Type 2 Iodothyronine Deiodinase Levels Are Higher in Slow-Twitch ...

T H Y R O I D - T R H - T S H

Type 2 Iodothyronine Deiodinase Levels Are Higher in Slow-Twitch than Fast-Twitch Mouse Skeletal Muscle and Are Increased in Hypothyroidism Alessandro Marsili, Waile Ramadan, John W. Harney, Michelle Mulcahey, Luciana Audi Castroneves, Iuri Martin Goemann, Simone Magagnin Wajner, Stephen A. Huang, Ann Marie Zavacki, Ana Luiza Maia, Monica Dentice, Domenico Salvatore, J. Enrique Silva, and P. Reed Larsen Thyroid Section (A.M., J.W.H., A.M.Z., P.R.L.), Division of Endocrinology, Diabetes, and Hypertension, Brigham and Women’s Hospital and Harvard Medical School, Boston, and Thyroid Program (M.M., L.A.C., S.A.H.), Department of Endocrinology, Children’s Hospital, Boston, Massachusetts 02115; Division of Endocrinology, Diabetes, and Metabolism (W.R., J.E.S.), Baystate Medical Center, Tufts University School of Medicine, Springfield, Massachusetts 01199; Thyroid Section (I.M.G., S.M.W., A.L.M.), Endocrine Division, Hospital de Clínicas de Porto Alegre, Universidade Federal do Rio Grande do Sul, Porto Alegre RS-90035-003, Brazil; and Department of Molecular and Clinical Endocrinology and Oncology (M.D., D.S.), University of Naples “Federico II,” 80131 Naples, Italy

Because of its large mass, relatively high metabolic activity and responsiveness to thyroid hormone, skeletal muscle contributes significantly to energy expenditure. Despite the presence of mRNA encoding the type 2 iodothyronine-deiodinase (D2), an enzyme that activates T4 to T3, very low or undetectable activity has been reported in muscle homogenates of adult humans and mice. With a modified D2 assay, using microsomal protein, overnight incubation and protein from D2 knockout mouse muscle as a tissue-specific blank, we examined slow- and fast-twitch mouse skeletal muscles for D2 activity and its response to physiological stimuli. D2 activity was detectable in all hind limb muscles of 8- to 12-wk old C57/BL6 mice. Interestingly, it was higher in the slow-twitch soleus than in fast-twitch muscles (0.40 ⫾ 0.06 vs. 0.076 ⫾ 0.01 fmol/min 䡠 mg microsomal protein, respectively, P ⬍ 0.001). These levels are greater than those previously reported. Hypothyroidism caused a 40% (P ⬍ 0.01) and 300% (P ⬍ 0.001) increase in D2 activity after 4 and 8 wk treatment with antithyroid drugs, respectively, with no changes in D2 mRNA. Neither D2 mRNA nor activity increased after an overnight 4 C exposure despite a 10-fold increase in D2 activity in brown adipose tissue in the same mice. The magnitude of the activity, the fiber specificity, and the robust posttranslational response to hypothyroidism argue for a more important role for D2-generated T3 in skeletal muscle physiology than previously assumed. (Endocrinology 151: 5952–5960, 2010)

S

keletal muscle comprises about 40 –50% of human total body mass, and it is the single largest contributor to resting energy expenditure and insulin-induced glucose disposal in adults (1, 2). A significant portion of the differences in metabolic rate among humans can be accounted for by differences in skeletal muscle energy expenditure (3). Thyroid hormone is the most potent

physiological regulator of basal metabolic rate or its surrogate, resting energy expenditure (4). Skeletal muscle is also an important target tissue for thyroid hormones (5, 6). T4, the main secretory product of the human thyroid gland, has weak intrinsic hormonal activity and must be deiodinated in the 5⬘ position to T3 to exert its biological effects. This reaction is catalyzed by

ISSN Print 0013-7227 ISSN Online 1945-7170 Printed in U.S.A. Copyright © 2010 by The Endocrine Society doi: 10.1210/en.2010-0631 Received June 4, 2010. Accepted August 30, 2010. First Published Online September 29, 2010

Abbreviations: BAT, Brown adipose tissue; Ct, cycle threshold; D2, type 2 deiodinase; D3, type 3 deiodinase; D2KO, Dio2 gene inactivated; DTT, dithiothreitol; MMI, methimazole; P1, postnatal d 1; TgD2, transgenic D2 in cardiac muscle.

5952

endo.endojournals.org

Endocrinology, December 2010, 151(12):5952–5960


two iodothyronine deiodinases, termed type 1 or type 2 (D2), which combined account for approximately 80% of the T3 produced daily in humans (7). Both T4 and T3 are irreversibly inactivated by the removal of an inner-ring iodine, a reaction catalyzed by the type 3 deiodinase (D3) (7). Therefore, these three enzymes are critical regulators of thyroid hormone action. Tissues expressing D2 such as brown adipose tissue, brain, and pituitary derive a significant fraction of their nuclear receptor T3 via intracellular D2-mediated T4 5⬘ deiodination (7). The presence of propylthiouracil-insensitive T4 to T3 conversion in humans and rodents as well as in vitro studies have also suggested that D2 may contribute to plasma T3 production, but it is not clear which tissue or tissues are the source of circulating T3 (7–9). D2 mRNA is expressed in human and mouse skeletal muscle (10 –12). However, the D2 enzymatic activity in both species is quite low, even undetectable in some studies (13–17). To define the potential role of D2 in any tissue, it is necessary to measure its activity rather than mRNA because much of the regulation of D2 activity occurs at the posttranslational level (7, 18). Because skeletal muscle comprises 40 –50% of body mass, if D2 contributes significantly to muscle T3, even low D2 activity could have a significant metabolic impact on overall body or resting energy expenditure through intrinsic effects on muscle metabolic rate. Estimates of the contribution of muscle D2 in humans to the circulating pool of T3 have been made indirectly from in vitro studies, but the validity of these observations depends heavily on the accuracy of the muscle D2 determinations (9). We have addressed this issue by developing a sensitive method, the initial results of which are presented here. To facilitate our methodological studies, we used muscle from two strains of genetically modified mice, one expressing high levels of human D2 in cardiac muscle (TgD2) and a second strain with the Dio2 gene inactivated (D2KO), thus providing an optimal null background for the assay.


