(Thr92Ala) Polymorphism Is Associated with Decreased Enzyme ...

0021-972X/05/$15.00/0 Printed in U.S.A.

The Journal of Clinical Endocrinology & Metabolism 90(6):3472–3478 Copyright © 2005 by The Endocrine Society doi: 10.1210/jc.2004-1977

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 Luis Henrique Canani, Clarissa Capp, Jose´ Miguel Dora, Erika L. Souza Meyer, Ma´rcia S. Wagner, John W. Harney, P. Reed Larsen, Jorge L. Gross, Antonio C. Bianco, and Ana Luiza Maia Endocrine Division (L.H.C., C.C., J.M.D., E.L.S.M., M.S.W., J.L.G., A.L.M.), Hospital de Clıńicas de Porto Alegre, Universidade Federal do Rio Grande do Sul, 90035-003 Porto Alegre, RS, Brazil; and Division of Endocrinology (J.W.H., P.R.L., A.C.B.), Diabetes and Hypertension, Brigham and Women’s Hospital and Harvard Medical School, Boston, Massachusetts 02115 The single-nucleotide polymorphism A/G in the type 2 deiodinase (D2) gene predicts a threonine (Thr) to alanine (Ala) substitution at codon 92 (D2 Thr92Ala) and is associated with insulin resistance in obese patients. Here, this association was investigated in 183 patients with type 2 diabetes mellitus, using homeostasis model assessment. The median fasting plasma insulin in Ala/Ala individuals was significantly higher than in patients with Ala/Thr or Thr/Thr genotypes (19.6 vs. 12.0 vs. 14.8 mIU/ml, respectively; P ⫽ 0.004). Assuming a recessive model, the homeostasis model assessment index was higher in the Ala/Ala group when compared with Ala/Thr-Thr/Thr group (8.50 vs. 4.85, P ⫽ 0.003). Although this polymorphism has not been

T

HE IODOTHYRONINE DEIODINASES types 1, 2, and 3 (D1, D2, and D3) constitute a family of oxidoreductases that catalyze the removal of iodine from the outer ring (D1 and D2, activation) or inner ring (D3, inactivation) of the thyroid hormones (1). All three deiodinases contain the uncommon amino acid selenocysteine in their active site, an essential residue for efficient catalysis (2– 4), and recently they have been identified as members of the thioredoxin-fold superfamily (5). T4 activation to T3 via D2 allows for changes in intracellular thyroid status in a tissue-specific manner independent of serum T3. This system has been extensively characterized in the brain, pituitary, and brown adipose tissue (BAT). As a result, mice with targeted disruption of the Dio2 gene have been found to have increased serum TSH levels, severe hearing impairment, and defective BAT adaptive thermogenesis (6 – 8). A role for D2 in human thermogenesis has been largely ignored because adult humans do not have substantial amounts of BAT. However, after cloning the human D2 cDNA, it has become clear that, unlike small mammals, D2 mRNA and activity are present in human heart and skeletal muscle (9, 10). Indirect clinical evidence supports the exciting First Published Online March 29, 2005 Abbreviations: BAT, Brown adipose tissue; BMI, body mass index; DM2, type 2 diabetes mellitus; DTT, dithiothreitol; HOMA, homeostasis model assessment; Vmax, maximum velocity; WHR, waist to hip ratio. JCEM is published monthly by The Endocrine Society (http://www. endo-society.org), the foremost professional society serving the endocrine community.

associated with changes in D2 kinetics as measured in HEK-293 cells transiently expressing D2 Thr92Ala, we investigated whether such association could be detected in human tissue samples. Remarkably, in thyroid and skeletal muscle samples from subjects homozygous for the Ala allele, D2 velocity was significantly lower than in subjects with Ala/Thr-Thr/Thr genotypes (P ⫽ 0.05 and 0.04, respectively). In conclusion, the A/G polymorphism is associated with greater insulin resistance in type 2 diabetes mellitus patients and with lower D2 velocity in tissue samples. These findings suggest that the D2generated T3 in skeletal muscle plays a role in insulin resistance. (J Clin Endocrinol Metab 90: 3472–3478, 2005)

possibility that D2 might play a previously unrecognized role in determining the thyroid status and metabolic activity of skeletal muscle, analogous to its role in BAT (11, 12). Chronic terbutaline treatment in adult humans was found to increase both resting energy expenditure (13) and the serum T3 to T4 ratio (11), suggesting the existence of an adrenergicdependent T4 to T3 conversion pathway. That this pathway is predominantly through D2 is supported by the fact that its gene is up-regulated severalfold by adrenergic stimulators and cAMP (14). In addition, in patients receiving T4 replacement at varying dosages, resting energy expenditure correlated directly with free T4 and inversely with serum TSH but, interestingly, not with serum T3 (12). These data are consistent with a role for T4 via D2-dependent intracellular T3 production in skeletal muscle as a significant physiological determinant of energy expenditure in humans. Although it is difficult to investigate the role of D2 in controlling fuel use and energy expenditure in healthy individuals, clinical studies with type 2 diabetes mellitus (DM2) patients offer a practical approach because these individuals require intense metabolic monitoring and are subjected to detailed scrutiny of fuel use. DM2 is a heterogeneous group of disorders usually characterized by varying degrees of insulin resistance and increased blood glucose concentrations. Ultimately, insulin resistance results either from uninhibited hepatic gluconeogenesis or decreased glucose disposal rate in tissues such as skeletal muscle and adipose tissue. GLUT4, the insulin-responsive glucose transporter, mediates the rate-limiting step of glucose metabolism

