Changes in Plasma Low-Density Lipoprotein (LDL)-and High-Density ...

5 downloads 0 Views 267KB Size Report
Thyroid function disorders lead to changes in lipoprotein metab- .... Changes in thyroid function tests, plasma lipids and lipoproteins, and body mass index in 47 ...
0021-972X/00/$03.00/0 The Journal of Clinical Endocrinology & Metabolism Copyright © 2000 by The Endocrine Society

Vol. 85, No. 5 Printed in U.S.A.

Changes in Plasma Low-Density Lipoprotein (LDL)- and High-Density Lipoprotein Cholesterol in Hypo- and Hyperthyroid Patients Are Related to Changes in Free Thyroxine, Not to Polymorphisms in LDL Receptor or Cholesterol Ester Transfer Protein Genes M. J. M. DIEKMAN, N. ANGHELESCU, E. ENDERT, O. BAKKER, W. M. WIERSINGA

AND

Department of Endocrinology and Metabolism, Academic Medical Center, University of Amsterdam, 1105 AZ Amsterdam, The Netherlands ABSTRACT Thyroid function disorders lead to changes in lipoprotein metabolism. Both plasma low-density lipoprotein cholesterol (LDL-C) and high-density lipoprotein cholesterol (HDL-C) increase in hypothyroidism and decrease in hyperthyroidism. Changes in LDL-C relate to altered clearance of LDL particles caused by changes in expression of LDL receptors on liver cell surfaces. Changes in cholesterol ester transfer activity partly explain changes in HDL-C. It has been suggested that the magnitude of these changes is related to polymorphisms of involved genes. The aim of the present study is to investigate whether the polymorphic AvaII restriction site in exon 13 of the LDL receptor gene and the polymorphic TaqIB site in intron 1 of the cholesterol ester transfer protein are associated with the magnitude of the changes in plasma LDL-C and HDL-C, respectively, in the transition from the hypo- or hyperthyroid to the euthyroid state. From a consecutive group of 66 untreated hypothyroid and 60 hyperthyroid patients, 47 Caucasians in each group were analyzed. Fasting LDL-C and HDL-C were measured at baseline and 3 months after restoration of the euthyroid state. Genotype was determined by means of PCR techniques. The homozygous presence of a restriction site was designated as ⫹/⫹, heterozygous as ⫹/⫺, and absence as ⫺/⫺. Trend analysis was done with ANOVA.

I

T IS WELL KNOWN that thyroid dysfunction leads to changes in lipoprotein metabolism. Plasma low-density lipoprotein cholesterol (LDL-C) and high-density lipoprotein cholesterol (HDL-C) levels increase in hypothyroidism and decrease in hyperthyroidism (1–3). Furthermore, clearance of chylomicron remnants is decreased in hypothyroidism (4). Changes in LDL-C are mainly attributable to altered clearance of LDL-C from plasma by changes in the number of LDL receptors on liver cell surfaces (5, 6). Because the promoter of the LDL receptor gene contains a thyroid hormone responsive element (TRE), T3 could modulate gene expression of the LDL receptor (7). HDL-C metabolism is complex, and changes in plasma levels are due, in part, to remodeling of HDL-C particles by hepatic lipase and choReceived August 5, 1999. Revision received February 1, 2000. Accepted February 7, 2000. Address correspondence and requests for reprints to: M. J. M. Diekman, Academic Medical Center, Department of Endocrinology and Metabolism F5-174, Meibergdreef 9, 1105 AZ Amsterdam Zuidoost, The Netherlands. E-mail: [email protected].