5953

backcrossed 11 times into the C57BL/6 background. All animals were maintained under 12 h light, 12-h dark cycle and the standard animal facility temperature and humidity.

D2 assay methodology Animals were euthanized with isofluorane (Phoenix Pharmaceutical, St. Joseph, MO) inhalation; gastrocnemius, vastus lateralis, vastus intermedius, soleus, and anterior tibialis muscles were dissected and processed immediately or snap frozen in liquid nitrogen and stored at ⫺80 C until assay. For initial studies using homogenates, we followed previously reported methods with slight modifications (14, 15). Briefly, muscle samples from wild-type or D2KO were homogenized on ice in 10 volumes of PE buffer (0.1 phosphate, 2 mM EDTA, pH 7) and 10 mM dithiothreitol using a Tissue-Tearor (BioSpec Products, Bartlesville, OK). Protease inhibitor was included as described by Grozovsky et al. (13). Triplicate samples of 250 ␮g of homogenate protein were incubated for 4 h at 37 C with 3⬘-5⬘ 125 I T4 (100,000 cpm) in a final volume of 100 ␮l of PE buffer and 10 mM dithiothreitol (DTT) in the presence of 0.1 ␮M unlabeled T3 and 1 mM 6-n-propylthiouracil. Microsomes for the assay described below were prepared from homogenates in 10 volumes of 0.4 M sucrose, 20 mM DTT, 10 mM HEPES, 1 mM EDTA buffer (pH 7.0) containing a protease inhibitor cocktail from Roche (Basel, Switzerland). Homogenate was centrifuged at 10,000 ⫻ g for 20 min at 4 C; the pellet was resuspended in the same buffer, and the centrifugation was repeated and the two supernatants were combined. This step removed mitochondria, nuclei, and cell debris. The supernatant was then centrifuged at 150,000 ⫻ g for 1 h at 4 C to collect the microsomal fraction. The microsomal pellet was resuspended in the homogenizing buffer and used fresh for further steps. A similar approach was used to prepare microsomes from ventricular muscle from transgenic mice (21) and from skeletal muscle of D2KO mice (20). Protein concentration was measured with the Bio-Rad protein assay (Bio-Rad Laboratories, Hercules, CA) following the manufacturer’s instructions (13). Assays were performed using 20 –100 ␮g of protein in 100 ␮l final volume of PE buffer containing 20 mM DTT and 1 mM propylthiouracil (see below). Reactions were stopped by addition of 0.1 ml of ice-cold 100% ethanol. After vortexing, the supernatant was collected by centrifugation and mixed with an equal volume of 0.02 M ammonium acetate (pH 4), and the 125I-labeled products were identified by HPLC as described previously (22).

Mouse treatments

Materials and Methods Reagents and materials Unless otherwise specified, all reagents were purchased from Sigma (St. Louis, MO). Outer ring-labeled T4 (PerkinElmer, Boston, MA; specific activity 4400 Ci/mmol) was purified on LH-20 columns to remove free iodide before each use.

Animals All animal experimental protocols were approved by the Animal Research Committee of Harvard Medical School. C57BL/6J and CD-1 mice were purchased from The Jackson Laboratories (Bar Harbor, ME). TgD2 mice and D2KO/ C57BL/6J are as previously described (19 –21). D2KO mice were

Hypothyroidism Eight-week-old male mice were made hypothyroid by providing them drinking water containing 0.1% methimazole (MMI) and 1% NaClO4 (MMI/ClO4) for 4 (short term) or 8 wk (long term) (23). Because all the mice were at the same age at the initiation of the hypothyroidism induction (8 wk), the mice studied after 4 wk of treatment were 4 wk younger (12 wk) than the mice studied after 8 wk of treatment (16 wk).

Cold exposure Eight- to 12-wk-old male mice were housed one animal per cage with minimal bedding at 4 C for 16 h. Controls were housed at 22 C.

D2 Activity in Skeletal Muscle

Quantitative mRNA analysis via real-time PCR Muscle RNA was isolated using TRIzol reagent (Invitrogen, Carlsbad, CA), following the manufacturer’s instructions. Then 2.5 ␮g of RNA was used for cDNA synthesis using the VILO cDNA synthesis kit following the manufacturer’s specifications (Invitrogen). The cDNA samples were diluted 1:5 and analyzed with an iCycler (Bio-Rad Laboratories) using SYBR Green. Specific primers were designed to function under the same cycling conditions (95 C for 10 min followed by 40 cycles at 95 C for 15 sec and 60 C for 1 min), generating products of comparable sizes. For each reaction, standard curves for reference genes were constructed based on six 4-fold serial dilutions of cDNA. All samples were run in triplicate, using cyclophilin A as the housekeeping gene (24, 25). The cycle threshold (Ct) value for cyclophilin A within a run was not different in euthyroid and hypothyroid slow and fast muscles. The results were expressed as 2⫺⌬Ct (where ⌬Ct is the difference in the PCR cycles between D2 and cyclophilin A). Primers used were the following: dio2 sense, 5⬘-CTTCCTCCTAGATGCCTACAAAC-3⬘; dio2 antisense, 3⬘-GGCATAATTGTTACCTGATTCAGG-5⬘; cyclophilin A sense, 5⬘CGCCACTGTCGCTTTTCG-3⬘; cyclophilin A antisense, 3⬘-ACTTTGTCTGCAAACAGCTC-5⬘. Both sets of these primers span at least one intron.

Statistical analysis Statistical significance was determined by an unpaired Student’s t test or a one-way ANOVA followed by the NewmanKeuls post hoc analysis for multiple comparisons using Prism 4.0 software (GraphPad Software Inc., San Diego, CA). The results shown are mean ⫾ SEM.