3472

Canani et al. • Deiodinase Polymorphism and Insulin Resistance

in skeletal muscle and adipose tissue (15). GLUT4 expression is up-regulated by thyroid hormone, and its overexpression in insulin-resistant db/db mice fed a high-fat diet dramatically improves glycemic control (15, 16). Recent studies have shown that specific polymorphisms in the deiodinase genes may have pathophysiological effects, interfering in the phenotypic expression of these enzymes (17). Of particular interest was a study describing a Dio2 single nucleotide polymorphism A/G polymorphism in humans, in which a threonine (Thr) change to alanine (Ala) at codon 92 (D2 Thr92Ala) was associated with an approximately 20% lower glucose disposal rate (18). The frequency of the variant allele was also found to be increased in some ethnic groups, such as Pima Indians and Mexican-Americans, who also have a higher prevalence of insulin resistance (18). In the present study, we demonstrate that although D2 activity in vitro is not affected by the Thr92Ala mutation per se, the Ala allele variant was associated with lower D2 velocity in human thyroid and skeletal muscle samples. Furthermore, the Ala/Ala genotype was associated with more severe insulin resistance in a sample population of DM2 patients. This suggests that D2 catalyzed conversion of T4 to T3 might play a role in determining insulin sensitivity, glucose use, and, hence, obesity in humans. Patients and Methods Study population The type 2 deiodinase Thr92Ala polymorphism was analyzed in blood samples from 183 DM2 patients attending the Endocrine Division’s outpatient clinic at Hospital de Clıńicas de Porto Alegre (Brazil). A standard questionnaire was used to collect information about age, age at DM diagnosis, and drug treatment. All patients underwent a complete physical examination and laboratory tests. They were weighed without shoes and in light outdoor clothes and had their height measured. Body mass index (BMI) was calculated as weight (kilograms) divided by the square of height (meters squared). To calculate waist to hip ratio (WHR), waist circumference was measured at the narrowest point, as viewed from the front, and hip circumference was measured at the widest point. Blood pressure was measured twice in the sitting position after a 10-min rest by means of a mercury sphygmomanometer (Korotkoff phases I and V). Hypertension was confirmed with blood pressure more than 140/90 mm Hg, or if the patient was taking antihypertensive drugs. Patients with elevated serum creatinine [⬎1.5 mg/dl (132.6 mmol/liter)], and those receiving insulin therapy were excluded. Samples of thyroid tissues were obtained from patients attending the Endocrine or the Head and Neck Surgery Divisions at Hospital de Clıńicas de Porto Alegre who had thyroidectomy because of benign or malignant thyroid nodules. The samples were collected from normal contralateral thyroid lobe or from tissue of focal thyroid neoplasm. Attending physicians not involved in the study had independently indicated surgery, and the tissues obtained were immediately frozen and stored until analysis. Skeletal muscle sample biopsies (sternocleidomastoid and rectus abdomenus) were obtained during routine surgery procedures performed by the same general surgery team. Exclusion criteria were a diagnosis of diabetes mellitus or impaired glucose intolerance, metastatic cancer, and hyper- or hypothyroidism. Samples collected were immediately snap-frozen in liquid nitrogen and stored at ⫺70 C until analysis. The information obtained from the study did not influence the patients’ diagnosis or treatment. The local Ethics Committee approved both protocols, and all patients signed an informed consent form.

Laboratory analysis A serum sample was collected after a 12-h fast. Glucose levels were determined by a glucose oxidase method, creatinine by the Jaffe´ reaction,

J Clin Endocrinol Metab, June 2005, 90(6):3472–3478

3473

glycosylated hemoglobin by an ion-exchange HPLC procedure (MerckHitachi L-9100 glycated hemoglobin analyzer, Merck, Darmstadt, Germany; reference range: 2.7– 4.3%), and triglyceride and cholesterol levels by enzymatic methods. Serum insulin was measured by radioassay (ElecsysR Systems 1010/2010/modular analytics E170, Roche Diagnostics, Indianapolis, IN). Low-density lipoprotein-cholesterol was calculated using the Friedewald equation. Insulin sensitivity was estimated by homeostasis model assessment [HOMA ⫽ fasting insulin (milliunits per milliliter) ⫻ fasting glucose (millimoles per liter)/22.5], as recently described and validated (19). The mean HOMA value of control subjects in our laboratory was 1.84 ⫾ 1.02 (20).

