Werner & Ingbar's The Thyroid: A Fundamental & Clinical Text, 9th Edition

20.Genetic Factors in Thyroid Disease

Simon H.S. Pearce

Pat Kendall-Taylor

The spectrum of clinical thyroid disease that is strongly influenced by genetic factors runs from congenital hypothyroidism and thyroid agenesis, through transient gestational hyperthyroidism and autoimmune thyroid diseases (AITDs), to benign and malignant thyroid tumors. Indeed, a family history is a particularly frequent finding in patients with all forms of thyroid autoimmunity and for many types of goiter. This chapter reviews the inherited (germ line) basis for thyroid diseases, including those disorders with a strict monogenic (Mendelian) basis and those more common conditions that are inherited as complex multigenic traits. The pathogenesis of thyroid tumors and the role of somatic gene mutations in thyroid disease is covered in detail elsewhere (see section on oncogenes in Chapters 70 and 71).


Monogenic disorders of the thyroid axis frequently present as congenital or childhood hypothyroidism. They are usefully classified according to the level of the defect, i.e., pituitary thyrotrophs, thyroid development or hormonogenesis, hormone transport, and end-organ responsiveness (Table 20.1). Developmental problems of the thyroid account for the majority of infants with congenital hypothyroidism and are broadly referred to as thyroid dysgenesis, which includes thyroid agenesis, hemiagenesis, hypoplasia, and ectopic thyroid (i.e., abnormalities of both morphogenesis and migration) (1,2). Less than 15% of infants with thyroid dysgenesis are found to have one of the currently defined molecular defects (3,4,5,6). Of these gene abnormalities, only infants with a PAX8 gene mutation appear to have a disorder that is manifest as thyroid dysgenesis alone (6). Interestingly, even in families with a characterized PAX8 molecular abnormality, the morphology of the thyroid abnormality is not always consistent between patients (6,7). Defects in thyroid hormonogenesis account for only about 10% of congenital hypothyroidism, and often present with goiter as an additional feature (see Chapter 48). There undoubtedly are other single gene defects, and most likely some complex ones causing or modifying both thyroid dysgenesis and hormonogenesis, that await discovery.




Gene Defect (Inheritance)


Reference No.


Hypothalamic-pituitary disorders


Insensitivity to TRH

Hypothyroidism, no response to TRH




Combined pituitary hormone deficiency

Hypothyroidism, serum TSH low to slightly high

Pit1 [POU1F1] (AR, AD)






-with rigid cervical spine




-with septo-optic dysplasia




Isolated TSH deficiency


TSH-β (AR)



Thyroid development


Thyroid dysgenesis

Hypothyroidism, ectopic or hypoplastic thyroid




Variable degree of hypothyroidism, pulmonary hypoplasia and choreoathetosis

TTF1 [NKX2–1] (haploinsufficient)



Thyroid agenesis

Hypothyroidism, with palatal clefts, spiky hair, choanal atresia [Bamforth-Lazarus syndrome]




Thyroid hormonogenesisb


Iodide transport defect

Hypothyroidism, goiter




Total organification defect

Hypothyroidism, goiter




Hypothyroidism, goiter




Partial organification

Transient neonatal hypothyroidism





Goiter, variable sensorineural deafness, [Pendred's syndrome]




Thyroglobulin defect

Variable hypothyroidism, goiter

Tg (AR)




Simple goiter

Tg (AD)



TSH-receptor signaling


Complete TSH resistance

Hypothyroidism, thyroid hypoplasia




Partial TSH resistance

Euthyroid hyperthyrotropinemia




Pseudohypoparathyroidism type 1a, parathyroid hormone resistance, osteodystrophy

GNAS1 (AD & paternal imprinting)



Familial gestational hyperthyroidism

Hyperthyroidism in pregnancy

TSH-R (AD) [K183R]



Hereditary toxic thyroid hyperplasia

Congenital hyperthyroidism and goiter

TSH-R activation (AD)



Toxic thyroid adenoma

Autonomous follicular adenoma

TSH-R activation (somatic)



McCune-Albright syndrome, polyostotic fibrous dyplasia, precocious puberty

GNAS1 activation (mosaic)



Thyroid hormone transport


Euthyroid hypothyroxinemia

TBG deficiency (complete or partial)




Euthyroid, excess serum total thyroid hormones

TBG excess

TBG duplication (XR)



Increased TTR affinity




Familial dysalbuminemic hyperthyroxinemia

Alb (AD)



Familial dysalbuminemic hypertriiodothyroninemia

Alb (AD)



Transmembrane defect

Low serum T4, high serum T3, psychomotor retardation, nystagmus


33, 33A

End-organ response



Resistance to thyroid hormone

High serum T4 and T3, normal to high serum TSH, goiter




Thyroid tumorigenesis


Medullary carcinoma

Familial medullary carcinoma




MEN2a with pheochromocytoma





MEN2b with ganglioneuroma and pheochromocytoma




Thyroid hamartoma/carcinoma and thyroiditis

Breast adenoma/carcinoma, skin hamartoma [Cowden's syndrome]




Differentiated thyroid carcinoma

Short stature, cataract, skin changes, type 2 diabetes [Werner syndrome]




Papillary carcinoma

with polyposis coli [Gardner's syndrome]




 Thyroid adenoma/carcinoma

Pigmentary adrenal adenoma, cardiac myxoma, [Carney complex]




Thyroid autoimmunity


Autoimmune hypothyroidism

Hypoparathyroidism, hypoadrenalism, candidiasis [APECED syndrome, APS1]




 Immune diabetes, autoimmune enteropathy [IPEX syndrome]






Hypothyroidism, renal failure, corneal deposition





Thyrotoxic periodic paralysis

Muscle paralysis when hypokalemic





aOMIM no. refers to the catalog of “Online Mendelian Inheritance in Man,” as found on www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=OMIM .

bAn additional form of dyshormonogenesis, dehalogenase deficiency, has been characterized at a biochemical, but not molecular, level.

Alb, albumin; APC, adenomatous polyposis coli; APECED, autoimmune polyendocrinopathy candidiasis and ectodermal dysplasia; APSI, autoimmune polyendocrinopathy syndrome type 1; CTNS, cystinosin; FOXP3, Forkhead box P3; GNAS1, Gs-alpha subunit gene; KCNE3, Voltage gated potassium channel-3; MCT8, Monocarboxylate transporter-8; NKX2-1, (no full form); PAX8, paired box gene-8; PDS, Pendrin; PRKAR1A, Regulatory c-AMP dependent protein kinase-1α; PTEN, Phosphatase and tensin homolog; RECQL2, RECQ-like2; RET, (no full form); THOX2, thyroid oxidase-2; THRB, Thyroid hormone receptor-β; TTF1, thyroid transcription factor-1; TTF2, thyroid transcription factor-2.

Inherited disorders of thyroid hormone transport comprise abnormalities of both the serum-binding and transmembrane passage of thyroid hormones. Abnormalities of serum transport are due to altered quantity or binding affinity of the serum binding proteins, thyroxine-binding globulin (TBG), transthyretin, and albumin. Thyroxine-binding globulin is the major carrier protein, and abnormalities range from complete TBG deficiency, altered affinity for thyroid hormones, to excess production of TBG (27,28). The TBG gene is located on the X chromosome, and the biochemical manifestations of abnormal TBG production are more marked in males. Alleles encoding TBG isoforms with altered binding affinity are particularly prevalent in certain racial groups (29). Mutations in the albumin gene cause familial dysalbuminemic hyperthyroxinemia (30,31). Interestingly, a kindred with an albumin variant that has a high affinity for triiodothyronine (T3) but not thyroxine (T4) has been reported, demonstrating that the binding of T4 and T3 has different molecular specificities within the albumin molecule (32).

With the widespread use of serum thyrotropin (TSH) and free thyroid hormone assays, these familial states of euthyroid hypothyroxinemia and euthyroid hyperthyroxinemia, which are only apparent on assay of serum total hormone concentrations, rarely cause significant diagnostic confusion. These conditions of altered biochemistry are in marked contrast to the devastating disorder recently described in male children with abnormal transmembrane thyroid hormone transport (33). These children present with hypotonia, severe psychomotor retardation, failure of speech, nystagmus, and a low serum T4 concentration in conjunction with a high serum T3 concentration (33). Serum TSH values are normal at birth and become moderately elevated, but the developmental abnormalities are not reversed by T4 treatment. Mutations in an X-chromosome membrane transporter, MCT8, are responsible for this disorder (33). The disorders of end-organ responsiveness to thyroid hormones are termed resistance to thyroid hormone (RTH) and are caused by a variety of dominantly inherited mutations of the thyroid hormone β receptor (34). RTH syndromes have a variable phenotype, which ranges from asymptomatic goiter or tachycardia to hyperactivity, learning difficulty, developmental delay, and retarded bone growth with short stature. The biochemical hallmarks of these conditions are high serum thyroid hormone and normal or slightly high serum TSH concentrations (see Chapter 81).