D2 activity (fmol /min/mg homogen. prt.)

Marsili et al.

Optimization of the D2 assay for skeletal muscle Overnight incubation of 1 ␮g TgD2 microsomal protein in the absence of DTT or other cofactors caused nearly

0.15

0.10

**

0.05

P14

P1

FIG. 1. D2 activity in skeletal muscle of newborn and 2-wk-old mice. D2 activity was assayed in homogenates (250 ␮g protein) from pools of gastrocnemius muscles of CD-1 mice at P1 and P14 (mean ⫾ SEM, n ⫽ 3), using previously described methods (15). prt., Protein.

50% labeled iodide release with no T3 production, which was not saturated by high T4 concentrations (Fig. 2A). This result demonstrates that the T4 deiodination is not occurring through an enzymatic process and that the iodide released is nonspecific. Adding 20 mM DTT resulted in formation of labeled T3 and iodide in a ratio close to 1 (Fig. 2B). Both products were nearly eliminated by the addition of 500 nM T4, confirming that this is a saturable process. Although these conditions caused nearly com-

A

Results

75 50 25 0

I-

T2

T3

T4

I-

T2

T3

T4

100 % total counts

B

1 nMT4 500 nMT4

100

75 50 25 0

C D2 activity (fmol/min/mg protein)

D2 activity is present in muscle from newborn mice but falls rapidly over 2 wk Because D2 activity has been reported in myocytes isolated from neonatal mice (13), we assayed homogenates of skeletal muscles from CD-1 newborn mice as described (15). Robust D2 activity was present at postnatal d 1 (P1), but this had decreased sharply by P14 in both gastrocnemius (Fig. 1) and anterior tibialis (data not shown). Monodeiodination of outer ring-labeled T4 should yield equal quantities of labeled T3 and I⫺. HPLC analysis of reaction products from P14 mouse muscle showed that in addition to a decrease in T3 generated, there was an increase in the ratio of labeled iodide to T3 from about 1:1 to about 3:1, indicating some nonspecific release of I⫺. Therefore, to improve the assay, we attempted to maximize the specific activity and minimize nonspecific iodide release. To monitor these experiments, we used the D2KO or TgD2 microsomal protein described above.

0.20

0.00

% total counts

5954

0.08

**

homogenates microsomes

0.06 0.04 0.02 0.00

FIG. 2. Effects of DTT and protein preparation on D2 assay in muscle tissues. A and B, Assays of 1 ␮g TgD2 microsomal protein with 1 or 500 nM T4 without (A) or with (B) 20 mM DTT. C, Comparison of D2 activity after overnight incubation of 100 ␮g of homogenate (open bar) or microsomal (closed bar) protein prepared from anterior tibialis muscles of 8-wk-old C57/BL6 mice (mean ⫾ SEM, n ⫽ 4).



5955

the rate of the reaction was 40% lower with a longer incubation time. None35 35 25 *** 30 30 theless there was still a net gain in the 20 25 25 total T3 production with time (Fig. 3D) 20 20 15 ** ** 15 15 because there is no increase in the 125I 10 10 10 T3 between 4 and 16 h during incuba** 5 5 5 0 0 0 tion with D2KO microsomes. Thus, the 1 2 4 TgD2 (µg) 125 I T3 present in the D2KO assay is a D2KO (µg) 20 20 20 contaminant in the original T4. B D F 3.5 35 Because D3 will cause inner-ring 3.0 50 30 deiodination producing rT3 from T4 2.5 25 40 2.0 20 and 3, 3⬘ T2 from T3, it could reduce the *** 30 * 1.5 15 T4 available for 5⬘ deiodination as well 20 1.0 10 10 as convert the T3 formed to 3, 3⬘ T2. 5 0.5 0 0.0 0 Because D3 is present in human skeletal 16 0 100 50 4 4 TgD2 (µg) 4 4 D2KO (µg) T3 added (nM) 50 Incubation time (hours) 100 25 muscle (26), unlabeled T3 must be added to reduce or minimize this probFIG. 3. Effects of protein concentration, incubation time, and T3 addition on D2 activity. A and B, Effect on D2 activity of increasing the TgD2 protein added to a constant amount of lem (14). Depending on the level of D3 D2KO microsomal protein (A) vs. a constant amount of TgD2 incubated with increasing activity, the loss of T3 produced by conquantities of D2KO microsomal protein (B). Results shown are mean ⫾ SEM. *, P ⬍ 0.05; version to 3, 3⬘ T2 could lead to an unas indicated. C and D, Comparison of total (C) and fractional (D) D2-mediated T4-to-T3 conversion over 4 and 16 h using identical quantities of the same TgD2/D2KO skeletal muscle derestimation of the 5⬘ deiodination of microsome mixture. Results shown are mean ⫾ SEM of triplicates. ***, P ⬍ 0.001. E and F, T4 (Fig. 3E). On the other hand, because Effect of incubation with 0, 50, or 100 nM unlabeled T3 on the products of D2-mediated T4we can quantitate the T3 and 3,3⬘T2 by to-T3 conversion in the presence of skeletal muscle microsomes from D2KO mice containing 125 125 125 D3. Data are based on the net I T3 (E) or the sum of I T3, and I T2 (F) derived from HPLC, we can also correct for D3-me125 I T4 (mean ⫾ SEM). **, P ⬍ 0.001 for difference from T3-supplemented reactions. diated T3 to 3, 3⬘ T2 conversion by adding this product to the T3 formed (Fig. plete exhaustion of the T4 substrate, thus underestimating 3F). Alternatively, D3 can be blocked by unlabeled T3. the enzyme velocity, they showed that TgD2 had activity Because we found no difference between adding 50 or 100 of at least 250 fmol/min 䡠 mg microsomal protein. To nM T3, we decided to use 50 nM T3 (about 40 times the demonstrate that these conditions would also be approMichaelis constant) to eliminate these D3-mediated efpriate for skeletal muscle and to compare the different fects (27). approaches of protein preparation, we assayed mouse anIn summary, the major methodological improvements terior tibialis muscle homogenates and microsomes. Unwere a 4-fold increase in incubation time, an increase in the der these conditions, the D2 specific activity was about D2 specific activity of the protein assayed, and the use of 6-fold higher in microsomes than in homogenates (0.070 ⫾ 0.007 vs. 0.012 ⫾ 0.001 fmol/min 䡠 mg protein, D2KO skeletal muscle as a tissue-specific blank. T3 production (%)