Genotype analysis Genotyping of the A/G polymorphism was performed by restriction fragment-length polymorphism analyses. Briefly, PCR was carried out in 96-well microtiter plates with a final reaction volume of 25 ␮l using specific oligonucleotides derived from published sequences of the human Dio2 gene, resulting in a predicted 256-pb fragment (18). Each PCR contained 50 –100 ng genomic DNA, 1 ␮mol/liter of each primer, 2 mmol/liter MgCl2, 50 mmol/liter KCl, 20 mmol Tris-HCl, 0.2 mmol/ liter dNTPs, and 1 U Taq polymerase (Invitrogen, Carlsbad, CA). The reaction conditions were as follows: an initial denaturation step at 94 C for 2 min, followed by 94 C for 1 min, annealing at 60 C, and extension at 72 C for 2 min. Thirty cycles were used, with a final additional extension step at 72 C for 4 min. Ten microliters of PCR product was then incubated at 37 C for 3 h with 5 U BsgI restriction enzyme. The product of the enzymatic digestion was analyzed on a 1.5% agarose gel stained with ethidium bromide and visualized by optical densitometry (ImageMaster VDS, Amersham Biosciences Inc., Piscataway, NJ). Individuals homozygous for the Thr-encoding allele were displayed as an uncut 256-pb fragment and the Ala allele were shown as a doublet of 186- and 70-pb bands. Patients were classified in groups of Thr/Thr or Ala/Thr and Ala/Ala according to the presence of the alleles. All amplification reactions were performed twice.

5⬘-Deiodinase assays D2 assay was performed as previously described (21). Briefly, tissue samples or cell sonicates were homogenized in 0.25 m sucrose in PE buffer (0.1 m potassium phosphate and 1 mm EDTA) with 10 mm dithiothreitol (DTT). The reaction mixtures containing 50 –300 ␮g tissue protein were incubated in a total volume of 300 ␮l with 125I-T4 (Amersham Biosciences Inc.) purified by LH-20 column chromatography (Pfizer, Inc., Ta¨by, Sweden), 1 or 4 nm unlabeled T4, 20 mm DTT, and 1 mm propylthiouracil in PE buffer at 37 C for 2 or 5 h. In kinetic experiments, T4 concentrations of 0.5–20 nm were used as substrate, and kinetic constants were calculated using double reciprocal plots. Reactions were terminated by the addition of 200 ␮l horse serum and 100 ␮l 50% trichloroacetic acid. After centrifugation at 13,000 rpm for 2 min, free 125I in the supernatant was counted with a ␥-scintillation counter. D1 assay was performed at the same conditions described above using 1 ␮m unlabeled T4 as substrate, 10 mm DTT, and no propylthiouracil. All reactions were performed in duplicate.

Tissue preparation and RNA isolation Thyroid tissue samples were immediately snap-frozen in liquid nitrogen and stored at ⫺70 C until RNA preparation. Total RNA was isolated from 50 –100 mg thyroid nontumor or skeletal muscle tissues using TRIzol reagent (Invitrogen) according to the manufacturer’s instructions. The purity and integrity of total RNA was assessed by UV spectrophotometer (GeneQuant II, Amersham Biosciences, Inc.) and denaturing agarose gel electrophoresis.

RT-PCR A semiquantitative RT-PCR technique was used to determine the expression levels of the D2 mRNA in samples of RNA isolated from thyroid tissues. RT-PCR was performed using the Superscript Preamplification System for First Strand cDNA Synthesis (Invitrogen) with 3 ␮g total RNA as template. Specific oligonucleotides derived from the coding region of published sequences of the human Dio2 gene (5⬘-

3474


ACTCGGTCA TTCTGCTCAAG-3⬘ and 5⬘-GAGAACTCTTCCACCAGTTTG-3⬘) were used to prime target cDNA, resulting in a predicted 368-bp product (22). Human ␤ 2-microglobulin primers set (5⬘-ATCCAGCGTACTCCAAAGATTCAG-3⬘ and 5⬘-AAATTGAAAGTTAACTTATGCACGC-3⬘), which generates a 623-bp product, was used as an internal control. Both fragments were coamplified with ␤2-microglobulin within the same reaction to evaluate intersample variation in cDNA content and PCR efficiency. A preliminary PCR amplification was performed to determine the range of cDNA concentrations and number of cycles over which the gene expression should be examined before reaching a plateau. The PCR included 1 ␮l RT product for Dio2 mRNA amplification and were carried with Taq DNA polymerase (Invitrogen) in a final 50-␮l volume. The amplification protocol had an initial step at 94 C for 3 min, followed by 94 C for 1 min, annealing at 60 C (D2), and extension at 72 C for 2 min. Thirty cycles were used, with a final additional extension step at 72 C for 4 min. The ␤2-microglobulin primers were included after the 10th cycle for D2. RT-PCR without cDNA samples was carried out as negative controls. All reactions were performed in duplicate. After amplification, 10 ␮l PCR products was analyzed on a 1.5% agarose gel stained with ethidium bromide and the densitometric results of D2 bands (arbitrary units) were normalized against the corresponding values of the ␤2-microglobulin band intensities. Densitometric quantification of the ethidium bromide stained bands was carried out using the ImageMaster VDS (Amersham Biosciences Inc.).