Several chromosomal deletion or rearrangement syndromes are associated with thyroid disorders. The most common disorders in which thyroid dysfunction is a frequent accompaniment are listed in Table 20.2. Patients with Down's syndrome, with trisomy of chromosome 21, have a high prevalence of both autoimmune and congenital thyroid disease. Autoimmune hypothyroidism (chronic autoimmune thyroiditis) is the most common thyroid problem, affecting 15% to 20% of adults with Down's syndrome (47,48,49). Graves' disease occurs less frequently, with a prevalence of 1% to 2%, which may not be increased compared with the general population (47,48). Congenital hypothyroidism is found in about 2% of infants with Down's syndrome. Turner's syndrome, which is characterized by complete or partial loss of X-chromosome material in a phenotypic female, is also associated with autoimmune thyroid disease (AITD). About 15% of patients with Turner's syndrome have autoimmune hypothyroidism, with 30% to 40% having high serum antithyroid antibody concentrations (50,51,52). There is a correlation between cytogenetic abnormalities and AITD in patients with Turner's syndrome, with overt hypothyroidism being found in up to 40% of patients with an Xq isochromosome (deleted short arm and duplicated long arm) (52,53,54). The DiGeorge syndrome (cardiac outflow tract defects, thymic hypoplasia, hypoparathyroidism, and facial anomalies) and the overlapping chromosome 22q11 deletion syndromes are associated with Graves' disease in up to 20% of cases (55,56,57). Two rarer chromosomal disorders, 1p terminal deletion and Smith-Magenis syndrome, are each associated with hypothyroidism in about one third of cases (Table 20.2). The pathogenesis of the hypothyroidism is ill defined, although it is likely to be developmental in nature.



Thyroid Phenotype

Other Features

Cytogenetic Abnormality


Reference No.

Down's syndrome

Autoimmune hypothyroidism

Mental retardation, cardiac anomalies, characteristic facies, others

Trisomy 21



DiGeorge/CATCH22 syndrome

Graves' disease

Cardiac outflow tract anomalies, thymic hypoplasia, hypoparathyroidism




Turner's syndrome

Autoimmune hypothyroidism, Thyroid antibodies

Web neck, ovarian dysgenesis, short stature, renal and aortic root abnormalities




Smith-Magenis syndrome


Mental retardation, eye anomalies, self-injury




1p terminal deletion


Mental retardation, microcephaly, large fontanelle, hearing loss, characteristic facies

1p36 del



aOMIM no. refers to the catalog of “Online Mendelian Inheritance in Man,” as found on www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=OMIM .


Complex disorders are determined by a combination of genetic and nongenetic factors (including environmental and other influences). Most common complex traits are thought to have a multigenic basis; i.e., in different individuals they are determined by a varying combination of susceptibility alleles in different genes (61). However, even for an apparently simple genetic trait, the phenotype may be dependent to some extent upon environmental influences. For instance, in mendelian forms of thyroid dyshormonogenesis the presence or severity of hypothyroidism and the rate of goiter development may be determined by dietary iodide availability. AITD is the most prevalent of the autoimmune conditions, with at least 2% of women being affected by overt thyroid dysfunction. Similarly, simple euthyroid goiter is a frequent finding in women, even in iodine-replete areas. Both these common disorders have a complex multigenic basis.

Simple (Euthyroid) Goiter

Simple goiter is defined as a diffuse or nodular enlargement of the thyroid that is not the result of neoplasia or inflammation in a subject who is euthyroid (62). It is a heterogeneous disorder that affects up to 15% of women and about 3% of men who live in iodine-sufficient regions (63,64). It is an important problem on a population level due to its prevalence, and also because some subjects with simple goiter may be predisposed to nonmedullary thyroid cancer (65,66,67). However, it is notable that in the majority of patients who have recurrent thyroid nodules after surgery for goiter the new nodules are polyclonal in nature, indicating that the condition is predominantly a hyperplastic rather than neoplastic process (68). Surprisingly, in iodine-deficient regions, goiter has a familial tendency (69), with a higher concordance in monozygotic than dizygotic twins (70). In the absence of iodine deficiency, simple goiter also has a strong genetic basis (71). In a large Danish cohort of female twins, the concordance rate in monozygotic twins was 42%, as compared with 13% in dizygotic twins (72), and data from this study suggested that about 80% of the susceptibility to simple goiter may be determined by genetic factors.

Heterozygous mutations in the thyroglobulin (Tg) molecule have been described in three kindreds with simple goiter, identified from 30 probands (19). In these families, about 50% of patients carrying the mutation had a goiter, with an increasing penetrance after the age of 25 years (19). Thus, at least 10% of cases of simple goiter may be due to mild defects in thyroid hormonogenesis (19,72). However, in contrast to the “dyshormonogenic” presentation as a recessive trait associated with hypothyroidism, these Tg mutations are manifest as euthyroid goiter in later life and are inherited in an autosomal-dominant fashion (19,72,73). In recent years, several large kindreds with dominantly inherited multinodular goiter have been identified, and genome-wide linkage studies have mapped novel susceptibility loci to chromosomes 14q32, Xp22, and 3q26 (designated MNG1 to MNG3) (74,75,76). Of these three loci, only MNG1 on 14q32 has been independently confirmed (77).

Of the three families in whom linkage was initially established, the phenotype of one kindred was distinct in having an onset of goiter in prepubertal children, rapid 123I uptake, and high serum TSH concentrations in some subjects, suggesting that the MNG3 locus may encode another mild form of dyshormonogenesis (76). Reassuringly, these linkages appear distinct from those for familial nonmedullary thyroid cancer (1p21, 2q21, 19p13) (78,79,80) (see section on oncogenes in Chapter 70). Thus, simple goiter is certainly a complex trait in terms of its molecular pathogenesis, with iodine availability as the major environmental component. Whether its inheritance is as a multiple monogenic trait comprising a series of different single gene disorders with a common phenotype, or is a truly complex trait with alleles at several genetic loci contributing to the disorder in each patient, is unknown. It will be interesting to determine whether the gene defects underlying MNG1 to MNG3 are subtle abnormalities of hormonogenesis or whether a proportion of cases will be due to inherited defects in the regulation of cell cycling or growth of thyroid follicular cells.

Autoimmune Thyroid Disease

AITD comprises a series of interrelated conditions including Graves' disease, Hashimoto's (goitrous) thyroiditis, atrophic autoimmune hypothyroidism, postpartum thyroiditis, and thyroid-associated ophthalmopathy. These different manifestations of AITD may occur sequentially, and sometimes synchronously, in the same patient. This clustering of different phenotypes of AITD within a patient suggests that these conditions have a common pathophysiologic basis (see section on pathogenesis of Graves' disease in Chapters 23 and 47). A widely accepted model for the pathogenesis of AITD suggests that each patient has a background inherited predisposition to autoimmunity, with additional environmental and hormonal factors that trigger or contribute to the development of disease. In support of this model, there is good evidence that both cigarette smoking and adverse psychosocial events are associated with the development of Graves' disease (81,82,83). Similarly, the female preponderance of AITD (63), the modulation of animal models of the disease with sex steroids (84), the amelioration of Graves' disease during pregnancy, and the occurrence of postpartum thyroiditis all support the important role of sex steroids in these disorders.

Genetic Epidemiology of Autoimmune Thyroid Disease

Twin studies show an increased concordance of Graves' disease and autoimmune hypothyroidism in monozygotic twins as compared to dizygotic twins (reviewed in reference 85). In a series of careful investigations of 8966 Danish twins, concordance for Graves' disease in monozygotic twins was 35% as compared with 3% in dizygotic twins (86). Similarly, there was 55% concordance of autoimmune hypothyroidism (combining both atrophic and goitrous forms of the disease) in monzygotic twins, but no concordance among dizygotic twins (87). A statistical model based on these data suggested that 79% of the predisposition to Graves' disease is due to genetic factors, with only 21% due to nongenetic (environmental and hormonal) factors (86), confirming the dominant role of heredity in the pathogenesis of the disease.

In addition to twin studies, a measure of the heritability of a disorder can be gained from the ratio of the risk to a relative of an affected proband compared with the background population prevalence (a value termed λs for siblings and λo for parents/offspring) (61). A study from Hungary found 5.3% of 435 probands with Graves' disease had siblings with the disease (21 sisters, 2 brothers), compared with a background population frequency of Graves' disease of 0.6% (88). This allows estimation of the λs for Graves' disease as 8.1 in this population. In a similar study carried out in England, the λs for Graves' disease was 9.9 (89). To put these values into context, the λs for type 1 diabetes mellitus is 15, and that for rheumatoid arthritis is 8 (90). An important message for patients with Graves' disease is that their female siblings and children have a 5% to 8% chance of also being affected by the disease, and an approximately similar risk of developing autoimmune hypothyroidism.

Heritability of Thyroid Antibodies

Soon after the first description of thyroid antibodies (91), their occurrence in 22 of 39 siblings (56) of probands with AITD was noted, and a dominant pattern of inheritance was suggested (92). The finding of thyroid antibodies in 30% to 50% of first-degree female relatives has been confirmed using sensitive assays for serum anti-thyroglobulin (Tg) and antithyroid peroxidase (TPO) antibodies (93). The prevalence of these antibodies in male relatives is less (~10% to 30); this pattern has been attributed to dominant inheritance with reduced penetrance in males (93). Naturally occurring TPO antibodies are not uniform in specificity, and tend to be directed against certain defined “immunodominant” epitopes of TPO. Interestingly, these patterns of TPO antibody response (epitopic fingerprints) can be inherited in a dominant fashion (94), and can persist for many years. It is striking that even in families in which several members have antibodies to TPO or Tg, clinically overt AITD is not the rule (95). This demonstrates that, although the generation of a B-cell immune response to thyroid antigens is one component of thyroid autoimmunity, it is not in itself sufficient to cause disease, and that other tissue-specific responses or immune system factors are also necessary.