E

T2+T3 production(%)

total T3 production (%)

C

T4 to T3 (%/h)

T4 to T3 (%)

T4 to T3 (%)

A

respectively; P ⬍ 0.01) (Fig. 2C). We used combinations of microsomal protein from the skeletal muscle of D2KO mice (20 –100 ␮g protein) enriched with TgD2 to analyze the assay performance. There was a linear increase in T3 production when increasing quantities of TgD2 were added to a constant amount of D2KO protein (Fig. 3A). In contrast, adding increasing quantities of D2KO microsomes to a constant amount of TgD2 reduced the T3 production efficiency by ⬃25% (Fig. 3B). This is likely to be due to a reduction of the free fraction of the T4 substrate in the buffer. While the efficiency of the D2 assay is decreased about 25% using 100 rather than 25 ␮g protein, the total activity generated is still 3 fold higher with the larger amount, indicating improved sensitivity. We next compared the T3 production after an overnight incubation (16 h) with that after 4 h (Fig. 3C). T3 production was only 2.5 fold higher after 16 h indicating that

D2 activity in mouse muscle samples using the improved method Based on the above results, we used the following procedure: 50 –100 ␮g of microsomal muscle protein are incubated for 16 h in 100 ␮l PE buffer (pH 7.0), containing 1 mM propylthiouracil, 20 mM DTT, 50 nM T3, 3 ⫻ 105 cpm 125I T4 (0.68 nM) plus 1 nM unlabeled T4. A matched quantity of microsomes from D2KO mouse skeletal muscle was used as a blank to determine the 125I T3 contaminating the T4 as well as any T3 that might be artifactually generated during the assay or HPLC. Under these conditions, the typical contamination was 1.6% ⫾ 0.01 of the total T4 added, and this was subtracted from the total T3 found. Using 100 ␮g of microsomal protein from a pool of soleus plus vastus intermedius muscles (largely slow

5956

Marsili et al.



TABLE 1. Comparison of the HPLC analyses of a D2 assay of 100 ␮g microsomal protein from a pool of WT or D2KO mouse mostly slow-twitch muscle (vastus intermedius ⫹ soleus; n ⫽ 4 animals per group) incubated for 16 h (mean ⫾ SEM of triplicates) Labeled product or substrate

WT (cpm)

KO (cpm)

I⫺ 3⬘ T1 3,3⬘ T2 3,5,3⬘ T3 rT3 T4 Total CPM

6150 ⫾ 454a 359 ⫾ 118 193 ⫾ 40b 3165 ⫾ 256a 1168 ⫾ 107 56017 ⫾ 792b 67052 ⫾ 559

3424 ⫾ 271 72 ⫾ 54 46 ⫾ 8 958 ⫾ 70 915 ⫾ 158 59772 ⫾ 842 65187 ⫾ 1044

Assay was performed in the presence of 50 nM unlabeled T3. WT, wild type; KO, knockout. a P ⬍ 0.01 vs. KO. b P ⬍ 0.05.

twitch muscles) from euthyroid 8- to 12-wk-old mice, the T3 production was about 2000 cpm over the D2KO background and readily quantifiable (Table 1). Iodide was typically approximately 5% of the total counts in D2KO microsomes and a similar level in the wild-type muscle. All conversion rates were corrected for the 50% reduction in the specific activity of the T3 due to 5⬘-monodeiodination of 3⬘,5⬘ 125I T4. In addition, the observed T3 production rate after 16 h incubation was corrected for the approximately 40% higher production rate during a 4- as opposed to a 16-h incubation (Fig. 3, C and D). This permits a comparison with results generated in other studies with homogenates that were generally incubated for 4 h. D2 activity is 5-fold higher in slow- as opposed to fast-twitch muscles Mouse hind limb muscles are a mixture of slow- and fast-twitch muscles with different characteristics and fiber composition (28, 29). D2 activity is approximately 5-fold higher in slow, red, type I, soleus muscle than in type II, white, fast-twitch or mixed muscles (P ⬍ 0.001) (Table 2). In agreement with this, in the largely red component of the quadriceps muscle (vastus intermedius), D2 activity was about 3-fold greater than in the white portion (vastus late-

ralis) (0.16 ⫾ 0.02 vs. 0.057 ⫾ 0.01 fmol/min 䡠 mg microsomal protein, respectively, P ⬍ 0.01). Hypothyroidism increases D2 activity in mouse skeletal muscle In previous studies of human and rat D2-expressing cells in culture, a decrease in the T4 substrate increases D2 activity due to a reduction in the rate of ubiquitination and proteasomal degradation (9, 30). The same occurs in intact rats (30). Four weeks of antithyroid drug treatment significantly reduced serum T4 but doubled the serum T3 and resulted in a 30-fold increase in TSH (Table 3). This was accompanied by a 35% increase in D2 activity (0.16 ⫾ 0.01 vs. 0.12 ⫾ 0.01 fmol/min 䡠 mg microsomal protein; P ⬍ 0.05, respectively) in fast-twitch muscles and a 45% increase (0.49 ⫾ 0.02 vs. 0.33 ⫾ 0.01; P ⬍ 0.001) in slow-twitch muscles (Fig. 4A). In slow twitch muscle, D2 mRNA was lower after short-term hypothyroidism (Fig. 4B and Table 3), perhaps related to the increase in serum T3 at that time. Eight weeks of treatment with antithyroid drugs decreased serum T4 to undetectable levels as well as causing a 50% fall in serum T3 (Table 3). This more severe hypothyroidism resulted in a 3-fold increase in D2 activity (P ⬍ 0.001) in both muscle types (0.28 ⫾ 0.03 vs. 0.10 ⫾ 0.01 and 1.56 ⫾ 0.07 vs. 0.52 ⫾ 0.01 fmol/min 䡠 mg microsomal protein, in fast and slow twitch muscles, respectively, Fig. 4C), again with no significant changes in D2 mRNA (Fig. 4D), confirming the increase is mostly due to a posttranslational effect. Overnight exposure to 4 C does not alter skeletal muscle D2 activity Cold exposure in rodents causes increased D2-mediated T4-to-T3 conversion in brown adipose tissue due to sympathetic stimulation. We evaluated the response of mouse skeletal muscle D2 to a similar stress. Housing animals at 4 C for 16 h did not change muscle D2 activity (Fig 5A) or mRNA, (Fig 5B), compared with room temperature (22 C) controls. In the same mice, brown adipose tissue (BAT) D2 activity increased from 3.2 ⫾ 0.8 to 29.9 ⫾ 1.1 fmol/min 䡠 mg protein.