Expression studies in mammalian cells Site-directed mutagenesis was performed to introduce T92A mutation in wild-type D2 (wtD2) cDNA cloned into the D10 mammalian expression vector (23) The mutant construct, named here D2 Thr92Ala, was generated using overlap extension PCR. The fragments obtained with the oligonucleotides Bg84/C33 (5⬘ CATGCCATGGGCATCCTCAGCGTAGACTTGCTGA 3⬘ and 5⬘ ACTGTTGTCACCTCCTTCTGCACTGGAGACATGCAC 3⬘) and C32/C36 (5⬘ GTGCATGTCTCCAGTGCAGAAGGAGGTGACAACAGT 3⬘ and 5⬘ GTTGTAGGAGAAGGGGCCCTTTCCTCCCAG-3⬘) were used as a template for another PCR using only the outer oligonucleotides Bg84/C36, and the resulting fragment was subcloned in NCO1 and APA1 sites of wt D2. The mutations were verified by automated sequencing. Constructs were transfected into HEK-293 cells by calcium phosphate precipitation and transfection efficiencies were monitored by assay of cotransfected human GH in the medium (24). To determine the half-life of both D2 proteins, we used cycloheximide (Sigma-Aldrich, St. Louis, MO) in 100 ␮m final concentration. At the appropriate times, cells were harvested, and D2 activity


was measured in sonicated cells. Each experiment was performed with triplicate dishes for each condition.

Statistical analysis Results are expressed as mean ⫾ sd or median and range. Clinical and laboratory data from Thr/Thr, Thr/Ala, and Ala/Ala genotype individuals were compared using ANOVA or the Kruskal-Wallis H test. To examine the main effect of the D2 Thr92Ala variant, the three genotypes (Thr/Thr, Thr/Ala, and Ala/Ala) were considered separately, followed by pooling the Thr/Thr and Thr/Ala groups. For comparisons between two groups, Student’s t test or Mann-Whitney U test was applied. Linear multiple regression analysis was performed with HOMA-r (logarithmic) as dependent variable and age, gender, BMI, use of medication, and genotype as independent variables. The analyses were done using SSPS version 10.0. P ⬍ 0.05 was considered significant.

Results The D2 Thr92Ala polymorphism is associated with more severe insulin resistance in patients with DM2

The characteristics of the 183 DM2 patients analyzed in this study were the following. The mean age was 58 ⫾ 10 yr, and the mean BMI was 29 ⫾ 4.8 kg/m2. Males comprised 42% (n ⫽ 79) of the sample, and 64% (n ⫽ 117) of all patients had arterial hypertension. Twenty-four patients (13%) were treated by diet alone. Sulfonylureas alone were used by 30.6% (n ⫽ 56), metformin alone by 20.2% (n ⫽ 37), and the combination of both by 36.1% (n ⫽ 66) of the patients. Thirty-five patients (19.1%) were homozygous for the Ala allele (Ala/Ala), 82 (44.8%) were heterozygous (Ala/Thr), and 66 (36.0%) were homozygous for the Thr allele (Thr/ Thr). The Thr allele frequency was 0.59, and Ala frequency was 0.41. Genotypes were in Hardy-Weinberg equilibrium. Table 1 summarizes the clinical and laboratory data of the patients grouped according to the D2 Thr92Ala polymorphism. The mean age, diabetes duration, and the proportion of males and Caucasians were similar among the genotypes. Systolic and diastolic arterial blood pressure, waist to hip

TABLE 1. Clinical and laboratorial characteristics of DM2 patients grouped according to type 2 deiodinase Thr92Ala genotype D2 Thr92Ala polymorphism

Age (yr) Duration of diabetes (yr) Male sex [n (%)] Caucasians (%) SU/Met/SU ⫹ Met (%) Systolic BP (mm Hg) Diastolic BP (mm Hg) BMI (kg/m2) WHR Fasting plasma glucose (mg/dl) Glycosylated hemoglobin A1c (%) Total cholesterol (mg/dl) HDL cholesterol (mg/dl) Triglycerides (mg/dl)a Serum creatinine (mg/dl) Fasting insulin (mIU/ml)a HOMAa

Ala/Ala (n ⫽ 35)

Ala/Thr (n ⫽ 82)

Thr/Thr (n ⫽ 66)

58 ⫾ 9.8 10.8 ⫾ 7.8 16 (45.7) 68.6 22.9/20/48.6 139 ⫾ 19 84 ⫾ 10 28.8 ⫾ 5.6 0.96 ⫾ 0.08 176 ⫾ 74 7.13 ⫾ 2.33 207 ⫾ 56 44 ⫾ 11 158 (54 – 692) 0.87 ⫾ 0.30 19.6 (3.8 –142) 8.05 (0.88 –144)

59 ⫾ 10 11.5 ⫾ 9.3 36 (43.9) 59.8 30.5/24.4/31.7 137 ⫾ 21 86 ⫾ 16 28.2 ⫾ 4.6 0.95 ⫾ 0.08 166 ⫾ 67 6.14 ⫾ 1.95 208 ⫾ 48 45 ⫾ 13 145 (47– 491) 0.96 ⫾ 0.30 12.0 (1.0 –72.1) 4.06 (0.30 – 47.9)