Among hypo- or hyperthyroid patients, subgroups with different genotypes did not differ in thyroid function pre- or post treatment. The mean decrease in LDL-C (mmol/L ⫾ SD) in hypothyroid patients with different AvaII genotypes did not differ: ⫺1.07 ⫾ 1.44 (⫺/⫺, N ⫽ 15), ⫺1.25 ⫾ 1.53 (⫹/⫺, N ⫽ 19), and ⫺1.18 ⫾ 1.01 (⫹/⫹, N ⫽ 13) mmol/L [not significant (NS)]; neither did the mean increase in hyperthyroid patients: 1.07 ⫾ 0.90 (⫺/⫺, N ⫽ 18), 0.92 ⫾ 1.00 (⫹/⫺, N ⫽ 21), and 1.20 ⫾ 0.45 (⫹/⫹ N, ⫽ 6) (NS). The mean decrease in HDL-C (mmol/ L ⫾ SD) in hypothyroid patients with different TaqIB genotypes did not differ: ⫺0.22 ⫾ 0.26 (⫺/⫺, N ⫽ 13), ⫺0.15 ⫾ 0.23 (⫹/⫺, N ⫽ 21), and ⫺0.12 ⫾ 0.22 (⫹/⫹, N ⫽ 9) (NS); neither did the mean increase in hyperthyroid patients: 0.29 ⫾ 0.39 (⫺/⫺, N ⫽ 7), 0.26 ⫾ 0.23 (⫹/⫺, N ⫽ 22), and 0.19 ⫾ 0.31 (⫹/⫹, N ⫽ 18) (NS). Changes in LDL-C and HDL-C correlated with the logarithm of the change in free T4 (fT4), expressed as the fT4 posttreatment/fT4 pretreatment ratio (r ⫽ ⫺0.81, P ⬍ 0.001; and r ⫽ ⫺0.62, P ⬍ 0.001, respectively). In conclusion, in the transition from hypo- or hyperthyroidism to euthyroidism, no association is found between AvaII genotype and changes in plasma LDL-C nor between TaqIB genotype and changes in HDL-C. Changes in LDL-C and HDL-C correlate with changes in fT4. (J Clin Endocrinol Metab 85: 1857–1862, 2000)

lesterol ester transfer protein (CETP) (8). Activity of both enzymes decreases in hypothyroidism and increases in hyperthyroidism, correlating with plasma HDL-C (9 –12). The magnitude of changes in plasma LDL-C and HDL-C levels, after restoration of the euthyroid state, varies from patient to patient (1). The extent of these changes depends both on the severity and duration of the thyroid dysfunction and on the degree of pretreatment hypercholesterolemia (13– 15). Diet, body weight, and smoking habits can also modify absolute LDL-C levels (16, 17). In addition to these endocrine and environmental factors, genetic constitution can explain some of the interindividual variation. An association between a polymorphic AvaII site in exon 13 of the LDL-R gene and the extent of cholesterol lowering upon restoration of the euthyroid state in hypothyroid patients has been reported (18). Absence of this site was associated with the most marked hypocholesterolemic response. A polymorphic site explaining variations in plasma HDL-C during thyroid dysfunction is not known. A candidate could be the TaqIB site in intron 1 of the CETP gene, because presence of this site is

1857

1858

JCE & M • 2000 Vol 85 • No 5

DIEKMAN ET AL.

TABLE 1. Changes in thyroid function tests, plasma lipids and lipoproteins, and body mass index in 47 hypothyroid and in 47 hyperthyroid patients, upon restoration of euthyroidism Hypothyroid (n ⫽ 47)

TSH fT4 T3 Total cholesterol LDL-C HDL-C Total chol/HDL-C Apoprotein A Apoprotein B Triglycerides BMI

Hyperthyroid (n ⫽ 47)

Untreated

Euthyroid

Untreated

Euthyroid

66.0 (4.5–162) 6.0 ⫾ 3.7 1.2 ⫾ 0.5 6.79 ⫾ 2.22 4.59 ⫾ 2.02 1.60 ⫾ 0.44 4.4 ⫾ 1.7 1.59 ⫾ 0.26 1.35 ⫾ 0.51 1.25 ⫾ 0.94 25.5 ⫾ 4.3

2.4 (0.18 – 8.5)a 16.3 ⫾ 3.7a 1.8 ⫾ 0.4a 5.43 ⫾ 1.20a 3.47 ⫾ 1.11a 1.43 ⫾ 0.36a 4.0 ⫾ 1.3a 1.53 ⫾ 0.24c 1.07 ⫾ 0.30a 1.14 ⫾ 0.66 25.1 ⫾ 4.2b

⬍0.01 (⬍0.01– 0.04) 42.8 ⫾ 16.7 5.2 ⫾ 2.3 4.49 ⫾ 1.08 2.69 ⫾ 3.70 1.25 ⫾ 0.33 3.7 ⫾ 1.0 1.39 ⫾ 0.31 0.82 ⫾ 0.27 1.11 ⫾ 0.41 22.5 ⫾ 3.9