TABLE 2. D2 activity (femtomoles per minute per milligram microsomal protein) of different hindlimb mouse skeletal muscles (mean ⫾ SEM)

a b

Muscle

Vastus lateralis

Gastrocnemius

Tibialis anterior

Vastus intermedius

Soleus

n Dominant fiber type D2 activity (fmol/min 䡠 mg)

6 Fast twitch 0.057 ⫾ 0.01

7 Fast twitch 0.07 ⫾ 0.01

10 Fast twitch 0.09 ⫾ 0.01

6 Fast and slow twitch 0.16 ⫾ 0.02a

6 Slow twitch 0.40 ⫾ 0.06b

P ⬍ 0.01 vs. fast muscles. P ⬍ 0.001 vs. all.



Treatment

Control

4 wk

8 wk

T4 (␮g per 100 ml) T3 (ng/ml) TSH (ng/ml)

1.65 ⫾ 0.13

0.55 ⫾ 0.18a

All ⬍ 0.5

0.87 ⫾ 0.05 1.7 ⫾ 0.31

1.65 ⫾ 0.10a 59.73 ⫾ 11.8b

0.36 ⫾ 0.02a N.D.

Animals were given 0.1% MMI plus 1% NaClO4 in drinking water. N.D., Not done. a P ⬍ 0.001 vs. euthyroid levels. b P ⬍ 0.05.

Discussion The assay of D2 activity in skeletal muscle is difficult for a number of reasons. It is a heterogeneous tissue, is not easily homogenized, and has much lower D2 expression than brain, pituitary, or BAT. Moreover, muscle cytosol, possibly due to its myoglobin content, creates an oxidative environment that enhances the nonspecific deiodination of T4 (31). This could be responsible for the nonspecific I⫺ release that has resulted in inappropriately high D2 estimates in some studies, which did not monitor T3 production by HPLC (9, 32). We found that for analysis of muscle samples with high D2 expression, such as those of newborn mice, homogenates can readily be used (Fig. 1). Assaying adult muscle,

D2 activity (fmol/min/mg microsomal prt.)

MMI ⴙ NaClO4

A = controls 0.35 0.30

ns

=4C

ns

0.25 0.20 0.15 0.10 0.05 0.00

B mRNA D2/cyclo A

TABLE 3. Serum thyroid hormone and TSH concentrations in control and MMI/NaClO4-treated mice after 4 or 8 wk (mean ⫾ SEM; n ⫽ 10 for each group)

5957

ns 1.4 1.2 1.0 0.8 0.6 0.4

ns

0.2 0.0

fast twitch

slow twitch

FIG. 5. Effect of cold exposure on D2 activity in mouse skeletal muscle. D2 activity (A) and quantitative RT-PCR D2/cyclophilin A mRNA (B) from fast- and slow-twitch muscles of 8-wk-old-C57/BL6 mice at room temperature or kept at 4 C for 16 h (n ⫽ 5 mice per group). Results shown are mean ⫾ SEM. ns, Not significant.

D2 activity (fmol /min/mg microsomal prt.)

D2 activity (fmol /min/mg microsomal prt.)

mRNA D2/cyclo A

mRNA D2/cyclo A

in which D2 is much lower, is better performed using microsomes, which reduce the cytosolic contribution to the nonspecific iodide release as well as increase the specific activity of D2 about 6-fold. Unfortunately, the yield of microsomes from muscle is not high, about 250 ␮g protein per 100 mg wet A C weight, which is a limitation when only = controls = controls small samples are available. *** = short term = long term *** 0.6 1.75 The use of TgD2 and the D2KO skelhypothyroid hypothyroid 1.50 0.5 etal muscle microsomes was quite help1.25 0.4 ful in the evaluation of assay variables. *** 1.00 0.3 There was no T4-to-T3 conversion be* 0.75 tween 4 and 16 h incubation in micro0.2 0.50 somes from D2KO mice. The TgD2 ob0.1 0.25 tained from cardiac muscle is similar to 0.0 0.00 skeletal muscle, and its high specific acB D ** 1.2 1.2 tivity allows the use of minute amounts 1.0 1.0 of myocardial microsomal protein (⬍1 0.8 part per 1000) still resulting in robust 0.8 0.6 activity, facilitating the methodological 0.6 0.4 analyses. Thus, we found that DTT vir0.4 tually eliminates nonspecific iodide re0.2 0.2 lease (Fig. 2B). We also found that D2 0.0 0.0 slow twitch activity was linear when TgD2 was fast twitch slow twitch fast twitch added to the same amount of D2KO FIG. 4. Hypothyroidism increases the D2 activity in mouse skeletal muscles by a posttranslational mechanism. D2 activity (A and C) and quantitative RT-PCR D2/cyclophilin A protein. Increasing the amount of total mRNA (B and D) from fast- and slow-twitch muscles of 8-wk-old-C57/BL6 mice euthyroid microsomal protein from 25 to 100 ␮g receiving either 4 wk (A and B) or 8 (C and D) wk of MMI/ClO4 treatment (n ⫽ 10 mice per modestly reduces the efficiency of the group). Results shown are mean ⫾ SEM. *, P ⬍ 0.05, **, P ⬍ 0.01, ***, P ⬍ 0.001 vs. D2 reaction (Fig. 3, A and B). Thus, the euthyroid. Please note the difference in the y-axis scale between panels A and C.