57 ⫾ 12 11.6 ⫾ 8.3 27 (40.9) 60.3 34.8/15.1/34.8 139 ⫾ 20 84 ⫾ 11 29.2 ⫾ 4.8 0.95 ⫾ 0.07 171 ⫾ 53 6.49 ⫾ 2.08 207 ⫾ 43 46 ⫾ 11 123 (47–575) 0.84 ⫾ 0.31 14.8 (1.0 –143) 5.47 (0.30 –51.2)

P

0.419 0.907 0.883 0.590 0.816 0.763 0.431 0.904 0.891 0.827 0.102 0.998 0.547 0.205 0.205 0.004 0.002

SU, Sulfonylureas; Met, metformin; BP, blood pressure; WHR, waist to hip ratio; HDL, high-density lipoprotein; HOMA, homeostasis model assessment. a Median (range). Kruskal-Wallis test.


ratio, and metabolic control were also similar across the three genotypes. The median fasting plasma insulin level in Ala/Ala individuals was significantly higher than in patients with Ala/ Thr or Thr/Thr genotypes, despite similar fasting plasma glucose (Table 1). Accordingly, patients with Ala/Ala genotype had a higher HOMA index when compared with patients harboring Ala/Thr or Thr/Thr genotypes (Table 1). Of note is that, as expected, the median HOMA index was significantly elevated 5.37 (0.27–144) in this general DM2 sample population than in control subjects (20). Assuming a recessive model, patients with Ala/Thr and Thr/Thr genotypes were grouped and compared with patients with Ala/Ala genotype (Fig. 1, A and B). The HOMA index was higher in the Ala/Ala group compared with Ala/ Thr-Thr/Thr group [8.50 (0.88 –144) vs. 4.85 (0.3– 82.7); P ⫽ 0.003] (Fig. 1A). Because insulin sensitivity is known to be influenced by multiple independent factors, a multiple regression linear analysis was performed with the HOMA-r (log10HOMA) as dependent variable. Table 2 shows that the Ala/Ala genotype remains significantly associated with insulin resistance after controlling for gender, age, BMI, and use of medication. Interestingly, the association between the Ala/Ala genotype and the HOMA index was even more pronounced among patients with BMI lower than 30 kg/m2 [9.71 (0.88 –144) vs. 4.37 (0.3– 82.7); P ⫽ 0.006] (Fig. 1B). D2 velocity is decreased in tissues from patients with the Ala/Ala genotype

To determine whether homozygosity for the Ala allele affects D2 expression and/or activity, we obtained fresh human thyroid and skeletal muscle tissue samples and processed them for D2 mRNA and activity. All subjects had normal serum TSH levels at the time the tissues were collected. D2 activity was measured at 1 nm T4 in 21 thyroid samples. In subjects with the Ala/Ala genotype, D2 velocity


3475

was half of that found in the subjects with Ala/Thr or Thr/ Thr genotypes (Ala/Ala vs. Ala/Thr ⫹Thr/Thr; 0.13 ⫾ 0.06 vs. 0.24 ⫾ 0.09 fmol/min䡠mg protein; P ⫽ 0.05) (Fig. 2A). D2 activity was also measured in 31 skeletal muscle (sternocleidomastoid and rectus abdomenus) samples. The D2 maximum velocity (Vmax) in this tissue was about 1.3 fmol/mg protein䡠min. Similar to thyroid tissue, muscle samples from patients with the Ala/Ala genotype displayed significantly lower D2 velocity than patients with Ala/ThrThr/Thr genotypes (Ala/Ala vs. Ala/Thr-Thr/Thr; 0.22 ⫾ 0.12 vs. 0.35 ⫾ 0.12 fmol/min/mg protein; P ⫽ 0.04) (Fig. 2A). Further analyses of the thyroid tissue samples using the Lineweaver-Burk double-reciprocal plots indicated that the Michaelis-Menten constant (Km) (T4) values for D2 in the Ala/Ala and Thr/Thr were similar (Fig. 2B). However, the D2 Vmax values in the tissues samples with an Ala/Ala genotype were 3–10-fold lower than in samples with the Thr/ Thr genotype (Fig. 2B). Therefore, a difference in D2 amounts or velocity accounts for the lower enzyme velocity in tissues from patients with Ala/Ala genotypes. To verify that changes in the D2 mRNA concentrations in tissues with the Ala/Ala genotype would explain the lower D2 activity, we estimated D2 mRNA levels by PCR in thyroid tissue samples from subjects with the three different genotypes. A total of 26 tissue samples were available for analysis, with the following distribution among the three genotypes: AL/Ala, six; AL/Thr,12; and Thr/Thr, eight. No differences in D2 mRNA levels were found among the three genotypes (0.635 ⫾ 0.343, 0.601 ⫾ 0.398, and 0.630 ⫾ 0.255 AU, respectively; P ⫽ 0.970). D1 activity was also measured in some of the thyroid tissue samples to verify that the reduction in D2 activity in the samples with an Ala/Ala genotype was not due to admixture with cellular thyroid tissue. Figure 2B shows that there were no differences in D1 activities between the Ala/Thr-Thr/Thr

FIG. 1. A, HOMA index, calculated as described in Patients and Methods, in DM2 patients grouped according to D2 Thr92Ala genotype, Thr/Thr-Thr/Ala, or Ala/Ala. B, HOMA index according to genotype groups in patients with BMI ⬍ 30 kg/m2. The box plots represent 10th, 25th, 50th, 75th, and 90th percentiles, and P values were obtained with Mann-Whitney U test.