1.0 (0.01– 8.3)a 14.5 ⫾ 3.1a 2.1 ⫾ 0.7a 5.81 ⫾ 1.21a 3.70 ⫾ 1.10a 1.48 ⫾ 0.42a 4.2 ⫾ 1.6b 1.57 ⫾ 0.33a 1.13 ⫾ 0.30a 1.32 ⫾ 0.61d 24.3 ⫾ 3.6a

Values as mean ⫾ SD, except TSH given as median (range); a P ⬍ 0.005; b P ⬍ 0.01; c P ⫽ 0.05; d P ⫽ 0.02 vs. the untreated stage. TSH (mU/L), fT4 (pmol/L), T3 (nmol/L), total cholesterol (mmol/L), LDL-C (mmol/L), HDL-C (mmol/L), apoprotein A (g/L), apoprotein B (g/L), triglycerides (mmol/L), BMI (kg/m2).

FIG. 1. RFLP analysis. Left panel, The PCR product of exon 13 of the LDL receptor gene (480 bp) contains always one AvaII site resulting in fragments of 150 and 330 bp. When an extra (polymorphic) AvaII site is present, the 330-bp fragment yields fragments of 195 and 135 bp. m, Marker lane; arrow, 250 bp. Right panel: The PCR product of intron 1 of CETP gene (1420 bp) has one TaqIB polymorphic site yielding bands of 750 and 670 bp when a ⫹ allele is present. The bands are not fully separated on this example gel (arrow); however, this does not interfere with the ability to determine the specific genotype.

associated with increased CETP concentrations and reduced HDL-C concentrations in healthy males (19). The aim of the present study is to reexamine the association between the AvaII site and changes in LDL-C in hypothyroidism on restoration of the euthyroid state and to evaluate whether this association is also observed in hyperthyroid patients. At the same time, we studied the relationship between the TaqIB site and the changes in plasma HDL-C level in these patients. Subjects and Methods Patients Consecutive patients with primary hypothyroidism (n ⫽ 66) or with primary hyperthyroidism (n ⫽ 60), referred to our out-patient clinic, were studied. To have a genetically more homogeneous group, we included only Caucasians in the final analysis. In the hypothyroid group, 7 patients were non-Caucasians, 6 patients were lost to follow-up, 6 patients had no AvaII genotyping, and 10 had no TaqIB genotyping; consequently, 47 patients [median age, 45 yr (range, 23–75); 13 males] were available for analysis of AvaII; and 43 patients [median age, 45 yr (range, 23–75); 11 males], for analysis of TaqIB polymorphism. The cause of hypothyroidism was chronic autoimmune

thyroiditis (n ⫽ 31), 131I treatment (n ⫽ 12), thyroidectomy (n ⫽ 2), prolonged overdose of thiamazol (n ⫽ 1), and subacute thyroiditis (n ⫽ 1). Subclinical hypothyroidism was present in 7; and overt hypothyroidism, in 40 patients. In the hyperthyroid group, 12 patients were non-Caucasians, 1 patient was lost to follow-up, and 2 patients had no AvaII genotyping, leaving 45 patients [median age, 47 yr (range, 20 –77); 7 male]) for AvaII and 47 [median age, 47 (range, 20 –77); 9 males] for TaqIB polymorphism analysis. Causes of hyperthyroidism were Graves’ disease (n ⫽ 32), toxic multinodular goiter (n ⫽ 14), and toxic adenoma (n ⫽ 1). Subclinical hyperthyroidism was present in 6; and overt hyperthyroidism, in 41 patients. None of the patients was on a special diet or used any medication known to interfere with lipoprotein metabolism; women using oral anticonceptives (4 in the hypothyroid group and 7 in the hyperthyroid group) continued to do so until the end of the study. Smoking habits were noted, as well as length and body weight. All patients were studied twice: once in the untreated state, and again at least 3 months after achieving the euthyroid state. Treatment was with levothyroxine sodium in the case of hypothyroidism or with antithyroid drugs or 131I in the case of thyrotoxicosis. Blood samples were collected, after an overnight fast, by venipuncture into evacuated tubes containing either EDTA (1 g/L) as an anticoagulant for measurement of lipid profiles, or sodium heparinate as an anticoagulant for thyroid hormone measurements. The study was approved by the institutional Medical Ethical Committee, and patients gave their written informed consent.

