| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Institut de Recherche Interdisciplinaire en Biologie Humaine et, Moléculaire, Université Libre de Bruxelles, Campus Erasme, B-1070 Bruxelles, Belgium
Address all correspondence and requests for reprints to: Gilbert Vassart or Jacques E. Dumont, Institut de Recherche Interdisciplinaire en Biologie Humaine et Moléculaire (IRIBHM), Université Libre de Bruxelles, Campus Erasme, 808 Route de Lennik, B-1070 Bruxelles, Belgium. E-mail: gvassart{at}ulb.ac.be or jedumont{at}ulb.ac.be.
With about 1/3200 newborn affected, congenital hypothyroidism is a significant public health issue that, in developed countries, has been addressed by efficient neonatal screening programs. In about 15% of the cases, the cause can be traced to a defect in one of the many genes implicated in thyroid hormone synthesis (thyroglobulin, thyroperoxidase, sodium-iodide symporter, pendrin, Thox2, iodotyrosine dehalogenase) (1). Dyshormonogeneses, as these situations are called, are transmitted as classical autosomal recessive mendelian traits. Thyroid dysgenesis, referring to developmental defects, encompasses the remaining 85% of the cases and can be further subdivided into agenesis, hypoplasia, ectopy, and hemiagenesis. Whereas about 5% of these cases have been shown to result from mutations in one of three thyroid developmental genes (TITF1, PAX8, FOXE1/TITF2) or the TSH receptor, the great majority present as sporadic (2). As such, thyroid dysgenesis represents one of the remaining enigma in the pathophysiology of thyroid diseases.
Over the past 20 yr, research conducted essentially by the group of Roberto Di Lauro (2) has lead to the identification of the role of these three developmental genes, both in the regulation of expression of thyroid-specific genes, and embryonic development of the thyroid anlagen. Knockout mice provided the first animal models of human thyroid dysgenesis. From the beginning, however, there have been difficulties in comparing the mode of transmission between mouse and man. For titf1, homozygous knockout mice are nonviable, with athyreosis and major central nervous system anomalies (3). In human, no homozygous case has been identified yet, but heterozygotes display minor thyroid problems, with variable respiratory problems and progressive development of extrapyramidal symptoms in relation with developmental anomalies of the brain (4, 5, 6). Interestingly, and contrary to what is reported in the paper of Amendola et al. in this issue (7), upon reinvestigation, the heterozygous titf1 knockout mice, initially considered wild type, demonstrated mild thyroid anomalies (slightly elevated TSH), together with limited trouble of motor coordination (4). For pax8, the mode of transmission is different: recessive in mice, with failure of follicular cell development in homozygotes (8), and dominant in human in whom heterozygotes display thyroid hypoplasia with variable penetrance and expressivity (9, 10, 11, 12, 13). For foxe1/titf2, there seems to be agreement between recessive transmission in mice (agenesis or ectopy) and the limited number of human cases (agenesis) (14, 15).
The variable penetrance and expressivity of heterozygous pax8 mutations in man pointed to the existence of modifier genes that, in a sense, is already indicative of a multigenic situation. In the current issue, Amendola et al. (7) present evidence that mice doubly heterozygous for pax8 and titf1 null mutations display interesting thyroid phenotypes reminiscent of some forms of thyroid dysgenesis (hypoplasia or hemiagenesis), provided the mutations are present on a specific genetic background (C57BL/6).When the same mutations are introduced in another background (129/Sv), the double heterozygotes are completely wild type. A simple backcross experiment allows the authors to propose that two modifier genes are likely at play to explain the strain differences. Future linkage studies will allow to locate these genes and hopefully to identify them. This leads the authors to suggest that a minimum of four genes, including titf1 and pax8, could be involved in the development of thyroid hypoplasia or hemiagenesis in these mice and, by extrapolation, possibly in humans.
Detailed analysis of the thyroid phenotype of the double heterozygous mice reveals that at least two different phenomena are likely involved. Affected animals display a reduced number of thyrocytes, either as a symmetrical hypoplasia, or in the form of hemiagenesis in about 30% of the cases. Demonstration that the defect is already present early in development indicates that single doses of TITF1 and PAX8, when present in association, are not enough to give rise to the normal complement of thyrocytes. Because in mice the size of the gland at birth is not affected by the absence of the TSH receptor (16), this aspect of the phenotype is clearly independent of the sensitivity of thyrocytes to TSH. However, some loss of sensitivity to the growth effects of TSH must be involved after birth because the thyroids of the double heterozygous mice remain hypoplastic despite being stimulated by increased TSH levels. Of interest, thyroid hemiagenesis in man is not necessarily associated with clinical or biological hypothyroidism (17), which suggests that growth compensation occurs and that different pathophysiological mechanisms may be involved in the experimental model and human disease. To complicate the picture, a monogenic cause of thyroid hemiagenesis has been identified in mice with homozygote invalidation of the sonic hedgehog gene (18) and in human familial cases with no PAX8 mutations (19).