Association of Autoimmune Thyroid Disease with Other Autoimmune Disorders

The occurrence of Hashimoto's thyroiditis in patients with type 1 diabetes mellitus, and in their family members, is well recognized. In large groups of families with type 1 diabetes in the United Kingdom and the United States, at least one case of AITD was reported in relatives of 22% and 40% of patients with type 1 diabetes, respectively (96,97). A higher-than-expected prevalence of AITD has been found in patients with other autoimmune disorders and in their families (98). In two studies of families with two or more rheumatoid arthritis cases, about 50% of the families had a patient with rheumatoid arthritis or a relative had AITD (99,100). Similarly, AITD is often found in patients with autoimmune Addison's disease, being present in 40% to 50% of patients with sporadic Addison's disease [i.e., those who do not have the autoimmune polyendocrinopathy, candidiasis, or ectodermal dysplasia (APECED) syndrome] (101,102). Other weaker associations exist between AITD and other autoimmune endocrinopathies (e.g., autoimmune premature ovarian failure) and nonendocrine autoimmune disorders (e.g., pernicious anemia, celiac disease, myasthenia gravis, multiple sclerosis) (103). This clustering of different autoimmune conditions suggests that several different autoimmune disorders are likely to have a disease susceptibility allele (or alleles) in common.

Monogenic Disorders with Autoimmune Thyroid Disease as a Component

AITD is a component of three monogenic syndromes. The most common of these rare disorders is the APECED (type 1 polyendocrinopathy) syndrome (Table 20.1). The cardinal features of this condition are autoimmune hypoparathyroidism, Addison's disease, and chronic mucocutaneous candidiasis, which begin in childhood or early adolescence (see reference 42 for a comprehensive review). Other autoimmune disorders such as type 1 diabetes, pernicious anemia, and hypogonadism occur in 15% to 30% of patients with the APECED syndrome, but autoimmune hypothyroidism is comparatively uncommon, affecting only about 5% of patients. The extreme rarity of Graves' disease in APECED is notable. APECED is an autosomal recessive syndrome caused by mutations in the autoimmune regulator (AIRE) gene on chromosome 21 (43). AIRE is a nuclear transcription factor that is expressed predominantly in dendritic antigen presenting cells in the thymus and peripheral lymphoid tissues. AIRE gene mutations lead to defective negative selection of potentially autoreactive thymocytes and ineffective peripheral antigen presentation for some antigens (e.g., Candida). One large family has been described with dominantly inherited severe and persistent candidiasis and hypothyroidism with goiter, a condition that is distinct from APECED. Linkage studies found that the disorder mapped to chromosome 2p, but the causative gene has not been identified (104). A third, rarer disorder is the immune dysregulation, polyendocrinopathy, and enteropathy (X-linked) syndrome (IPEX). This is a devastating disorder of male infants with failure to thrive due to autoimmune enteropathy (44). Autoimmune hypothyroidism and type 1 diabetes develop in the first year of life in about 50% and 90% of affected males, respectively. IPEX is caused by mutations in the FOXP3 gene on chromosome Xp11. FOXP3 is expressed in T lymphocytes and is critical to the development of regulatory T cells, without which inappropriate activation and proliferation of CD4 lymphocytes occurs (44). These monogenic syndromes point towards two cell types, namely the dendritic antigen-presenting cell and the CD4 lymphocyte, whose normal functioning is critical to the maintenance of immune tolerance to thyroid tissues.

Autoimmune Thyroid Diseases as Complex Genetic Traits

In contrast to the unusual monogenic forms of autoimmunity mentioned above, most cases of AITD, along with other common autoimmune disorders, are now thought to have a complex genetic basis; i.e., the genetic predisposition to the disease is determined by a series of interacting susceptibility alleles of several different genes (90). These various genetic loci may also have differing influences on the predisposition to AITD in different populations (locus heterogeneity), which makes the identification of disease susceptibility genes a more difficult task. There are two standard approaches for identifying disease genes for either monogenic or complex traits (61). First, candidate gene studies involve examining polymorphic markers within a particular gene, which has been selected because it is thought that disruption of its function may result in the phenotype. There has been some success in using this approach to identify susceptibility genes for AITD. A second approach is linkage scanning, in which widely spaced anonymous genetic markers (usually microsatellite repeat polymorphisms between genes) are used to detect chromosomal segments with evidence for linkage in affected families. Linkage analysis is more robust than association analysis, because it may detect genetic effects many millions of nucleotides away and does not rely on the same allele being linked to disease in each kindred. However, linkage studies may lack sensitivity to detect loci with small effects and are poor at localizing a disease allele. Several genetic tests now combine elements of familial linkage and association analysis (linkage in the presence of association), the most widely used analysis being the transmission disequilibrium test (105).

Luckily for investigators interested in AITD, linkage studies of type 1 diabetes have been prominent in the field of complex trait mapping. It is now clear from genome-wide linkage studies encompassing more than 750 families with type 1 diabetes that there are seven loci where there is firm or suggestive evidence of genetic linkage (106,107). Of these loci, one is the major histocompatibility complex (MHC) on chromosome 6p21, another is the insulin gene region on chromosome 11p15, and a third is the cytotoxic T lymphocyte antigen-4 (CTLA-4) gene on chromosome 2q33 (see later in the chapter). The other four linked loci, and a further 12 putative type 1 diabetes loci, have all been defined on the basis of linkage to anonymous markers, and there is no defined allelic association. These studies of type 1 diabetes have advanced the field by demonstrating that a common autoimmune disorder has a complex (multigenic) genetic basis (90). Indeed, similar findings have subsequently been made for rheumatoid arthritis, multiple sclerosis, and systemic lupus erythematosus, albeit with smaller numbers of families (108,109). There is likely to be a similar number of genetic loci implicated in the pathogenesis of AITD, and because of the familial clustering of the disease with type 1 diabetes, some susceptibility alleles are likely to be shared by both disorders.

Genetic studies of AITD have proceeded in recent years with a series of linkage and association studies, and by fine-mapping studies. Although some progress has been made, there is still no detailed molecular view of the pathogenesis of AITD. The limiting factors for these genetic approaches have been small sample sizes, disease loci with small effects, and the expense and relatively slow speed of genotyping large numbers of people for multiple markers.

The Major Histocompatibility Complex

The MHC, which contains the human leukocyte antigen (HLA) genes, is located on chromosome 6p21. It is subdivided into three regions: the class I region, which encodes the HLA antigens A, B, and C; the class II region, which encodes HLA antigens DR, DQ, and DP, each with one or more α and β chains; and the class III region, encoding several immunoregulatory molecules including complement components, heat shock protein 70 (HSP70), and tumour necrosis factors (TNF) (Fig. 20.1). The class II region also contains the peptide transporters associated with antigen processing (TAP) and large multifunctional protease (LMP) genes. Tight linkage disequilibrium (i.e., conserved haplotypes) exists between the alleles of the MHC region. Major histocompatibility complex class II molecules play a critical part in the initiation of adaptive immune responses. Peptide antigens can only be recognized by T-cell antigen receptors when they are attached to the binding groove of an MHC molecule on the surface of an antigen-presenting cell.

FIGURE 20.1. Linkage disequilibrium map of the human major histocompatility complex (MHC) region. A gene and single nucleotide polymorphism (SNP) map showing seven regions (*) with highly conserved haplotype blocks, the linkage disequilibrium measure (D′), and recombination rates across the MHC region. The terminal end of the 6p arm is to the left in the upper chain, and the centromere is to the right of the lower chain. The numerous genes in the MHC region can be seen, with some areas of particularly strong linkage disequilibrium. (From Walsh EC, Mather KA, Schaffner SF, et al. An integrated haplotype map of the human major histocompatibility complex. Am J Hum Genet 2003;73:580, with permission.)

The first recognition of an association between Graves' disease and alleles of MHC class I molecules was the finding of a higher frequency of the HLA-B8 allele in patients with Graves' disease (47), as compared with normal subjects (21) (110). Several studies in white populations confirmed this association (111,112). However, a stronger association of Graves' disease was found with the MHC class II allele, HLA-DR3, which is in strong linkage disequilibrium with HLA-B8 (113). Many case-control studies in white populations have since found Graves' disease to be associated with HLA-DR3, with relative risks between 2.5 and 5 (reviewed in references 85 and 114). More recently, this association of MHC with Graves' disease was confirmed in a study using the transmission disequilibrium test, which showed preferential transmission of the HLA DRB1*0304-DQB1*02-DQA1*0501 (DR3-DQ2) haplotype (115). Although there is no doubt about the association of Graves' disease with the HLA-DR3 haplotype in whites, the primary disease susceptibility allele in the region remains unknown. An association of Graves' disease, particularly in males, with the HLA-DQA1*0501 allele, which was stronger than, and independent of, the HLA-DR3 status, has also been reported (116). This independent association of HLA-DQA1*0501 with Graves' disease has been supported by some studies (115,117,118), but not by others (119,120,121). In other populations, Graves' disease has been found to be associated with several different HLA alleles. For example, it has been found to be associated with HLA-B35, -B46, -A2, and -DPB1*0501 in Japanese patients (122,123,124); HLA-A10, -B8, and -DQw2 in Indians (125); HLA-DR1 and -DR3 in South African blacks (126); the DRB3*020/DQA1*0501 haplotype in black Americans (127); and HLA-B46, -DR9, -DRB1*303, and -DQB1*0303 in Hong Kong Chinese (128). Furthermore, case-control studies have also shown an association of Graves' disease with alleles of several different non-HLA genes within the MHC complex, including the HSP70TNFTAP, and LMP genes (121,129,130,131,132). Thus, it is likely that other, non-HLA genes within the MHC locus contribute to susceptibility to AITD. Due to the strong linkage disequilibrium within this region (Fig. 20.1), it is difficult to determine the independent effect of a particular allele in disease susceptibility, and much larger association studies are needed to resolve these issues.