3476



TABLE 2. Multivariate linear regression analysis of predictors for insulin resistance [HOMA (log10) as dependent variable] in DM2 patients Variables

Standardized coefficient ␤

Ala/Ala genotype Age (yr) BMI (kg/m2) Sex Use of medication

⫺0.190 0.041 0.086 0.764 ⫺0.204

and Ala/Ala genotypes, indicating that the differences in D2 were not due to tissue sampling variations. Functional studies of Thr92Ala D2 transiently expressed in HEK-293 cells

To evaluate whether the substitution of Ala for Thr at position 92 alters the kinetic properties of the enzyme, plasmids encoding either wild-type or D2Thr92Ala constructs were transiently expressed in HEK-293 cells, and D2 velocity was determined in cell sonicates under conditions described in Patients and Methods. We did not detect significant kinetic differences between the two proteins, which display comparable Km (T4), maximal enzyme velocity, half-life, and sensitivity to substrate-induced loss of activity (Fig. 3, A–C). Discussion

The present studies show that the D2 Thr92Ala polymorphism is associated with marked insulin resistance in an unselected population of DM2 patients. In individuals with the Ala/Ala genotype, the D2 activity in thyroid, and skeletal muscle is about half of that in subjects with Ala/Thr or Thr/Thr genotypes, suggesting that decreased D2-generated

FIG. 2. A, D2 activity in human thyroid and skeletal muscle samples. P values refer to comparison of Thr/Thr (䡬)-Ala/Thr (Œ) vs. Ala/Ala (ƒ) genotypes. B, D2 kinetics from thyroid samples from patients with Ala/Ala or Thr/ Thr genotypes. The larger plot shows the median Vmax values from three unrelated patients of each genotype. Inset, Representative Lineweaver-Burk plot of T4 deiodination catalyzed by D2 from thyroid samples from patients with Ala/Ala or Thr/Thr genotype. C, D1 activity on human thyroid samples from subjects harboring the Ala/Thr-Thr/Thr or Ala/Ala genotypes.

95% Confidence interval

0.473 ⫺0.004 ⫺0.012 ⫺0.108 ⫺0.437

1.582 0.008 0.014 0.147 ⫺0.069

P

0.017 0.599 0.912 0.764 0.001

T3 production might be implicated in the genesis of insulin resistance. Insulin resistance plays a major role in the pathogenesis of type 2 diabetes. Although major genetic factors remain to be determined, common polymorphisms in the PC-1, insulin receptor substrates 1 and 2, and PPAR-␥ 2 genes have been linked to insulin resistance and dyslipidemia (25). Recently, the D2 Thr92Ala polymorphism was found to be associated with insulin resistance in nondiabetic women who underwent euglycemic-hyperinsulinemic clamps (18). Here, we demonstrate using HOMA that this association is not only present but also somewhat more pronounced in nonobese DM2 patients. Patients homozygous for the Ala allele were found to have higher fasting plasma insulin levels and HOMA index, and this association was even more pronounced in patients with BMI lower than 30 kg/m2, a generally accepted cutoff for obesity. These findings are particularly relevant because the D2 ThrD2Ala variant might be a novel marker for insulin resistance that is apparently not related to other well-known risk factors such as obesity. Although HOMA index is only an estimate of insulin resistance, it is simple to perform and shows a good correlation


FIG. 3. Biochemical properties of wild-type (Wt) and mutant (Thr92Ala) D2 enzymes. A, Lineweaver-Burk plot of T4 deiodination catalyzed by WtD2 or Thr92 Ala. B, Disappearance of D2 activity in HEK-293 cells transiently expressing WtD2 or Thr92Ala plasmids in the presence of 100 ␮M cycloheximide. Activity is expressed relative to the mean of the vehicle (DMSO) ⫾ SD in three separate experiments, each performed in triplicate. C, Effects of T4-induced changes in activity on WtD2 and D2 Thr92Ala transiently expressed in HEK293 cells. Cells were incubated in the absence (⫺) or presence (⫹) of 30 nM T4. D2 activity was measured after 2 h of treatment. Bars, Mean ⫾ SE of three different experiments, each performed in triplicate.

with the reference method, the euglycemic clamp (19, 26) and, therefore, is considered a good approach for cohort and epidemiological studies (27). A major novel observation in these studies is the positive association between the Ala/Ala genotype and lower D2 activity in tissue samples. This provides a potential mechanistic explanation for the association between insulin resis-