Methods Plasma lipids and thyroid function. T4 and total T3 were measured by in-house RIA methods (20). FT4 was measured by a two-step fluoroimmuno assay (DELFIA, Wallac, Inc., Turku, Finland); TSH was measured by immunofluorometric assay (DELFIA, Wallac, Inc.). Hypothyroidism was defined as an increased plasma TSH. Hyperthyroidism was defined as a decreased plasma TSH (reference range, 0.4 – 4.0 mU/L) in combination with an increased plasma free T4 (fT4; reference range, 10 –23 pmol/L) or total T3 (reference range, 1.3–2.7 nmol/L). The euthyroid state for previous hypothyroid patients was defined as a normal TSH in combination with a normal plasma fT4 and total T3; for previous thyrotoxic patients, as a normal plasma fT4 and total T3 in the absence of an increased TSH. Total cholesterol in plasma was measured with an enzymatic method (CHOD-PAP, catalog no. 1442350; Roche Diagnostics B.V., Almere, The Netherlands) on a Cobas Bio centrifugal analyzer (Roche Diagnostics B.V.); HDL-C (after precipitation of very low-density lipoprotein cholesterol and LDL-C with heparin-Mn2⫹), by the enzymatic CHOD-PAP method. LDL-C was calculated with the Friedewald formula; triglycerides were measured by an enzymatic method (GPOPAP, catalog no 701912, Boehringer); apolipoprotein A-1 and B were assayed with an immunonephelometric method on a Behring nephelometric analyzer (Behring Diagnostics, Rijswijk, The Netherlands) according to the protocol and with reagents of the manufacturer.

CHOLESTEROL CHANGES IN THYROID PATIENTS

1859

TABLE 2. Demographic characteristics of hypothyroid and hyperthyroid patients, according to AvaII genotype of the LDL receptor and to Taq1B genotype of CETP Hypothyroid

Genotype AvaII LDLr N (males) Age (yr) N smokers Taq1B CETP N (males) Age (yr) N smokers

Hyperthyroid

⫺/⫺

⫹/⫺

⫹/⫹

⫺/⫺

⫹/⫺

⫹/⫹

15 (5) 47 (23– 69) 6 (40%)

19 (4) 42 (28 – 69) 7 (37%)

13 (4) 50 (28 –75) 5 (38%)

18 (5) 47 (25–71) 8 (44%)

21 (2) 48 (24 –76) 10 (48%)

6 (0) 43 (20 –77) 2 (33%)

13 (4) 42 (28 – 65) 8 (65%)

21 (5) 47 (23–75) 8 (38%)

9 (2) 40 (31–58) 3 (33%)

7 (1) 46 (20 – 63) 3 (43%)

22 (6) 38 (23–76) 9 (41%)

18 (2) 49 (25–77) 8 (44%)

FIG. 2. Bar diagrams for changes (⌬) in fT4, LDL-C, apolipoprotein (apo) B, and BMI according to previous hypo- or hyperthyroid state and AvaII genotype of the LDL receptor. Changes in TSH are given as Box and Whisker plots. Genetic analysis of polymorphisms. Genomic DNA was extracted from peripheral leucoctyes according to standard procedures (21). PCR was performed with primer sets and under conditions as reported (22, 23). The PCR products were digested with restriction endonuclease AvaII (Anabaena variabilis) or TaqIB (Thermophilus aquatus) (Boehringer). The digests were electroforesed on 1% agarose gels, stained with ethidium bromide, and visualized with ultraviolet detection (Eagle Eye II, Stratagene, La Jolla, CA). Absence of the restriction site was noted as (⫺) and presence as (⫹).

Statistical analysis. Data were analyzed using the statistical package SPSS, Inc. version 6.0 (Chicago, IL). TSH values are given as median and range because of the skewed distribution of the data and were analyzed by the Kruskall Wallis test. Paired data on lipoproteins in the transition from the hypo- or hyperthyroid state to the euthyroid state were compared by Student’s t test. Data on thyroid function tests and lipoproteins between groups with different genotypes were compared by means of ANOVA. All patients, whether hypo- or hyperthyroid, were analyzed together, along a continuum of ⌬ free T4 (fT4) (expressed as the fT4

1860

DIEKMAN ET AL.