The results of HLA association studies in autoimmune thyroiditis have been less consistent. In whites, autoimmune hypothyroidism has been associated with HLA alleles that include HLA-B8, -DR3, -DR4, -DR5, -DQA1* 0201/*0301, and -DQB1*03 (121,133,134,135,136,137). Small sample sizes and phenotypic heterogeneity (for example, goitrous vs. atrophic autoimmune hypothyroidism) make it difficult to draw a firm conclusion from these studies, but they suggest a modest association of autoimmune hypothyroidism with HLA-DR3, -DR4, and -DR5. Postpartum thyroiditis has also been found to be associated with HLA-DR4 (135,138) and -DR5 (139,140), suggesting a close relationship between these two disorders.

Although case-control studies in whites have consistently demonstrated an association of Graves' disease with HLA alleles, the results of familial linkage studies of the HLA locus in the disease have been discrepant. Among early studies, several found excess sharing of HLA alleles in sibs affected with the disease, suggesting linkage (88,141), but linkage was not confirmed in other studies (reviewed in reference 89). However, all the studies were small, and there was phenotypic heterogeneity in the kindreds in some studies (e.g., inclusion of members with both Graves' disease and autoimmune hypothyroidism) (89). Based on a transmission disequilibrium test approach in a larger cohort of Graves' disease families, Graves' disease was linked to HLA (115). Linkage analysis in a cohort of sib-pairs with Graves' disease in the United Kingdom also found modest evidence to support linkage of the disease to the MHC region (142). However, three recent genome-wide scans in families with AITD have failed to detect linkage at chromosome 6p21 markers (143,144,145). Therefore, the contribution of genes in the MHC region to the genetic susceptibility to Graves' disease is comparatively small, perhaps accounting for 10% to 20% of the inherited susceptibility in certain populations (142). This is in marked contrast to type 1 diabetes, in which MHC is the dominant locus, accounting for 30% to 40% of the total genetic susceptibility (106,107). In addition, most of the susceptibility to type 1 diabetes within the MHC locus is thought to be encoded by HLA-DQB and -DRB alleles, with lesser contributions from other adjacent genes (146,147). For AITD, it is likely that other, non-HLA genes within 6p21 determine a more substantial part of the inherited predisposition encoded by this region of the genome.

Cytotoxic T Lymphocyte Antigen-4

CTLA-4 is an immunoregulatory molecule that is expressed on the surface of activated T lymphocytes. Several lines of evidence support its role as a key negative regulator of T-cell activation. CTLA-4 knockout mice develop a rampant lymphoproliferative disorder resulting in splenomegaly, lymphadenopathy, and death from autoimmunity before 3 to 4 weeks of age (148). The administration of CTLA-4 blocking antibodies also precipitates or exacerbates autoimmunity in several murine models (149). Conversely, a soluble fusion protein of CTLA-4 and the immunoglobulin G1 Fc region (CTLA-4-Ig) ameliorates different experimental autoimmune disorders, including diabetes (150). Furthermore, the engagement of CTLA-4 with its ligands arrests the progression of activation-induced T-cell cycling (151), possibly by inducing apoptosis of activated T cells (152). This regulatory role of CTLA-4 in T-cell activation makes the CTLA-4 gene an attractive candidate locus for autoimmune disorders.

In 1995, Yanagawa and coworkers reported a significant association of Graves' disease with an allele of a microsatellite polymorphism (designated CTLA-4[AT]n) within the 3′-untranslated region of the CTLA-4 gene (153). Subsequently, the G allele of a single-nucleotide polymorphism that encoded a threonine to alanine change within the signal peptide of CTLA-4 (CTLA-4[49]A/G) was also found to be associated with Graves' disease (154). The association of these two CTLA-4 polymorphisms with Graves' disease has been confirmed by subsequent studies in many different populations, with odds ratios between 1.4 and 3.2 (reviewed in reference 89). Of note, one study in Tunisians found the association with Graves' disease was with the A allele at CTLA-4[49]A/G (155), suggesting that despite the change in the coding sequence encoded by this polymorphism, it was probably not the disease allele. The G allele at the CTLA-4[49]A/G polymorphism was also found to be associated with autoimmune hypothyroidism in several different populations, although with less consistency than with Graves' disease (reviewed in reference 156). This is most likely to be due to the small size of some of the studies, and perhaps the phenotypic heterogeneity in ascertainment of autoimmune hypothyroidism.

Despite the clear finding of association of CTLA-4 alleles with Graves' disease and autoimmune hypothyroidism in many studies from different populations, initial family studies failed to detect linkage at CTLA-4 (143,157). However, detailed analysis of the CTLA-4 locus in a cohort of sib pairs in the United Kingdom revealed strong evidence of linkage to CTLA-4 (peak nonparametric linkage score 3.4), conferring up to one third of the total genetic susceptibility to Graves' disease in this population (142). Significant linkage of CTLA-4 to thyroid autoantibody status was also subsequently reported (158), but genome-wide scans in Chinese families with Graves' disease and Japanese families with AITD did not show linkage at CTLA-4 (144,145).

In order to identify the underlying susceptibility polymorphism within CTLA-4, a detailed fine-mapping study of the entire CTLA-4 locus using markers for 108 single nucleotide polymorphisms was done (Fig. 20.2) (159). Using a regression model, the marker most highly associated with the disease (designated CT60) was located in the 3′ untranslated region of the gene, and the susceptibility allele had an odds ratio of about 1.5 for both Graves' disease and autoimmune hypothyroidism, but a lower contribution to type 1 diabetes (odds ratio ~1.2). The CT60 polymorphism appears to regulate a splice variant of an upstream exon, which determines the expression of membrane-bound versus soluble forms of the mature CTLA-4 molecule. It is hypothesized that under-expression of the mRNA for the soluble CTLA-4 isoform (encoded by the CT60 allele) leads to an imbalance in positive versus “negative” costimulatory signaling to activated T lymphocytes (Fig. 20.3). The CTLA-4 CT60 susceptibility allele has a high population prevalence (60% carrier rate in whites in the United Kingdom), and is likely to be a susceptibility allele not just for AITD and type 1 diabetes, but also for other autoimmune disorders. The high frequency of this low penetrance “disease” allele/haplotype in the normal population is the probable explanation as to why several linkage studies failed to identify the locus (160). A key question is whether this allele can be used to predict the outcome of treatment of Graves' disease, or the development of Graves' ophthalmopathy (161,162). Finally, the effect of nongenetic factors (e.g., smoking, pregnancy, sex hormones) on the expression of the soluble form of CTLA-4 are needed to understand better the interaction of the genetic predisposition and “environmental” factors.

FIGURE 20.2. Fine-mapping of markers of single nucleotide polymorphisms markers at CTLA-4 and adjacent loci. A plot of -log P values in favor of association against the physical map of the markers in the adjacent transcripts CD28, CTLA-4, and ICOS on chromosome 2q33, each of which has a role in lymphocyte costimulation. The most evidence for association is found at the CT60 polymorphism within the 3′ untranslated region of CTLA-4. Alleles of this polymorphism were found to correlate with relative amounts of membrane-bound and soluble CTLA-4 production. (From Ueda H, Howson JM, Esposito L, et al. Association of the T-cell regulatory gene CTLA4 with susceptibility to autoimmune disease. Nature 2003;423:506, with permission.)

FIGURE 20.3. Schematic drawing of an antigen-presenting cell interacting with a T lymphocyte. The dendritic antigen-presenting cell presents a cleaved peptide antigen bound to the groove of the MHC class II molecule. This complex is recognized by a T lymphocyte with an antigen receptor (CD3 complex) with high affinity for the peptide-MHC complex. For the lymphocyte to become activated, a second signal must be delivered. This occurs when CD28 molecules on the T cell bind to B7 (CD80 and CD86) molecules on the antigen-presenting cell. This binding results in T-cell proliferation and activation. In contrast, T cells become quiescent or die when B7 molecules bind to CTLA-4 molecules. Either membrane-bound or soluble CTLA-4 may inhibit B7 binding to CD28, thereby stopping the costimulatory activation of the lymphocyte via the B7-CD28 interaction. Subjects with the susceptibility allele for autoimmune thyroid disease at CTLA-4 produce relatively small amounts of soluble CTLA-4 messenger RNA (159); if reflected at the protein level, this could allow a greater B7 binding to CD28, and therefore greater activation of T cells.

Other Candidate Autoimmune Thyroid Disease Genes and Results of Linkage Scanning

Case-control studies of many other candidate immune system and thyroid-specific genes for AITD have not led to substantial advances in our understanding of pathogenesis. Immunoregulatory genes, including the immunoglobulin heavy-chain (Gm), T-cell receptor β-chain, interleukin-1 receptor antagonist, interleukin-4, interferon-γ, and Fas ligand genes have been studied, but no allele of any of these genes has been consistently associated with AITD (reviewed in references 89 and 163). Similarly, studies of the TSH receptor, the thyroid receptor-β and TPO genes have not yielded significant associations (89). Studies of the vitamin D receptor and the vitamin D–binding protein gene have suggested a possible role in Graves' disease (164,165) but await replication.