5958

Marsili et al.


net T3 counts formed will still be higher using 100 as opposed to 50 ␮g of muscle microsomes. Catalysis by D2 was not linear with time between 4 and 16 h, the average rate of the reaction decreasing from 3% conversion/h over 4 h to 1.7%/h over 16 h (Fig. 3D). We have also noted a similar slow decrease in D2 activity of this magnitude during the assay of D2 in homogenates of D2-expressing cells (Larsen, P. R., and J. H. Harney, unpublished data). This is probably due to oxidation, although ubiquitination might also play a role (33, 34). Nonetheless, because the net 125I T3 production over background is still substantially higher after 16 h, the longer assay increases the sensitivity (Fig. 3C). Using the above conditions, we found production of 0.40 ⫾ 0.06 fmol T3/min/mg microsomal protein (mean ⫾ SEM) in oxidative, type-I, slow twitch soleus and much lower activities (0.076 ⫾ 0.01 fmol/min/mg microsomal protein) in type II fast-twitch muscles (namely vastus lateralis, gastrocnemius and anterior tibialis). Even in the same quadriceps muscle, the vastus intermedius which has an higher content of type-I fibers than its white component, the vastus lateralis, has significantly more D2 activity (28). There are three published estimates of D2 activity in mouse muscle using homogenates. In one, the authors concluded that D2 is undetectable in adult murine quadriceps (13). In the other two papers, both from the same authors, values between undetectable and 0.02 fmol/min/mg protein were found (17, 35). Our results show detectable and reproducibly measurable D2 activities in all muscles that we studied; taking the average of the these activities, levels are at least 10 –35 fold higher than in (17), which can only in part be explained by the 6-fold higher D2 specific activity of microsomes than homogenates (Fig. 2C). By introducing additional changes in the microsome preparation technique it is conceivable that even higher specific activities can be achieved (36). There are several reports of D2 activity in homogenates of human skeletal muscle evaluated by HPLC. D2 activity was not measurable in the rectus abdominis muscle (a mixture of slow and fast fibers) collected postmortem (26, 37); in other studies D2 activity was measurable in 50% of the quadriceps muscle biopsies, with levels of about 0.009 fmol/min 䡠 mg protein in healthy subjects and not different from that in critically ill patients (16). Moreover, D2 activity in rectus abdominis was about 0.003 fmol/min 䡠 mg, with a significant increase (3-fold) in critically ill patients (15). Recently D2 activity in hypothyroid human vastus lateralis muscle has been reported to be 0.007 fmol/ min 䡠 mg protein, with no difference in the same patients when replaced with T4 (14). Taken together, these values are much lower than those we found in mice, even taking into account, for the human study, the fact that the vastus lateralis assayed has about 40% type I oxidative fibers (38). Again, the difference cannot be explained by the use


of microsomes instead of homogenates because this can account for only about a 6-fold specific activity increase. We have applied this method to human skeletal muscles, obtaining significantly higher results than previously reported (work in progress). We do not know what factor or factors causes the higher D2 activity in euthyroid slow- as opposed to fasttwitch muscles (Table 2). There is a previous report of higher D2 mRNA in euthyroid red vs. white muscles (12), which we have confirmed (Fig. 4), perhaps due to increased Dio2 gene transcription. The higher D2 activity in slow muscle may explain the observation of higher T3 tissue content in red compared with white muscles of euthyroid rats despite a similar T4 content (39). Slow-twitch muscles have significant differences from fast muscles; they have predominant oxidative metabolism and higher content of myosin heavy chain I, sarcoendoplasmic reticulum Ca2⫹-ATPase-2, and myogenin than fast fibers. Slow muscles are more responsive to thyroid hormones and also have an higher metabolic rate (2). Our current data suggest that D2-mediated T4-to-T3 conversion may be a more important source of T3 in slow than fast muscles. Previous reports showed an up-regulation of brain and pituitary D2 activity during hypothyroidism or iodine deficiency in rats, this response being proportional to the duration of the treatment (40). In agreement with this, we found an increase in skeletal muscle D2 activity of about 40% after 4 wk of treatment with methimazole and NaClO4, and about 3-fold after 8 wk (Fig. 4, A and C). The increase in D2 activity was not accompanied by increased D2 mRNA expression, as assessed by quantitative RTPCR analysis (Fig. 4, B and D), consistent with the increase being mostly due to a posttranslational effect, presumably due to an increase in the half-life of D2 protein secondary to decreased ubiquitination and proteasomal degradation (7, 18, 41). After short-term hypothyroidism, serum T3 was significantly higher than in controls, suggesting increased thyroidal T3 secretion stimulated by the increased TSH combined with intrathyroidal iodine deficiency (Table 3). Prolonging the treatment leads to a reduction of both serum T3 and T4 (Table 3). However, there is still no increase in D2 mRNA despite the 3-fold increase in D2 activity (Fig. 4, C and D). These data are in contrast with recent studies in humans, in which severe hypothyroidism did not affect either D2 activity or mRNA in muscle (14). There is no clear explanation for this apparent species difference. Interestingly, previous studies showed that reducing medium T4 increased D2 activity in human embryonic kidney cells transfected with human D2 as well as in the human mesothelioma cell, MSTO 211H, expressing endogenous D2. This suggests that there is no difference between human