JCE & M • 2000 Vol 85 • No 5

FIG. 3. Bar diagrams for changes (⌬) in fT4, HDL-C, apo A, triglycerides, and BMI, according to previous hypo- or hyperthyroid state and TaqIB genotype of CETP. Changes in TSH are given as Box and Whisker plots. posttreatment/fT4 pretreatment ratio). Multivariate linear regression analysis of ⌬ LDL-C on log (⌬ fT4) was performed with different intercepts and slopes for each AvaII genotype. Multivariate ANOVA (F-test) was used to test whether the three separate lines coincided. The same type of analysis was used to study the relationship between ⌬ HDL-C, log (⌬ fT4), and TaqIB genotype. A two-tailed probability value less than 0.05 was considered to be a significant difference for the F-test (major endpoint). A two-tailed probability value less than 0.01 was considered to be a significant difference for all other comparisons.

Results

Changes in thyroid function tests, lipids, lipoproteins, and body mass index (BMI) for the hypo- and hyperthyroid patient groups are given in Table 1. According to the direction of thyroid dysfunction, the typical changes in plasma LDL-C and HDL-C levels were observed. Wide variations were seen in the individual lipid responses, upon restoration of the euthyroid state: changes in LDL-C ranged from almost 0 to ⫺5 mmol/L and from ⫺0.8 to 2.8 mmol/L in hypothyroid and hyperthyroid patients, respectively. Changes in HDL-C varied from ⫹0.3 to [mimes]0.75 mmol/L and from [mimes]0.25 to ⫹ 1 mmol/L in hypothyroid and hyperthyroid patients, respectively. The frequency of the presence of an AvaII site was 43% in a Caucasian American population (22, 24), and the frequency

of the presence of a TaqIB site was 59% in a healthy male population (23). The observed frequencies of different alleles for the AvaII and TaqIB restriction sites in our total patient group (hypo- and hyperthyroid patients) were 42% and 54%, respectively. Typical examples of the AvaII and TaqIB polymorphisms, as assessed by inspection of the gels, are given in Fig. 1. When hypothyroid and hyperthyroid patient groups were subdivided according to their AvaII or TaqIB genotypes, no differences in sex, age, or smoking habits were found among the different subgroups (Table 2). Changes in thyroid function tests and BMI were not different either (Figs. 2 and 3). The magnitude of the changes in LDL-C and apolipoprotein B did not differ among the AvaII genotypes. Neither were there differences between the degree of changes in HDL-C, apolipoprotein A, or triglycerides among the TaqIB genotypes. When all patients, whether hypo- or hyperthyroid, were analyzed together, along a continuum of ⌬ f T4 (expressed as the f T4 posttreatment/f T4 pretreatment ratio), neither was AvaII genotype associated with the size of the difference in LDL-C (P ⫽ 0.24) level nor was TaqIB genotype associated with the size in difference in HDL-C levels (P ⫽ 0.54) (see Figs. 4 and 5); log (⌬ f T4) correlated both with ⌬ LDL-C (⌬

CHOLESTEROL CHANGES IN THYROID PATIENTS

1861

LDL-C ⫽ 0.05–2.20 log (⌬ f T4), r ⫽ ⫺0.81, P ⬍ 0.001) and with ⌬ HDL-C (⌬ HDL-C ⫽ 0.05– 0.36 log (⌬ f T4), r ⫽ ⫺0.62, P ⬍ 0.001) Discussion

The variable extent of the changes in LDL-C and HDL-C, upon restoration of the euthyroid state in hypothyroid patients, as observed in clinical practice, existed also in our patients. Changes in LDL-C were, however, not associated with the AvaII polymorphism. This is in contrast with the results of Wisemann et al (18). Compared with their study, our hypothyroid patients had a similar age and sex distribution, and the degree of hypothyroidism and changes in lipoproteins were equal. The frequency of the presence of an AvaII site in the hypothyroid patients was similar. To exclude an influence of different genetic background attributable to different ethnicity, we included only Caucasians in the final analysis. The ethnic origin of the Wisemann study group is not mentioned; a different racial background of their patients might explain the discrepancy with our study results. We extended our observations to hyperthyroid patients but could not find a relationship there either. This supports our conclusion that there is a lack of association between AvaII polymorphism and the magnitude of changes in LDL-C at different plasma thyroid hormone concentrations. The hypothesis put forward by Wisemann et al., that the polymorphic AvaII restriction site indicates a TRE in its surroundings, is original but highly speculative. The putative presence of such a hormone response element far downstream from the promoter site of the gene would be exceptional. Gene regulation outside the promoter site is a rather infrequent finding in genes. Moreover, recent studies have demonstrated a TRE in the promoter of the LDL receptor