In recent years, four different cohorts of patients have been examined for genetic linkage to a large number of anonymous chromosomal markers by US, UK, Japanese, and Chinese investigators (142,143,144,145,157,158,166,167,168). The results of these studies need to be viewed in the context that with 102 families (US study), 82 sib-pairs (UK study), 123 sib-pairs (Japanese study), and 54 families (Chinese study), no individual study has been large enough to detect susceptibility loci with modest effects (169). Thus, only highly significant linkages, or those that are replicated in a second cohort of patients, are likely to stand the test of time. Despite this caution, several promising chromosomal loci have emerged (Table 20.3) (170). In particular, loci for AITD on chromosomes 5q31–q33, 8q23, 18q21, and 20q13 have each been replicated to some degree by a second study (143,144,145,166,167,168,171), suggesting that novel disease susceptibility alleles are likely to be identified from at least some of these loci. Of particular interest, the chromosome 8 linkage appears to be strongest in patients with autoimmune hypothyroidism, and it overlaps the Tg locus, and in one study preliminary evidence for association within the Tg gene was found (172,173).


Chromosomal Locib

U.S. Study (Reference Nos. 143,157,158,168)

Japanese Study (Reference No. 144)

Chinese Study (Reference No. 145)

UK Study (Newcastle) (Reference No. 142,161,166,167)

UK Study (Birmingham) (Reference No. 115,131)









2q33 (CTLA-4)










6p21 (MHC)




6p (AITD1)

















12q22 (HT2)




13q32 (HT1)




14q31 (GD1)













20q11 (GD2)







Xq21 (GD3)



AITD1, Autoimmune thyroid disease-susceptibility to 1; CTLA-4, cytotoxic T lymphocyte antigen-4; GD1, Graves' disease-susceptibility to 1; GD2, Graves' disease-susceptibility to 2; GD3, Graves' disease-susceptibility to 3; HT1, Hashimoto's thyroiditis-susceptibility to 1; HT2, Hashimoto's thyroiditis-susceptibility to 2; MHC, major histocompatibility complex; ND, not done; TDT, transmission disequilibrium test.

aSignificance levels for linkage are designated according to the criteria of Lander and Kruglyak:—, no linkage; +, nominal linkage (LOD score >1.0); ++, suggestive linkage (LOD score >1.9); +++, significant linkage (LOD score >3.3). Equivalent nonparametric linkage scores are >1.2, >2.2, and >3.6 for nominal, suggestive and significant linkage, respectively (169).

bLoci found linked in more than one dataset are shown in bold.

cLinkage with thyroid autoantibody production (158).

dEvidence for linkage found only in the Graves' disease families with apparent dominant mode of transmission (a tentative finding).

Summary of Autoimmune Thyroid Disease Genetics

AITD can be considered a complex trait with a predominant genetic component. Several gene loci determine susceptibility to the disease, with a major contribution from CTLA-4. Currently, there is moderate evidence that other loci are involved to a lesser degree, including one or more genes in the MHC region. As work progresses, it is likely that other disease genes and immune system pathways will be identified that predispose to the disease. The evaluation of gene-environment interaction at a molecular level is likely to cast further light on disease pathogenesis.


In the last few years there have been unprecedented advances in knowledge about the sequence and organization of the human and murine genomes, such that there are now comprehensive gene maps of each chromosome, and tissue-specific expression profiles at both the mRNA and protein level are rapidly being developed. In time, these advances will more fully inform our knowledge of the physiology of normal thyroid function and its disturbance in many more disease states. Impressive progress in understanding both monogenic and complex thyroid disorders at a genomic level has been made in the last decade. The next decade holds the promise of allowing the integration of genetic and epidemiologic approaches to understanding disease, through a determination of the underlying molecular pathology.


1. Kopp P. Perspective: genetic defects in the etiology of congenital hypothyroidism. Endocrinology 2002;143:2019.

2. Van Vliet G. Development of the thyroid gland: lessons from congenitally hypothyroid mice and men. Clin Genet 2003;63: 445.

3. Clifton-Bligh RJ, Wentworth JM, Heinz P, et al. Mutation of the gene encoding human TTF-2 associated with thyroid agenesis, cleft palate and choanal atresia. Nat Genet 1998;19:399.

4. Krude H, Schutz B, Biebermann H, et al. Choreoathetosis, hypothyroidism, and pulmonary alterations due to human NKX2–1 haploinsufficiency. J Clin Invest 2002;109:475.

5. Pohlenz J, Dumitrescu A, Zundel D, et al. Partial deficiency of thyroid transcription factor 1 produces predominantly neurological defects in humans and mice. J Clin Invest 2002;109: 469.

6. Macchia PE, Lapi P, Krude H, et al. PAX8 mutations associated with congenital hypothyroidism caused by thyroid dysgenesis. Nat Genet 1998;19:83.

7. Congdon T, Nguyen LQ, Nogueira CR, et al. A novel mutation (Q40P) in PAX8 associated with congenital hypothyroidism and thyroid hypoplasia: evidence for phenotypic variability in mother and child. J Clin Endocrinol Metab 2001;86: 3962.

8. Collu R, Tang J Castagné J, et al. A novel mechanism for isolated central hypothyroidism: inactivating mutations in the thyrotropin-releasing hormone receptor. J Clin Endocrinol Metab 1997;82:1561.

9. Radovick S, Nations M, Du Y, et al. A mutation in the POU-specific domain of Pit1 and hypopituitarism without pituitary hypoplasia. Science 1992;257:1115.

10. Wu W, Cogan JD, Pfäffle RW, et al. Mutations in PROP1 cause familial combined pituitary hormone deficiency. Nat Genet 1998;18:147.

11. Netchine I, Soubrier M-L, Krude H, et al. Mutations in LHX3 result in a new syndrome revealed by combined pituitary hormone deficiency. Nat Genet 2000:25:182.

12. Dattani MT, Martinez-Barbera JP, Thomas PQ, et al. Mutations in the homeobox gene HESX1/hesx1 associated with septo-optic dysplasia in human and mouse. Nat Genet 1998; 19:125.

13. Hayashiziki Y, Hinoka Y, Tatsumi K, et al. Deoxyribonucleic acid analysis of five families with familial inherited thyroid stimulating hormone deficiency. J Clin Endocrinol Metab 1990; 71:792.

14. Fujiwara H, Tatsumi K, Miki K, et al. Congenital hypothyroidism caused by a mutation in the Na+/I- symporter. Nat Genet 1997;16:124.

15. Abramowicz M, Targovnik H, Varela V, et al. Identification of a mutation in the coding sequence of the human thyroid peroxidase gene causing congenital goiter. J Clin Invest 1992;90: 1200.

16. Moreno JC, Bikker H, Kempers MJE, et al. Inactivating mutations in the gene for thyroid oxidase 2 (THOX2) and congenital hypothyroidism. N Engl J Med 2002;347:95.

17. Everett LA, Glaser B, Beck JC, et al. Pendred syndrome is caused by mutations in a putative sulphate transporter gene (PDS). Nat Genet 1997;17:411.

18. Ieiri T, Cochaux P, Targovnik H, et al. A 3′ spliced site mutation in the thyroglobulin gene responsible for congenital goiter with hypothyroidism. J Clin Invest 1991;88:1901.

19. Corral J, Martín C, Pérez R, et al. Thyroglobulin gene point mutation associated with non-endemic simple goitre. Lancet 1993;341:462.

20. Biebermann H, Schoneberg T, Krude H, et al. Mutations of the human thyrotropin receptor gene causing thyroid hypoplasia and persistent congenital hypothyroidism. J Clin Endocrinol Metab 1997;82:3471.

21. Sunthornthepvarakul T, Gottschalk ME, Hayashi Y, et al. Resistance to thyrotropin caused by mutations in the thyrotropin-receptor gene. N Engl J Med 1995;332:155.

22. Weinstein LS, Gejman PV, Friedman E, et al. Mutations of the Gs alpha-subunit gene in Albright hereditary osteodystrophy detected by denaturing gradient gel electrophoresis. Proc Natl Acad Sci USA 1990;87:8287.

23. Rodien P, Bremont C, Sanson ML, et al. Familial gestational hyperthyroidism caused by a mutant thyrotropin receptor hypersensitive to human chorionic gonadotropin. N Engl J Med 1998;339:1823.

24. Duprez L, Parma J, Van Sande J, et al. Germline mutations in the thyrotropin receptor gene cause non-autoimmune autosomal dominant hyperthyroidism. Nat Genet 1994;7:396.

25. Parma J, Duprez L, Van Sande J, et al. Somatic mutations in the thyrotropin receptor gene cause hyperfunctioning thyroid adenomas. Nature 1993;365:649.

26. Mastorakos G, Mitsiades NS, Doufas AG, et al. Hyperthyroidism in McCune-Albright syndrome with a review of thyroid abnormalities sixty years after the first report. Thyroid 1997; 7:433.

27. Mori Y, Seino S, Takeda K, et al. A mutation causing reduced biological activity and stability of thyroxine-binding globulin probably as a result of abnormal glycosylation of the molecule. Mol Endocrinol 1989;3:575.

28. Mori Y, Miura Y, Takeuchi H, et al. Gene amplification as a cause of inherited thyroxine-binding globulin excess in two Japanese families. J Clin Endocrinol Metab 1995;80:3758.

29. Takeda K, Mori Y, Sobieszczyk S, et al. Sequence of the variant thyroxine-binding globulin of Australian aborigines. Only one of two amino acid replacements is responsible for its altered properties. J Clin Invest 1989;83:1344.