3477

tance and the D2 Thr92Ala polymorphism. In humans, skeletal muscle is the primary site of insulin-dependent glucose disposal (28). A lower D2 activity would decrease D2-generated T3 in skeletal muscle and could create a state of relative intracellular hypothyroidism, decreasing the expression of genes involved in energy use, such as GLUT4, leading to insulin resistance. The mechanism of reduced D2 activity is not clear. In agreement with previous studies (16), we did not detect any significant changes in the biochemical properties of the mutant enzyme (Fig. 3A), suggesting that although this variant could be a marker for abnormal Dio2 expression, the mutation per se does not explain the reduced D2 velocity. Other D2 properties, such as half-life and sensitivity to substrate exposure, were also unaffected (Fig. 3, B and C). The singlepoint mutation (A/G) in the Dio2 gene is a nonconservative variant that predicts a Thr to Ala substitution at codon 92, which is not near the catalytic active site of the enzyme (5). In addition, this region is not phylogenetically conserved. The homologous amino acid is a proline in rodents and a glycine in chicken D2 (29). In contrast, humans and amphibians share a Thr in this position. Thus, it seems likely that the lower D2 activity is the result of linkage disequilibrium between the Thr92Ala polymorphism and a functional polymorphism elsewhere in this allele of the Dio2 gene or in a gene nearby. This is supported by a recent report of a positive association of common single-nucleotide polymorphisms in the Dio2 gene, the D2 Thr92Ala polymorphism, and two others located in the noncoding region of the gene, with mental retardation in iodine-deficient areas of China (30). However, it is intriguing that no differences were observed in D2 mRNA levels among the genotypes. Although the method used here, RT-PCR, is semiquantitative, it provides a reasonable approximation to compare amounts of a given mRNA in different samples. If this is true, an alternative possible explanation is that the Thr92Ala mutation causes a defect in translation or protein stability, which was detected in tissues, but not in the in vitro expression system. Some factors unrelated to the deiodinase polymorphism could have interfered with the findings of the present investigation. For example, medications for DM2 could have played a role because some are known to affect with insulin sensitivity. However, we minimized such a possibility by excluding insulin-treated patients, and the proportion of patients on metformin and/or sulfonylureas was similar among the genotype groups. Furthermore, the D2 genotype remains significantly associated with HOMA index in a linear regression model analysis that included use of medication as a variable. Regarding the D2 activity in human tissues, even though the number of samples analyzed was small, these subjects were not receiving any medication known to affect insulin secretion and/or sensitivity, or deiodinase function. In conclusion, the Dio2 single-nucleotide polymorphism A/G is associated with more severe insulin resistance in DM2 patients. This polymorphism is associated with lower D2 velocity in the skeletal muscle and thyroid sample tissues. These findings indicate that the intracellular D2-generated T3 in skeletal muscle may play a role in determining insulin resistance. Thus, skeletal muscle D2 may be a potential

3478


pharmacological target for treating patients with insulin resistance. Acknowledgments We thank Drs. Alceu Migliavacca and Jose´ Ricardo Guimara˜es for the thyroid and muscle sample biopsies. Received October 11, 2004. Accepted March 21, 2005. Address all correspondence and requests for reprints to: Ana Luiza Maia, M.D., Ph.D., Serviço de Endocrinologia, Hospital de Clıńicas de Porto Alegre, Rua Ramiro Barcelos 2350, 90035-003 Porto Alegre, RS, Brazil. E-mail: [email protected]. This work was supported by the Coordenaçaõ de Aperfeiçoamento de Pessoal de Nı´vel Superior, by Projeto de Nućleos de Exceleˆncia do Ministe´rio de Cieˆncia e Tecnologia, by Conselho Nacional de Desenvolvimento Cientı´fico e Tecnolo´gico, Brazil, and by National Institutes of Health Grants DK36256 and DK65055.


13. 14.

15. 16.

17.

18.

References 1. 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 2. Berry MJ, Maia AL, Kieffer JD, Harney JW, Larsen PR 1992 Substitution of cysteine for selenocysteine in type I iodothyronine deiodinase reduces the catalytic efficiency of the protein but enhances its translation. Endocrinology 131:1848 –1852 3. Croteau W, Whittemore SL, Schneider MJ, St. Germain DL 1995 Cloning and expression of a cDNA for a mammalian type III iodothyronine deiodinase. J Biol Chem 270:16569 –16575 4. 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 functional analysis of the placental selenoenzyme. J Clin Invest 96:2421–2430 5. Callebaut I, Curcio-Morelli C, Mornon JP, Gereben B, Buettner C, Huang S, Castro B, Fonseca TL, Harney JW, Larsen PR, Bianco AC 2003 The iodothyronine selenodeiodinases are thioredoxin-fold family proteins containing a glycoside hydrolase-clan GH-A-like structure. J Biol Chem 278:36887–36896 6. 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 7. 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 8. Ng L, Goodyear RJ, Woods CA, Schneider MJ, Diamond E, Richardson GP, Kelley MW, Germain DL, Galton VA, Forrest D 2004 Hearing loss and retarded cochlear development in mice lacking type 2 iodothyronine deiodinase. Proc Natl Acad Sci USA 101:3474 –3479 9. Salvatore D, Bartha T, Harney JW, Larsen PR 1996 Molecular biological and biochemical characterization of the human type 2 selenodeiodinase. Endocrinology 137:3308 –3315 10. 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 11. Scheidegger K, O’Connell M, Robbins DC, Danforth Jr E 1984 Effects of chronic ␤-receptor stimulation on sympathetic nervous system activity, energy expenditure, and thyroid hormones. J Clin Endocrinol Metab 58:895–903 12. al-Adsani H, Hoffer LJ, Silva JE 1997 Resting energy expenditure is sensitive

19.