FIG. 4. Relationship between the change (⌬) in plasma f T4 (expressed as the f T4 posttreatment/f T4 pretreatment ratio on a logarithmic scale) and the change in plasma LDL-C in 47 hypothyroid (open symbols) and 47 hyperthyroid (solid symbols) patients, upon reaching euthyroidism, according to their AvaII genotypes of the LDL receptor gene. Regression line for the total group: ⌬ LDL-C ⫽ 0.05–2.20 log (⌬ f T4); r ⫽ ⫺0.81, P ⬍ 0.001. The f T4 posttreatment/f T4 pretreatment ratio is less than 1 in the hyperthyroid and greater than 1 in the hypothyroid group.

FIG. 5. Relationship between the change (⌬) in plasma f T4 (expressed as the f T4 posttreatment/f T4 pretreatment ratio on a logarithmic scale) and the change in plasma HDL-C in 43 hypothyroid (open symbols) and 47 hyperthyroid (solid symbols) patients, upon reaching euthyroidism, according to their TaqIB genotypes of the CETP gene. Regression line for the total group: ⌬ HDL-C ⫽ 0.05– 0.36 log (⌬ f T4); r ⫽ ⫺0.62, P ⬍ 0.001. The f T4 posttreatment/f T4 pretreatment ratio is less than 1 in the hyperthyroid and greater than 1 in the hypothyroid group.

gene (7). What then might explain the clear differences in response of LDL-C to treatment between different patients? Changes in diet or exercise are unlikely explanations, because this would imply a change in life style, which probably does not occur after diagnosing and treating a benign thyroid disease. Are there other candidate genes that might influence the relationship between f T4 and LDL-C? Variation in the gene coding for liver deiodinase type 1 might lead to different intracellular liver concentrations of T3 with equal plasma f T4 levels. Genetic variation at the TRE in the promote site of the LDL receptor is another possibility. Changes in HDL-C were also independent of TaqIB polymorphism. This particular polymorphic site was examined because it seems to be biologically relevant. In male patients with coronary heart disease, a relationship between the presence of this polymorphic site and the decrease in vascular luminal diameter has been found, indicating faster progression of atherosclerosis, compared with individuals in whom this site is absent. This relationship disappeared after treatment with HMGCoA reductase inhibitors (25). Such a biological significance is not known for the above mentioned AvaII site. In conclusion, polymorphisms for AvaII and TaqIB in the genes for LDL receptor and cholesterol ester transfer protein do not seem to influence the changes in plasma LDLs and HDLs occurring after treatment of hypo- and hyperthyroidism. The main determinant of changes in LDL-C and HDL-C, upon restoration of the euthyroid state, are the changes in plasma f T4. Acknowledgments We thank Dr. A. Soutar for helpful comments.