30. Moses AC, Rosen HN, Moller DE, et al. A point mutation in transthyretin increases affinity for thyroxine and produces euthyroid hyperthyroxinemia. J Clin Invest 1990;86:2025.

31. Sunthornthepvarakul T, Angkeow P, Weiss RE, et al. An identical missense mutation in the albumin gene results in familial dysalbuminemic hyperthyroxinemia in 8 unrelated families. Biochem Biophys Res Commun 1994;202:781.

32. Sunthornthepvarakul T, Likitmaskul S, Ngowngarmratana S, et al. Familial dysalbuminemic hypertriiodothyroninemia: a new, dominantly inherited albumin defect. J Clin Endocrinol Metab 1998;83:1448.

33. Friesema EC, Grueters A, Bierbermann H, et al. Severe X-linked psychomotor retardation caused by mutations in a thyroid hormone transporter. Lancet 2004 (in press).

33a. Dumitrescu AM, Liao XH, Best TB, et al. A novel syndrome combining thyroid and neurological abnormalities is associated with mutations in a monocarboxylate transporter gene. Am J Hum Genet 2004:74:168.

34. Weiss RE, Refetoff S. Resistance to thyroid hormone. Reviews in Endocrine and Metabolic Disorders 2000;1:97.

35. Mulligan LM, Kwok JB, Healey CS, et al. Germ-line mutations of the RET proto-oncogene in multiple endocrine neoplasia type 2A. Nature 1993;363:458.

36. Eng C, Clayton D, Schuffenecker I, et al. The relationship between specific RET proto-oncogene mutations and disease phenotype in multiple endocrine neoplasia type 2. JAMA 1996;276:1575.

37. Eng C, Smith DP, Mulligan LM, et al. Point mutation within the tyrosine kinase domain of the RET proto-oncogene in multiple endocrine neoplasia type 2B and related sporadic tumours. Hum Mol Genet 1994;3:237.

38. Marsh DJ, Coulon V, Lunetta KL, et al. Mutation spectrum and genotype-phenotype analyses in Cowden disease and Bannayan-Zonana syndrome, two hamartoma syndromes with germline PTEN mutation. Hum Mol Genet 1998;7:507.

39. Ishikawa Y, Sugano H, Matsumoto T, et al. Unusual features of thyroid carcinomas in Japanese patients with Werner syndrome and possible genotype-phenotype relations to cell type and race. Cancer 1999;85:1345.

40. Bell B, Mazzaferri EL. Familial adenomatous polyposis (Gardner's syndrome) and thyroid carcinoma. A case report and review of the literature. Dig Dis Sci 1993;38:185.

41. Kirschner LS, Carney JA, Pack SD, et al. Mutations of the gene encoding the protein kinase A type I-alpha regulatory subunit in patients with the Carney complex. Nat Genet 2000;26:89.

42. Perheentupa J. APS-I/APECED: the clinical disease and therapy. Endocrinol Metab Clin North Am 2002;31:295.

43. Bjorses P, Aaltonen J, Horelli-Kuitunen N, et al. Gene defect behind APECED: a new clue to autoimmunity. Hum Mol Genet 1998;7:1547.

44. Wildin RS, Smyk-Pearson S, Filipovich AH. Clinical and molecular features of the immunodysregulation, polyendocrinopathy, enteropathy, X linked (IPEX) syndrome. J Med Genet 2002;39:537.

45. Gahl WA, Thoene JG, Schneider JA. Cystinosis. N Engl J Med 2002;347:111.

46. Dias Da Silva MR, Cerutti JM, Arnaldi LA, et al. A mutation in the KCNE3 potassium channel gene is associated with susceptibility to thyrotoxic hypokalemic periodic paralysis. J Clin Endocrinol Metab 2002;87:4881.

47. Karlsson B, Gustafsson J, Hedov G, et al. Thyroid dysfunction in Down's syndrome: relation to age and thyroid autoimmunity. Arch Dis Child 1998;79:242.

48. Friedman DL, Kastner T, Pond WS, et al. Thyroid dysfunction in individuals with Down syndrome. Arch Intern Med 1989; 149:1990.

49. Murdoch JC, Ratcliffe WA, McLarty DG, et al. Thyroid function in adults with Down's syndrome. J Clin Endocrinol Metab 1977;44:453.

50. Medeiros CC, Marini SH, Baptista MT, et al. Turner's syndrome and thyroid disease: a transverse study of pediatric patients in Brazil. J Pediatr Endocrinol Metab 2000;13:357.

51. Elsheikh M, Wass JA, Conway GS. Autoimmune thyroid syndrome in women with Turner's syndrome—the association with karyotype. Clin Endocrinol (Oxf) 2001;55:223.

52. Radetti G, Mazzanti L, Paganini C, et al. Frequency, clinical and laboratory features of thyroiditis in girls with Turner's syndrome. Acta Paediatr 1995;84:909.

53. de Kerdanet M, Lucas J, Lemee F, et al. Turner's syndrome with X-isochromosome and Hashimoto's thyroiditis. Clin Endocrinol (Oxf) 1994;41:673.

54. Sparkes RS, Motulsky AG. The Turner syndrome with isochromosome X and Hashimoto's thyroiditis. Ann Intern Med 1967; 67:132.

55. Kawame H, Adachi M, Tachibana K, et al. Graves' disease in patients with 22q11.2 deletion. J Pediatr 2001;139:892.

56. Adachi M, Tachibana K, Masuno M, et al. Clinical characteristics of children with hypoparathyroidism due to 22q11.2 microdeletion. Eur J Pediatr 1998;157:34.

57. Kawamura T, Nimura I, Hanafusa M, et al. DiGeorge syndrome with Graves' disease: a case report. Endocr J 2000;47:91.

58. Heilstedt HA, Ballif BC, Howard LA, et al. Physical map of 1p36, placement of breakpoints in monosomy 1p36, and clinical characterization of the syndrome. Am J Hum Genet 2003; 72:1200.

59. Greenberg F, Lewis RA, Potocki L, et al. Multi-disciplinary clinical study of Smith-Magenis syndrome (deletion 17p11.2). Am J Med Genet 1996;62:247.

60. Slager RE, Newton TL, Vlangos CN, et al. Mutations in RAI1 associated with Smith-Magenis syndrome. Nat Genet 2003;33: 466.

61. Lander ES, Schork NJ. Genetic dissection of complex traits. Science 1994;265:2037.

62. Hegedüs L, Bonnema SJ, Bennedbæk FN. Management of simple nodular goiter: current status and future perspectives. Endocr Rev 2003;24:102.

63. Tunbridge WM, Evered DC, Hall R, et al. The spectrum of thyroid disease in a community; the Whickham survey. Clin Endocrinol (Oxf) 1977;7:481.

64. Knudsen N, Perrild H, Christiansen E, et al. Thyroid structure and size and two year follow up of solitary cold thyroid nodules in an unselected population with borderline iodine deficiency. Eur J Endocrinol 2000;142:224.

65. Ron E, Kleinerman RA, Boice JD, et al. A population-based case-control study of thyroid cancer. J Nat Cancer Inst 1987;79:1.

66. Preston-Martin S, Bernstein L, Pike MC, et al. Thyroid cancer among young women related to prior thyroid disease and pregnancy history. Br J Cancer 1987;55:191.

67. Lesueur F, Stark M, Tocco T, et al. Genetic heterogeneity in familial nonmedullary thyroid carcinoma: exclusion of linkage to RET, MNG1, and TCO in 56 families. J Clin Endocrinol Metab 1999;84:2157.

68. Harrer P, Broeker M, Zint A, et al. Thyroid nodules in recurrent multinodular goiters are predominantly polyclonal. J Endocrinol Invest 1998;21:380.

69. Hadjidakis SG, Koutras DA, Daikos GK. Endemic goitre in Greece: family studies. J Med Genet 1964;1:82.

70. Malamos B, Koutras DA, Kostamis P, et al. Endemic goitre in Greece: a study of 379 twin pairs. J Med Genet 1967;4:16.

71. Greig WR, Boyle JA, Duncan A, et al. Genetic and non-genetic factors in simple goitre formation: evidence from a twin study. QJM 1967;36:175.

72. Brix TH, Kyvik KO, Hegedus L. Major role of genes in the etiology of simple goiter in females: a population-based twin study. J Clin Endocrinol Metab 1999;84:3071.

73. González-Sarmiento R, Corral J, Mories MT, et al. Monoallelic deletion in the 5′ region of the thyroglobulin gene as a cause of sporadic nonendemic simple goiter. Thyroid 2001;8:789.

74. Bignell GR, Canzian F, Shayeghi M, et al. Familial nontoxic multinodular goiter locus maps to chromosome 14q but does not account for familial nonmedullary thyroid cancer. Am J Hum Genet 1997;61:1123.

75. Capon F, Tacconelli A, Giardina E, et al. Mapping a dominant form of multinodular goiter to chromosome Xp22. Am J Hum Genet 2000;67:1004.

76. Takahashi T, Nozaki J, Komatsu M, et al. A new locus for a dominant form of multinodular goiter on 3q26.1-q26.3. Biochem Biophys Res Commun 2001;284:650.

77. Neumann S, Willgerodt H, Ackermann F, et al. Linkage of familial euthyroid goiter to the multinodular goiter-1 locus and exclusion of the candidate genes thyroglobulin, thyroperoxidase and Na/I symporter. J Clin Endocrinol Metab 1999;84: 3750.

