source of plasma T3 in euthyroid humans. J Clin Invest 115:2524–2533 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Croteau W, Bodwell JE, Richardson JM, St. Germain DL 1998 Conserved cysteines in the type 1 deiodinase selenoprotein are not essential for catalytic activity. J Biol Chem 273:25230–25236 [DOI] [PubMed] [Google Scholar]
  28. Kawamata Y, Fujii R, Hosoya M, Harada M, Yoshida H, Miwa M, Fukusumi S, Habata Y, Itoh T, Shintani Y, Hinuma S, Fujisawa Y, Fujino M 2003 A G protein-coupled receptor responsive to bile acids. J Biol Chem 278:9435–9440 [DOI] [PubMed] [Google Scholar]
  29. Zurlo F, Larson K, Bogardus C, Ravussin E 1990 Skeletal muscle metabolism is a major determinant of resting energy expenditure. J Clin Invest 86:1423–1427 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. al-Adsani H, Hoffer LJ, Silva JE 1997 Resting energy expenditure is sensitive to small dose changes in patients on chronic thyroid hormone replacement. J Clin Endocrinol Metab 82:1118–1125 [DOI] [PubMed] [Google Scholar]
  31. Mentuccia D, Proietti-Pannunzi L, Tanner K, Bacci V, Pollin TI, Poehlman ET, Shuldiner AR, Celi FS 2002 Association between a novel variant of the human type 2 deiodinase gene Thr92Ala and insulin resistance: evidence of interaction with the Trp64Arg variant of the beta-3-adrenergic receptor. Diabetes 51:880–883 [DOI] [PubMed] [Google Scholar]
  32. Mentuccia D, Thomas MJ, Coppotelli G, Reinhart L, Mitchell BD, Shuldiner AR, Celi FS, The Thr92Ala deiodinase type 2(DIO2) variant is not associated with type 2 diabetes or indices of insulin resistance in the old order of Amish. Program of the 86th Annual Meeting of The Endocrine Society, New Orleans, LA, 2004 (Abstract P3-382) [DOI] [PubMed] [Google Scholar]
  33. Grarup N, Andersen MK, Andreasen CH, Albrechtsen A, Borch-Johnsen K, Jorgensen T, Auwerx J, Schmitz O, Hansen T, Pedersen O 2007 Studies of the common DIO2 Thr92Ala polymorphism and metabolic phenotypes in 7342 Danish white subjects. J Clin Endocrinol Metab 92:363–366 [DOI] [PubMed] [Google Scholar]
  34. Fiorito M, Torrente I, De Cosmo S, Guida V, Colosimo A, Prudente S, Flex E, Menghini R, Miccoli R, Penno G, Pellegrini F, Tassi V, Federici M, Trischitta V, Dallapiccola B 2007 Interaction of DIO2 T92A and PPARγ2 P12A polymorphisms in the modulation of metabolic syndrome. Obesity (Silver Spring) 15:2889–2895 [DOI] [PubMed] [Google Scholar]
  35. Silva JE 1995 Thyroid hormone control of thermogenesis and energy balance. Thyroid 5:481–492 [DOI] [PubMed] [Google Scholar]
  36. Lebon V, Dufour S, Petersen KF, Ren J, Jucker BM, Slezak LA, Cline GW, Rothman DL, Shulman GI 2001 Effect of triiodothyronine on mitochondrial energy coupling in human skeletal muscle. J Clin Invest 108:733–737 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Reitman ML, He Y, Gong DW 1999 Thyroid hormone and other regulators of uncoupling proteins. Int J Obes Relat Metab Disord 23(Suppl 6):S56–S59 [DOI] [PubMed] [Google Scholar]
  38. Harper ME, Seifert EL 2008 Thyroid hormone effects on mitochondrial energetics. Thyroid 18:145–156 [DOI] [PubMed] [Google Scholar]
  39. Dillmann WH 2002 Cellular action of thyroid hormone on the heart. Thyroid 12:447–452 [DOI] [PubMed] [Google Scholar]
  40. Reis M, Farage M, de Meis L 2002 Thermogenesis and energy expenditure: control of heat production by the Ca(2+)-ATPase of fast and slow muscle. Mol Membr Biol 19:301–310 [DOI] [PubMed] [Google Scholar]
  41. Mills I, Barge RM, Silva JE, Larsen PR 1987 Insulin stimulation of iodothyronine 5′-deiodinase in rat brown adipocytes. Biochem Biophys Res Commun 143:81–86 [DOI] [PubMed] [Google Scholar]
  42. Czech MP, Malbon CC, Kerman K, Gitomer W, Pilch PF 1980 Effect of thyroid status on insulin action in rat adipocytes and skeletal muscle. J Clin Invest 66:574–582 [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Wu P 2007 Thyroid disorders and diabetes. It is common for a person to be affected by both thyroid disease and diabetes. Diabetes Self Manag 24:80–82, 85–87 [PubMed] [Google Scholar]
  44. Gouveia CH, Christoffolete MA, Zaitune CR, Dora JM, Harney JW, Maia AL, Bianco AC 2005 Type 2 iodothyronine selenodeiodinase is expressed throughout the mouse skeleton and in the MC3T3-E1 mouse osteoblastic cell line during differentiation. Endocrinology 146:195–200 [DOI] [PubMed] [Google Scholar]
  45. Mizuma H, Murakami M, Mori M 2001 Thyroid hormone activation in human vascular smooth muscle cells: expression of type II iodothyronine deiodinase. Circ Res 88:313–318 [DOI] [PubMed] [Google Scholar]
  46. Maeda A, Toyoda N, Yasuzawa-Amano S, Iwasaka T, Nishikawa M 2003 Type 2 deiodinase expression is stimulated by growth factors in human vascular smooth muscle cells. Mol Cell Endocrinol 200:111–117 [DOI] [PubMed] [Google Scholar]
  47. Huang SA, Mulcahey MA, Crescenzi A, Chung M, Kim B, Barnes CA, Kuijt W, Tu HM, Harney JW, Larsen PR 2005 Transforming growth factor-β promotes inactivation of extracellular thyroid hormones via transcriptional stimulation of type 3 iodothyronine deiodinase. Mol Endocrinol 19:3126–3136 [DOI] [PubMed] [Google Scholar]
  48. Brown DD 2005 The role of deiodinases in amphibian metamorphosis. Thyroid 15:815–821 [DOI] [PubMed] [Google Scholar]
  49. Dentice M, Luongo C, Huang S, Ambrosio R, Elefante A, Mirebeau-Prunier D, Zavacki AM, Fenzi G, Grachtchouk M, Hutchin M, Dlugosz AA, Bianco AC, Missero C, Larsen PR, Salvatore D 2007 Sonic hedgehog-induced type 3 deiodinase blocks thyroid hormone action enhancing proliferation of normal and malignant keratinocytes. Proc Natl Acad Sci USA 104:14466–14471 [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Endocrinology are provided here courtesy of The Endocrine Society

RESOURCES