and rodent D2 in terms of posttranslational regulation by T4 (9). Additionally, in a previous human study, D2 activity was up-regulated in critically ill patients in association with low T4 and T3 levels and an elevated TSH (15). Exposing mice to 4 C for 16 h did not increase D2 activity or mRNA in either fast or slow muscle; however, in the same animals, BAT D2 activity is 10-fold increased (up to ⬃30 fmol/min 䡠 mg protein), establishing the presence of the expected physiological stimulatory response. D2 mRNA in BAT was also increased to the same extent as D2 activity. We did not assess D2 activity over shorter or longer intervals of cold exposure, so that a role for muscle D2 in the short-term thermogenic response cannot definitively be ruled out. It is notable that the ␣T3-receptor-deficient mice (Thra-0/0), which have impaired BAT thermogenesis and have higher D2 mRNA (12) and activity (36) in red muscles compared with wild-type animals. Despite the lack of effect of cold exposure on D2 activity in mice, a contribution of skeletal muscle D2 to nonshivering facultative thermogenesis is possible in larger mammals, in which BAT is less important for thermogenesis, and this will require further investigation. In conclusion, using a sensitive D2 assay using microsomal protein, we found substantially higher levels of D2 activity in mouse skeletal muscle than previously reported. D2 mRNA and activity are higher in slow than fast muscle. D2 activity, but not mRNA, increases during hypothyroidism (as in other D2 containing tissues like brain, pituitary, and BAT), suggesting a posttranslational regulation by T4, but it is not stimulated by overnight cold exposure. This approach to D2 assay in skeletal muscle should be a useful tool for further studies to clarify the metabolic relevance of the fiber specificity of D2 expression and the contribution of local T3 production to muscle physiology.

Acknowledgments Special thanks go to Ting Chen and Thuy Van Tran for technical assistance and Dr. Chandrajit Raut for muscle specimens. Address all correspondence and requests for reprints to: P. Reed Larsen, M.D., Thyroid Section, Division of Endocrinology, Diabetes, and Hypertension, Brigham and Women’s Hospital, Harvard Institutes of Medicine, Room 644, 77 Avenue Louis Pasteur, Boston, Massachusetts 02115. E-mail: plarsen@ partners.org. This work was supported by the National Institutes of Health Grants DK44128, DK076099, and FIC TW007559 and Conselho Nacional de Desenvolvimento Cientifico e Tecnologico, Brazil. A.M. was partially supported by a fellowship stipend from the Department of Endocrinology and Kidney, University Hospital of Pisa, Pisa, Italy.


5959

Disclosure Summary: S.A.H., A.M.Z., A.L.M., and P.R.L. are supported by National Institutes of Health grants. The other authors have nothing to disclose.

References 1. Lowell BB, Spiegelman BM 2000 Towards a molecular understanding of adaptive thermogenesis. Nature 404:652– 660 2. Simonides WS, van Hardeveld C 2008 Thyroid hormone as a determinant of metabolic and contractile phenotype of skeletal muscle. Thyroid 18:205–216 3. 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 4. Silva JE 2006 Thermogenic mechanisms and their hormonal regulation. Physiol Rev 86:435– 464 5. Cle´ment K, Viguerie N, Diehn M, Alizadeh A, Barbe P, Thalamas C, Storey JD, Brown PO, Barsh GS, Langin D 2002 In vivo regulation of human skeletal muscle gene expression by thyroid hormone. Genome Res 12:281–291 6. Visser WE, Heemstra KA, Swagemakers SM, Ozgur Z, Corssmit EP, Burggraaf J, van Ijcken WF, van der Spek PJ, Smit JW, Visser TJ 2009 Physiological thyroid hormone levels regulate numerous skeletal muscle transcripts. J Clin Endocrinol Metab 94:3487–3496 7. Bianco AC, Salvatore D, Gereben B, Berry MJ, Larsen PR 2002 Biochemistry, cellular and molecular biology, and physiological roles of the iodothyronine selenodeiodinases. Endocr Rev 23:38 – 89 8. Nguyen TT, Chapa F, DiStefano 3rd JJ 1998 Direct measurement of the contributions of type I and type II 5⬘-deiodinases to whole body steady state 3,5,3⬘-triiodothyronine production from thyroxine in the rat. Endocrinology 139:4626 – 4633 9. 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 10. Salvatore D, Bartha T, Harney JW, Larsen PR 1996 Molecular biological and biochemical characterization of the human type 2 selenodeiodinase. Endocrinology 137:3308 –3315 11. Croteau W, Davey JC, Galton VA, St. Germain DL 1996 Cloning of the mammalian type II iodothyronine deiodinase. A selenoprotein differentially expressed and regulated in human and rat brain and other tissues. J Clin Invest 98:405– 417 12. Pelletier P, Gauthier K, Sideleva O, Samarut J, Silva JE 2008 Mice lacking the thyroid hormone receptor-␣ gene spend more energy in thermogenesis, burn more fat, and are less sensitive to high-fat dietinduced obesity. Endocrinology 149:6471– 6486 13. Grozovsky R, Ribich S, Rosene ML, Mulcahey MA, Huang SA, Patti ME, Bianco AC, Kim BW 2009 Type 2 deiodinase expression is induced by peroxisomal proliferator-activated receptor-gamma agonists in skeletal myocytes. Endocrinology 150:1976 –1983 14. Heemstra KA, Soeters MR, Fliers E, Serlie MJ, Burggraaf J, van Doorn MB, van der Klaauw AA, Romijn JA, Smit JW, Corssmit EP, Visser TJ 2009 Type 2 iodothyronine deiodinase in skeletal muscle: effects of hypothyroidism and fasting. J Clin Endocrinol Metab 94: 2144 –2150 15. 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 16. Rodriguez-Perez A, Palos-Paz F, Kaptein E, Visser TJ, DominguezGerpe L, Alvarez-Escudero J, Lado-Abeal J 2008 Identification of molecular mechanisms related to nonthyroidal illness syndrome in skeletal muscle and adipose tissue from patients with septic shock. Clin Endocrinol (Oxf) 68:821– 827 17. Kwakkel J, van Beeren H, Ackermans M, Platvoet-Ter Schiphorst M, Fliers E, Wiersinga W, Boelen A 2009 Skeletal muscle deiodinase type 2 regulation during illness in mice. J Endocrinol 203:263–270 18. Steinsapir J, Bianco AC, Buettner C, Harney J, Larsen PR 2000

5960

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

Marsili et al.