1862

DIEKMAN ET AL. References

1. Heimberg M, Olubadewo JO, Wilcox HG. 1985 Plasma lipoproteins and regulation of hepatic metabolism of fatty acids in altered thyroid states. Endocr Rev. 6:590 – 607. 2. Muls E, Blaton V, Rossenue M, Lesaffre E, Lamberigts G, De Moor P. 1982 Serum lipids and apolipoproteins A-I, A-II and B in hyperthyroidism before and after treatment. J Clin Endocrinol Metab. 55:459 – 464. 3. Muls E, Rossenue M, Blaton V, Lesaffre E, Lamberigts G, De Moor P. 1984 Serum lipids and apolipoproteins A-I, A-II and B in primary hypothyroidism before and during treatment. Eur J Clin Invest. 14:12–15. 4. Weintraub M, Grosskopf I, Trostanesky Y, Charach G, Rubinstein A, Stern N. 1999 Thyroxine replacement therapy enhances clearance of chylomicron remnants in patients with hypothyroidism. J Clin Endocrinol Metab. 84:2532– 2536. 5. Chait A, Bierman EL, Albers J. 1979 Regulatory role of T3 in the degradation of LDL by cultured human skin fibroblast. J Clin Endocrinol Metab. 48:887– 889. 6. Soutar AK, Knight BL. 1990 Structure and regulation of the LDL receptor and its gene. Br Med Bull. 46:891–916. 7. Bakker O, Hudig F, Meijssen S, Wiersinga WM. 1998 Effects of triiodothyronine and amiodarone on the promoter of the human LDL receptor gene. Biochem Biophys Res Commun. 249:517–521. 8. Tall AR. 1993 Plasma cholesterol transfer protein. J Lipid Res. 34:1255–1274. 9. Dullaart RPF, Hoogenberg K, Groener JEM, Dikkeschei LD, Erkelens DW, Doorenbos H. 1990 The activity of cholesteryl ester transfer protein is decreased in hypothyroidism: a possible mechanism for an altered composition of high density lipoproteins. Eur J Clin Invest. 20:581–587. 10. Ritter MC, Kannan CR, Bagdade JD. 1996 The effects of hypothyroidism and replacement therapy on cholesteryl ester transfer. J Clin Endocrinol Metab. 81:797– 800. 11. Tan KC, Shiu SW, Kung AW. 1998 Effect of thyroid dysfunction on highdensity lipoprotein subfraction metabolism: roles of hepatic lipase and cholesteryl ester transfer protein. J Clin Endocrinol Metab. 83:2921–2924. 12. Tan KC, Shiu SW, Kung AW. 1998 Plasma cholesteryl ester transfer protein activity in hyper- and hypothyroidism. J Clin Endocrinol Metab. 83:140 –143.

JCE & M • 2000 Vol 85 • No 5

13. Kung AW, Pang RW, Lauder I, Lam KS. 1995 Changes in serum lipoprotein(a) and lipids during treatment of hyperthyroidism. Clin Chem. 41:226 –231. 14. Tanis BC, Westendorp RGJ, Smelt AHM. 1996 Effect of thyroid substitution on hypercholesterolaemia in patients with subclinical hypothyroidism: a reanalysis of intervention studies. Clin Endocrinol (Oxf). 44:643– 649. 15. Verdugo C, Perrot L, Ponsin G, Valentin C, Berthezene F. 1987 Time-course of alterations of high density lipoproteins (HDL) during thyroxine administration to hypothyroid women. Eur J Clin Invest. 17:313–316. 16. The Expert Panel. 1998 Report of the national cholesterol education program expert panel on detection, evaluation and treatment of high blood cholesterol in adults. Arch Intern Med. 148:36 – 69. 17. Mueller B, Zulewski H, Huber P, Ratcliff JG, Staub JJ. 1995 Impaired action of thyroid hormone associated with smoking in women with hypothyroidism. N Engl J Med. 333:964 –969. 18. Wiseman SA, Powell JT, Humphries SE, Press M. 1993 The magnitude of the hypercholesterolemia of hypothyroidism is associated with variation in the low density lipoprotein receptor gene. J Clin Endocrinol Metab. 77:108 –112. 19. Kuivenhoven JA, de Knijff P, Boer JMA, et al. 1997 Heterogeneity at the CETP gene locus: influence on plasma CETP concentrations and HDL cholesterol levels. Arterioscler Thromb Vasc Biol. 17:560 –568. 20. Wiersinga WM, Chopra IJ. 1982 Radioimmunoassays of thyroxine (T4), 3,5,3⬘triiodothyronine (T3) and 3,3⬘-diiodothyronine (T2). Methods Enzymol. 84:272–303. 21. Moore D. 1995 Preparation and analysis of DNA. In: Ausubel FM, Brent R, Kingston RE, et al., eds. Current protocols in molecular biology, ed. 1. Vol. 1. New York: John Wiley & Sons, Inc.; Chapter 2, section 2.1.1. 22. Hobbs HH, Esser V, Russell DW. 1987 AvaII polymorphisms in the human LDL receptor gene. Nucleic Acids Res. 15:379. 23. Draya D, Lawn R. 1987 Multiple RLFPs at the human cholesteryl ester transfer protein (CETP) locus. Nucleic Acids Res. 11:4698. 24. Leitersdorf E, Chakravarti A, Hobbs HH. 1989 Polymorphic DNA haplotypes at the LDL receptor gene locus. Am J Hum Genet. 44:409 – 421. 25. Kuivenhoven JA, Jukema JW, Zwinderman AH, et al. 1998 The role of a common variant of the cholesteryl ester transfer protein gene in the progression of coronary atherosclerosis. N Engl J Med. 338:86 –93.