78. Malchoff CD, Sarfarazi M, Tendler B, et al. Papillary thyroid carcinoma associated with papillary renal neoplasia: genetic linkage analysis of a distinct heritable tumor syndrome. J Clin Endocrinol Metab 2000;85:1758.

79. Canzian F, Amati P, Harach R, et al. A gene predisposing to familial thyroid tumors with cell oxyphilia maps to chromosome 19p13.2. Am J Hum Genet 1998;63:1743.

80. McKay JD, Lesueur F, Jonard L, et al. Localization of a susceptibility gene for familial nonmedullary thyroid carcinoma to chromosome 2q21. Am J Hum Genet 2001;69:440.

81. Vestergaard P. Smoking and thyroid disorders—a meta-analysis. Eur J Endocrinol 2002;146:153.

82. Winsa B, Adami HO, Bergstrom R, et al. Stressful life events and Graves' disease. Lancet 1991;338:1475.

83. Sonino N, Girelli ME, Boscaro M, et al. Life events in the pathogenesis of Graves' disease. A controlled study. Acta Endocrinol 1993;128:293.

84. Okayasu I, Kong YM, Rose NR. Effect of castration and sex hormones on experimental autoimmune thyroiditis. Clin Immunol Immunopathol 1981;20:240.

85. Brix TH, Kyvik KO, Hegedus L. What is the evidence of genetic factors in the etiology of Graves' disease? A brief review. Thyroid 1998;8:627.

86. Brix TH, Kyvik KO, Christensen K, et al. Evidence for a major role of heredity in Graves' disease: a population-based study of two Danish twin cohorts. J Clin Endocrinol Metab 2001;86:930.

87. Brix TH, Kyvik KO, Hegedus L. A population-based study of chronic autoimmune hypothyroidism in Danish twins. J Clin Endocrinol Metab 2000;85:536.

88. Stenszky V, Kozma L, Balazs C, et al. The genetics of Graves' disease: HLA and disease susceptibility. J Clin Endocrinol Metab 1985;61:735.

89. Vaidya B, Kendall-Taylor P, Pearce SHS. The genetics of autoimmune thyroid disease. J Clin Endocrinol Metab 2002; 87: 5385.

90. Wandstrat A, Wakeland E. The genetics of complex autoimmune diseases: non-MHC susceptibility genes. Nature Immunol 2001;2:802.

91. Roitt IM, Doniach D, Campbell PN, et al. Auto-antibodies in Hashimoto's disease (lymphadenoid goitre). Lancet 1956;ii:820.

92. Hall R, Owen SG, Smart GA. Evidence for a genetic predisposition to formation of thyroid autoantibodies. Lancet 1960;ii: 187.

93. Phillips D, McLachlan S, Stephenson A, et al. Autosomal dominant transmission of autoantibodies to thyroglobulin and thyroid peroxidase. J Clin Endocrinol Metab 1990;70:742.

94. Jaume JC, Guo J, Pauls DL, et al. Evidence for genetic transmission of thyroid peroxidase autoantibody epitopic fragments. J Clin Endocrinol Metab 1999;84:1424.

95. Phillips D, Prentice L, Upadhyaya M, et al. Autosomal dominant inheritance of autoantibodies to thyroid peroxidase and thyroglobulin—studies in families not selected for autoimmune thyroid disease. J Clin Endocrinol Metab 1991;72:973.

96. Heward J, Gough SCL. Genetic susceptibility to the development of autoimmune disease. Clin Sci (London)1997;93:479.

97. Payami H, Joe S, Thomson G. Autoimmune thyroid disease in type 1 diabetic families. Genet Epidemiol 1989;6:137.

98. Torfs CP, King MC, Huey B, et al. Genetic interrelationship between insulin-dependent diabetes mellitus, the autoimmune thyroid diseases, and rheumatoid arthritis. Am J Hum Genet 1986;38:170.

99. Walker DJ, Griffiths M, Griffiths ID. Occurrence of autoimmune diseases and autoantibodies in multicase rheumatoid arthritis families. Ann Rheum Dis 1986;45:323.

100. Grennan DM, Dyer PA, Clague R, et al. Family studies in RA—the importance of HLA-DR4 and of genes for autoimmune thyroid disease. J Rheumatol 1983;10:584.

101. Nerup J. Addison's disease: clinical studies a report of 108 cases. Acta Endocrinol (Copenh) 1974;76:127.

102. Dittmar M, Kahaly GJ. Polyglandular autoimmune syndromes: immunogenetics and long-term follow-up. J Clin Endocrinol Metab 2003;88:2983.

103. Broadley SA, Deans J, Sawcer SJ, et al. Autoimmune disease in first-degree relatives of patients with multiple sclerosis. Brain 2000;123:1102.

104. Atkinson TP, Schäffer AA, Grimbacher B, et al. An immune defect causing dominant chronic mucocutaneous candidiasis and thyroid disease maps to chromosome 2p in a single family. Am J Hum Genet 2001;69:791.

105. Spielman RS, McGinnis RE, Ewens WJ. Transmission test for linkage disequilibrium: the insulin gene region and insulin-dependent diabetes mellitus (IDDM). Am J Hum Genet 1993; 52:506.

106. Davies JL, Kawaguchi Y, Bennett ST, et al. A genome-wide search for human type 1 diabetes susceptibility genes. Nature 1994;371:130.

107. Cox NJ, Wapelhorst B, Morrison VA, et al. Seven regions of the genome show evidence of linkage to type 1 diabetes in a consensus analysis of 767 multiplex families. Am J Hum Genet 2001;69:820.

108. Cornelis F, Faure S, Martinez M, et al. New susceptibility locus for rheumatoid arthritis suggested by a genome-wide linkage study. Proc Natl Acad Sci USA 1998;95:10746.

109. Shai R, Quismorio FP, Li L, et al. Genome-wide screen for systemic lupus erythematosus susceptibility genes in multiplex families. Hum Mol Genet 1999;8:639.

110. Grumet FC, Payne RO, Konishi J, et al. HL-A antigens as markers for disease susceptibility and autoimmunity in Graves' disease. J Clin Endocrinol Metab 1974;39:1115.

111. Seignalet J, Mirouze J, Jaffiol C, et al. HLA in Graves' disease and in diabetes mellitus insulin-dependent. Tissue Antigens 1975;6:272.

112. Bech K, Lumholtz B, Nerup J, et al. HLA antigens in Graves' disease. Acta Endocrinol (Copenh) 1977;86:510.

113. Farid NR, Sampson L, Noel EP, et al. A study of human leukocyte D locus related antigens in Graves' disease. J Clin Invest 1979;63:108.

114. Tomer Y, Davies TF. The genetic susceptibility to Graves' disease. Baillieres Clin Endocrinol Metab 1997;11:431.

115. Heward JM, Allahabadia A, Daykin J, et al. Linkage disequilibrium between the human leukocyte antigen class II region of the major histocompatibility complex and Graves' disease: replication using a population case control and family-based study. J Clin Endocrinol Metab 1998;83:3394.

116. Yanagawa T, Mangklabruks A, Chang Y, et al. Human histocompatability leukocyte antigen-DQA1*0501 allele associated with genetic susceptibility to Graves' disease in a Caucasian population. J Clin Endocrinol Metab 1993;76:1569.

117. Barlow AB, Wheatcroft N, Watson P, et al. Association of HLA-DQA1*0501 with Graves' disease in English Caucasian men and women. Clin Endocrinol (Oxf) 1996;44:73.

118. Maciel LM, Rodrigues SS, Dibbern RS, et al. Association of the HLA-DRB1*0301 and HLA-DQA1*0501 alleles with Graves' disease in a population representing the gene contribution from several ethnic backgrounds. Thyroid 2001;11:31.

119. Cuddihy RM, Bahn RS. Lack of an independent association between the human leukocyte antigen allele DQA1*0501 and Graves' disease. J Clin Endocrinol Metab 1996;81:847.

120. Philippou G, Krimitzas A, Kaltsas G, et al. HLA DQA1*0501 and DRB1*0301 antigens do not independently convey susceptibility to Graves' disease. J Endocrinol Invest 2001;24:88.

121. Hunt PJ, Marshall SE, Weetman AP, et al. Histocompatibility leucocyte antigens and closely linked immunomodulatory genes in autoimmune thyroid disease. Clin Endocrinol (Oxf) 2001;55:491.

122. Kawa A, Nakamura S, Nakazawa M, et al. HLA-BW35 and B5 in Japanese patients with Graves' disease. Acta Endocrinol (Copenh) 1977; 86:754.

123. Dong RP, Kimura A, Okubo R, et al. HLA-A and DPB1 loci confer susceptibility to Graves' disease. Hum Immunol 1992; 35:165.

124. Onuma H, Ota M, Sugenoya A, et al. Association of HLA-DPB1*0501 with early-onset Graves' disease in Japanese. Hum Immunol 1994;39:195.

125. Tandon N, Mehra NK, Taneja V, et al. HLA antigens in Asian Indian patients with Graves' disease. Clin Endocrinol (Oxf) 1990;33:21.

126. Omar MA, Hammond MG, Desai RK, et al. HLA class I and II antigens in South African blacks with Graves' disease. Clin Immunol Immunopathol 1990;54:98.

127. Chen QY, Nadell D, Zhang XY, et al. The human leukocyte antigen HLA DRB3*020/DQA1*0501 haplotype is associated with Graves' disease in African Americans. J Clin Endocrinol Metab 2000;85:1545.

128. Wong GW, Cheng SH, Dorman JS. The HLA-DQ associations with Graves' disease in Chinese children. Clin Endocrinol (Oxf) 1999;50:493.