20. 21. 22. 23. 24.

25. 26.

27. 28.

29. 30.

to small dose changes in patients on chronic thyroid hormone replacement. J Clin Endocrinol Metab 82:1118 –1125 Ree RN, Kourides AA, Ridgway EC, Weintraub BD, Maloof F 1976 The effect of glucocorticoid administration of human pituitary secretion of thyrotropin and prolactin. J Clin Endocrinol Metab 43:338 Bartha T, Kim SW, Salvatore D, Gereben B, Tu HM, Harney JW, Rudas P, Larsen PR 2000 Characterization of the 5⬘-flanking and 5⬘-untranslated regions of the cyclic adenosine 3⬘,5⬘-monophosphate-responsive human type 2 iodothyronine deiodinase gene. Endocrinology 141:229 –237 Ezaki O 1997 Regulatory elements in the insulin-responsive glucose transporter (GLUT4) gene. Biochem Biophys Res Commun 241:1– 6 Ikemoto S, Thompson KS, Takahashi M, Itakura H, Lane MD, Ezaki O 1995 High fat diet-induced hyperglycemia: prevention by low level expression of a glucose transporter (GLUT4) minigene in transgenic mice. Proc Natl Acad Sci USA 92:3096 –3099 Peeters RP, van Toor H, Klootwijk W, de Rijke YB, Kuiper GG, Uitterlinden AG, Visser TJ 2003 Polymorphisms in thyroid hormone pathway genes are associated with plasma TSH and iodothyronine levels in healthy subjects. J Clin Endocrinol Metab 88:2880 –2888 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 ␤-3-adrenergic receptor. Diabetes 51:880 – 883 Bonora E, Targher G, Alberiche M, Bonadonna RC, Saggiani F, Zenere MB, Monauni T, Muggeo M 2000 Homeostasis model assessment closely mirrors the glucose clamp technique in the assessment of insulin sensitivity: studies in subjects with various degrees of glucose tolerance and insulin sensitivity. Diabetes Care 23:57– 63 Seligman BG, Biolo A, Polanczyk CA, Gross JL, Clausell N 2000 Increased plasma levels of endothelin 1 and von Willebrand factor in patients with type 2 diabetes and dyslipidemia. Diabetes Care 23:1395–1400 Wagner MS, Morimoto R, Dora JM, Benneman A, Pavan R, Maia AL 2003 Hypothyroidism induces type 2 iodothyronine deiodinase expression in mouse heart and testis. J Mol Endocrinol 31:541–550 Mandel SJ, Berry MJ, Kieffer JD, Harney JW, Warne RL, Larsen PR 1992 Cloning and in vitro expression of the human selenoprotein, type I iodothyronine deiodinase. J Clin Endocrinol Metab 75:1133–1139 Gossen M, Bujard H 1992 Tight control of gene expression in mammalian cells by tetracycline-responsive promoters. Proc Natl Acad Sci USA 89:5547–5551 Brent GA, Larsen PR, Harney JW, Koenig RJ, Moore DD 1989 Functional characterization of the rat growth hormone promoter elements required for induction by thyroid hormone with and without a co-transfected ␤ type thyroid hormone receptor. J Biol Chem 264:178 –182 Laakso M 2004 Gene variants, insulin resistance, and dyslipidaemia. Curr Opin Lipidol 15:115–120 Emoto M, Nishizawa Y, Maekawa K, Hiura Y, Kanda H, Kawagishi T, Shoji T, Okuno Y, Morii H 1999 Homeostasis model assessment as a clinical index of insulin resistance in type 2 diabetic patients treated with sulfonylureas. Diabetes Care 22:818 – 822 Wallace TM, Levy JC, Matthews DR 2004 Use and abuse of HOMA modeling. Diabetes Care 27:1487–1495 Ferrannini E, Galvan AQ, Gastaldelli A, Camastra S, Sironi AM, Toschi E, Baldi S, Frascerra S, Monzani F, Antonelli A, Nannipieri M, Mari A, Seghieri G, Natali A 1999 Insulin: new roles for an ancient hormone. Eur J Clin Invest 29:842– 852 Gereben B, Bartha T, Tu HM, Harney JW, Rudas P, Larsen PR 1999 Cloning and expression of the chicken type 2 iodothyronine 5⬘-deiodinase. J Biol Chem 274:13768 –13776 Guo TW, Zhang FC, Yang MS, Gao XC, Bian L, Duan SW, Zheng ZJ, Gao JJ, Wang H, Li RL, Feng GY, St. Clair D, He L 2004 Positive association of the DIO2 (deiodinase type 2) gene with mental retardation in the iodine-deficient areas of China. J Med Genet 41:585–590

JCEM is published monthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society serving the endocrine community.