The loss of sensitivity to TSH of double heterozygotes is not limited to growth effects. Amendola et al. (7) demonstrate a decrease in the production of thyroglobulin and propose that this is the immediate cause of hypothyroidism. Of interest, other effects of TSH are well preserved: thyrocytes show distinct signs of hypertrophy and display increased expression of sodium-iodide symporter.
In sum, the affected mice studied by Amendola et al. (7) demonstrates that association of haploinsufficiency for the titf1 and pax8 genes with C57BL/6 strain-specific alleles can mimic human forms of thyroid dysgenesis characterized by hypoplasia or hemiagenesis. As such, this leads the authors to propose that thyroid dysgenesis would be a multigenic condition, thus explaining its apparent sporadic character.
Appealing as it may sound, the situation is a bit more complicated. In their introduction, Amendola et al. (7) recall that the rule for monozygotic twins with thyroid dysgenesis is to be discordant (20). This simple experiment of nature implies that involvement of multiple genes is not enough to cause thyroid dysgenesis because monozygotic twins are expected to harbor identical copies of their genes, at all loci. However, it does not necessarily contradict the idea of multiple genes being involved. All it tells us is that either epigenetic mechanisms, or somatic mutations must be at play. Early somatic mutations with dominant effect in a gene important for thyroid development could theoretically explain discordance between twins. However, reminiscent of the situation in the McCune Albright syndrome, the extreme rarity of concordant twins or truly hereditary cases imply that the corresponding mutation must be lethal, if it were present in the germline. Although somatic mosaicism cannot be formally excluded, more likely in our opinion is the suggestion that stochastic epigenetic mechanisms causing variable degree of gene or allele silencing would be involved. According to this hypothesis, a certain amount of random noise would modulate quantitatively the expression of genes that are naturally expressed at low level (21). The term metastable epiallele has been proposed to describe the phenomenon (22). The result is that different cells of the same anlagen could express either the two alleles of a gene, or a single allele, or no allele at all. If such a phenomenon becomes epigenetically fixed in the somatic lineages of an organ bud, it may lead to defective development in the rare cases where low expression of one or more key genes would have been selected, by chance. Random silencing (or derepression) leading to monoallelic expression has been documented for a handful of genes (23, 24, 25, 26, 27) and may be related to this phenomenon, which would readily explain discordance between monozygotic twins for many developmental defects (28, 29, 30), including thyroid dysgenesis (20). Interestingly, there is evidence that monozygotic twinning, by itself, could be linked to epigenetic alterations (31).
By demonstrating that alteration of gene dosage at two loci, in a given genetic background, can cause thyroid hypoplasia or hemiagenesis, the study by Amendola et al. (7) opens the possibility that epigenetic mechanisms leading to stochastic variations of expression at multiple loci could be responsible for the sporadic character of thyroid dysgenesis. Experimental validation of this hypothesis is far from trivial. Indeed, depending on the type of dysgenesis, it would require studying gene expression or methylation in the hypoplastic, or ectopic tissue, or ... in the absent gland! Investigation of the signal transduction pathways involved in the switch from TSH-independent to TSH-dependent thyroid growth, during early life, could provide clues.
Received September 27, 2005.
Accepted for publication September 28, 2005.
 |
References |
|---|
 Top References |
|---|
This article has been cited by other articles:
 |
|
 |
|
  J. Deladoey, N. Belanger, and G. Van Vliet Random Variability in Congenital Hypothyroidism from Thyroid Dysgenesis over 16 Years in Quebec J. Clin. Endocrinol. Metab., August 1, 2007; 92(8): 3158 - 3161. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Dentice, V. Cordeddu, A. Rosica, A. M. Ferrara, L. Santarpia, D. Salvatore, L. Chiovato, A. Perri, L. Moschini, C. Fazzini, et al. Missense Mutation in the Transcription Factor NKX2-5: A Novel Molecular Event in the Pathogenesis of Thyroid Dysgenesis J. Clin. Endocrinol. Metab., April 1, 2006; 91(4): 1428 - 1433. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