Substrate-induced down-regulation of human type 2 deiodinase (hD2) is mediated through proteasomal degradation and requires interaction with the enzyme’s active center. Endocrinology 141: 1127–1135 Christoffolete MA, Linardi CC, de Jesus L, 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 coldinduced overexpression of the uncoupling protein 1 gene but fail to increase brown adipose tissue lipogenesis and adaptive thermogenesis. Diabetes 53:577–584 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 Pachucki J, Hopkins J, Peeters R, Tu H, Carvalho SD, Kaulbach H, Abel ED, Wondisford FE, Ingwall JS, Larsen PR 2001 Type 2 iodothyronin deiodinase transgene expression in the mouse heart causes cardiac-specific thyrotoxicosis. Endocrinology 142:13–20 Richard K, Hume R, Kaptein E, Sanders JP, van Toor H, De Herder WW, den Hollander JC, Krenning EP, Visser TJ 1998 Ontogeny of iodothyronine deiodinases in human liver. J Clin Endocrinol Metab 83:2868 –2874 Zavacki AM, Ying H, Christoffolete MA, Aerts G, So E, Harney JW, Cheng SY, Larsen PR, Bianco AC 2005 Type 1 iodothyronine deiodinase is a sensitive marker of peripheral thyroid status in the mouse. Endocrinology 146:1568 –1575 Castillo M, Freitas BC, Rosene ML, Drigo RA, Grozovsky R, Maciel RM, Patti ME, Ribeiro MO, Bianco AC 2010 Impaired metabolic effects of a thyroid hormone receptor ␤-selective agonist in a mouse model of diet-induced obesity. Thyroid 20:545–553 Freitas BC, Gereben B, Castillo M, Kallo´ I, Zeo¨ld A, Egri P, Liposits Z, Zavacki AM, Maciel RM, Jo S, Singru P, Sanchez E, Lechan RM, Bianco AC 2010 Paracrine signaling by glial cell-derived triiodothyronine activates neuronal gene expression in the rodent brain and human cells. J Clin Invest 120:2206 –2217 Peeters RP, Wouters PJ, Kaptein E, van Toor H, Visser TJ, Van den Berghe G 2003 Reduced activation and increased inactivation of thyroid hormone in tissues of critically ill patients. J Clin Endocrinol Metab 88:3202–3211 Salvatore D, Low SC, Berry M, Maia AL, Harney JW, Croteau W, St. Germain DL, Larsen PR 1995 Type 3 lodothyronine deiodinase: cloning, in vitro expression, and the functional analysis of the placental selenoenzyme. J Clin Invest 96:2421–2430 Hitomi Y, Kizaki T, Watanabe S, Matsumura G, Fujioka Y, Haga S, Izawa T, Taniguchi N, Ohno H 2005 Seven skeletal muscles rich in slow muscle fibers may function to sustain neutral position in the rodent hindlimb. Comp Biochem Physiol B Biochem Mol Biol 140: 45–50 Ariano MA, Armstrong RB, Edgerton VR 1973 Hindlimb muscle fiber populations of five mammals. J Histochem Cytochem 21:51–55


30. Burmeister LA, Pachucki J, St Germain DL 1997 Thyroid hormones inhibit type 2 iodothyronine deiodinase in the rat cerebral cortex by both pre- and posttranslational mechanisms. Endocrinology 138: 5231–5237 31. Tseng YL, Latham KR 1984 Iodothyronines: oxidative deiodination by hemoglobin and inhibition of lipid peroxidation. Lipids 19:96 – 102 32. 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 33. Gereben B, Zavacki AM, Ribich S, Kim BW, Huang SA, Simonides WS, Zeo¨ld A, Bianco AC 2008 Cellular and molecular basis of deiodinase-regulated thyroid hormone signaling. Endocr Rev 29: 898 –938 34. Zavacki AM, Arrojo E Drigo R, Freitas BC, Chung M, Harney JW, Egri P, Wittmann G, Fekete C, Gereben B, Bianco AC 2009 The E3 ubiquitin ligase TEB4 mediates degradation of type 2 iodothyronine deiodinase. Mol Cell Biol 29:5339 –5347 35. Kwakkel J, Chassande O, van Beeren HC, Fliers E, Wiersinga WM, Boelen A 2010 Thyroid hormone receptor ␣ modulates lipopolysaccharide-induced changes in peripheral thyroid hormone metabolism. Endocrinology 151:1959 –1969 36. Ramadan W HS, Marsili A, Larsen P, Silva J Mouse skeletal muscle microsomes have type-2 iodothyronine deiodinase (D2) mRNA and activity. Possible role in hypermetabolism of TR deficient mice. Program of the 80th Annual Meeting of the American Thyroid Association, Palm Beach, FL, 2009, p 177 (Abstract O65) 37. Peeters RP, Wouters PJ, van Toor H, Kaptein E, Visser TJ, Van den Berghe G 2005 Serum 3,3⬘,5⬘-triiodothyronine (rT3) and 3,5,3⬘triiodothyronine/rT3 are prognostic markers in critically ill patients and are associated with postmortem tissue deiodinase activities. J Clin Endocrinol Metab 90:4559 – 4565 38. Staron RS, Hagerman FC, Hikida RS, Murray TF, Hostler DP, Crill MT, Ragg KE, Toma K 2000 Fiber type composition of the vastus lateralis muscle of young men and women. J Histochem Cytochem 48:623– 629 39. Janssen JW, van Hardeveld C, Kassenaar AA 1979 A methodological study in measuring T3 and T4 concentration in red and white skeletal muscle and plasma of euthyroid rats. Acta Endocrinol (Copenh) 90:81– 89 40. 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 41. Cheron RG, Kaplan MM, Larsen PR 1979 Physiological and pharmacological influences on thyroxine to 3,5,3⬘-triiodothyronine conversion and nuclear 3,5,3⬘-triiodothyronine binding in rat anterior pituitary. J Clin Invest 64:1402–1414