129. Ratanachaiyavong S, Demaine AG, Campbell RD, et al. Heat shock protein 70 (HSP70) and complement C4 genotypes in patients with hyperthyroid Graves' disease. Clin Exp Immunol 1991;84:48.

130. Badenhoop K, Schwarz G, Schleusener H, et al. Tumor necrosis factor beta gene polymorphisms in Graves' disease. J Clin Endocrinol Metab 1992;74:287.

131. Heward JM, Allahabadia A, Sheppard MC, et al. Association of the large multifunctional proteasome (LMP2) gene with Graves' disease is a result of linkage disequilibrium with the HLA haplotype DRB1*0304-DQB1*02-DQA1*0501. Clin Endocrinol (Oxf) 1999;51:115.

132. Kamizono S, Hiromatsu Y, Seki N, et al. A polymorphism of the 5′ flanking region of tumour necrosis factor alpha gene is associated with thyroid-associated ophthalmopathy in Japanese. Clin Endocrinol (Oxf) 2000;52:759.

133. Moens H, Farid NR, Sampson L, et al. Hashimoto's thyroiditis is associated with HLA-DRw3. N Engl J Med 1978;299:133.

134. Tandon N, Zhang L, Weetman AP. HLA associations with Hashimoto's thyroiditis. Clin Endocrinol (Oxf) 1991;34:383.

135. Thompson C, Farid NR. Post-partum thyroiditis and goitrous (Hashimoto's) thyroiditis are associated with HLA-DR4. Immunol Lett 1985;11:301.

136. Bogner U, Badenhoop K, Peters H, et al. HLA-DR/DQ gene variation in nongoitrous autoimmune thyroiditis at the serological and molecular level. Autoimmunity 1992;14:155.

137. Badenhoop K, Schwarz G, Walfish PG, et al. Susceptibility to thyroid autoimmune disease: molecular analysis of HLA-D region genes identifies new markers for goitrous Hashimoto's thyroiditis. J Clin Endocrinol Metab 1990;71:1131.

138. Jansson R, Safwenberg J, Dahlberg PA. Influence of the HLA-DR4 antigen and iodine status on the development of autoimmune postpartum thyroiditis. J Clin Endocrinol Metab 1985; 60:168.

139. Vargas MT, Briones-Urbina R, Gladman D, et al. Antithyroid microsomal autoantibodies and HLA-DR5 are associated with postpartum thyroid dysfunction: evidence supporting an autoimmune pathogenesis. J Clin Endocrinol Metab 1988;67: 327.

140. Parkes AB, Darke C, Othman S, et al. Major histocompatibility complex class II and complement polymorphisms in postpartum thyroiditis. Eur J Endocrinol 1996;134:449.

141. Uno H, Sasazuki T, Tamai H, et al. Two major genes, linked to HLA and Gm, control susceptibility to Graves' disease. Nature 1981;292:768.

142. Vaidya B, Imrie H, Perros P, et al. The cytotoxic T lymphocyte antigen-4 is a major Graves' disease locus. Hum Mol Genet 1999;8:1195.

143. Tomer Y, Barbesino G, Greenberg DA, et al. Mapping the major susceptibility loci for familial Graves' and Hashimoto's diseases: evidence for genetic heterogeneity and gene interactions. J Clin Endocrinol Metab 1999;84:4656.

144. Sakai K, Shirasawa S, Ishikawa N, et al. Identification of susceptibility loci for autoimmune thyroid disease to 5q31-q33 and Hashimoto's thyroiditis to 8q23-q24 by multipoint affected sib-pair linkage analysis in Japanese. Hum Mol Genet 2001;10:1379.

145. Jin Y, Teng W, Ben S, et al. Genome-wide scan of Graves' disease: evidence for linkage on chromosome 5q31 in Chinese Han pedigrees. J Clin Endocrinol Metab 2003;88:1798.

146. Zavattari P, Lampis R, Motzo C, et al. Conditional linkage disequilibrium analysis of a complex disease superlocus, IDDM1 in the HLA region, reveals the presence of independent modifying gene effects influencing the type 1 diabetes risk encoded by the major HLA-DQB1, -DRB1 disease loci. Hum Mol Genet 2001;10:881.

147. Johansson S, Lie BA, Todd JA, et al. Evidence of at least two type 1 diabetes susceptibility genes in the HLA complex distinct from HLA-DQB1, -DQA1 and -DRB1. Genes Immun 2003;4:46.

148. Waterhouse P, Penninger JM, Timms E, et al. Lymphoproliferative disorders with early lethality in mice deficient in Ctla-4. Science 1995;270:985.

149. Luhder F, Hoglund P, Allison JP, et al. Cytotoxic T lymphocyte-associated antigen 4 (CTLA-4) regulates the unfolding of autoimmune diabetes. J Exp Med 1998;187:427.

150. Lenschow DJ, Ho SC, Sattar H, et al. Differential effects of anti-B7–1 and anti-B7–2 monoclonal antibody treatment on the development of diabetes in the nonobese diabetic mouse. J Exp Med 1995;181:1145.

151. Krummel MF, Allison JP. CTLA-4 engagement inhibits IL-2 accumulation and cell cycle progression upon activation of resting T cells. J Exp Med 1996;183:2533.

152. Scheipers P, Reiser H. Fas-independent death of activated CD4(+)T lymphocytes induced by CTLA-4 crosslinking. Proc Natl Acad Sci U S A 1998;95:10083.

153. Yanagawa T, Hidaka Y, Guimaraes V, et al. CTLA-4 gene polymorphism associated with Graves' disease in a Caucasian population. J Clin Endocrinol Metab 1995;80:41.

154. Nisticò L, Buzzetti R, Pritchard LE, et al. The CTLA-4 gene region of chromosome 2q33 is linked to, and associated with, type 1 diabetes. Hum Mol Genet 1996;5:1075.

155. Hadj Kacem H, Bellassoued M, Bougacha-Elleuch N, et al. CTLA-4 gene polymorphisms in Tunisian patients with Graves' disease. Clin Immunol 2001;101:361.

156. Nithiyananthan R, Heward JM, Allahabadia A, et al. Polymorphism of the CTLA-4 gene is associated with autoimmune hypothyroidism in the United Kingdom. Thyroid 2002;12:3.

157. Barbesino G, Tomer Y, Concepcion E, et al. Linkage analysis of candidate genes in autoimmune thyroid disease. I. Selected immunoregulatory genes. J Clin Endocrinol Metab 1998;83:3290.

158. Tomer Y, Greenberg DA, Barbesino G, et al. CTLA-4 and not CD28 is a susceptibility gene for thyroid autoantibody production. J Clin Endocrinol Metab 2001;86:1687.

159. Ueda H, Howson JM, Esposito L, et al. Association of the T-cell regulatory gene CTLA4 with susceptibility to autoimmune disease. Nature 2003;423:506.

160. Risch N, Merikangas K. The future of genetic studies of complex human diseases. Science 1996;273:1516.

161. Vaidya B, Imrie H, Perros P, et al. Cytotoxic T lymphocyte antigen-4 (CTLA-4) gene polymorphism confers susceptibility to thyroid associated orbitopathy. Lancet 1999;354:743.

162. Allahabadia A, Heward JM, Nithiyananthan R, et al. MHC class II region, CTLA4 gene, and ophthalmopathy in patients with Graves' disease. Lancet 2001;358:984.

163. Stuck BJ, Pani MA, Besrour F, et al. Fas ligand gene polymorphisms are not associated with Hashimoto's thyroiditis and Graves' disease. Hum Immunol 2003;64:285.

164. Ban Y, Taniyama M, Ban Y. Vitamin D receptor gene polymorphism is associated with Graves' disease in the Japanese population. J Clin Endocrinol Metab 2000;85:4639.

165. Pani MA, Regulla K, Segni M, et al. A polymorphism within the vitamin D–binding protein gene is associated with Graves' disease but not with Hashimoto's thyroiditis. J Clin Endocrinol Metab 2002;87:2564.

166. Pearce SHS, Vaidya B, Imrie H, et al. Further evidence for a susceptibility locus on chromosome 20q13.11 in families with dominant transmission of Graves' disease. Am J Hum Genet 1999;65:1462.

167. Vaidya B, Imrie H, Perros P, et al. Evidence for a new Graves' disease susceptibility locus at chromosome 18q21. Am J Hum Genet 2000;66:1710.

168. Tomer Y, Ban Y, Concepcion E, et al. Common and unique susceptibility loci in Graves' and Hashimoto diseases: results of whole-genome screening in a data set of 102 multiplex families. Am J Hum Genet 2003;73:000.

169. Risch N. Linkage strategies for genetically complex traits. II. The power of affected relative pairs. Am J Hum Genet 1990;46:229.

170. Lander E, Kruglyak L. Genetic dissection of complex traits: guidelines for interpreting and reporting linkage results. Nat Genet 1995;11:241.

171. Akamizu T, Hiratani H, Ikegami S, et al. Association study of autoimmune thyroid disease at 5q23-q33 in Japanese patients. J Hum Genet 2003;48:236.

172. Ban Y, Greenberg DA, Concepcion E, et al. Amino acid substitutions in the thyroglobulin gene are associated with susceptibility to human and murine autoimmune thyroid disease. Proc Natl Acad Sci USA 2003;100:15119.

173. Gough S. The thyroglobulin gene: the third locus for autoimmune thyroid disease or a false dawn? Trends Molecular Medicine 2004;10:302.