A Mouse Model Demonstrates a Multigenic Origin of Congenital Hypothyroidism
http://www.100md.com
《内分泌学杂志》
Stazione Zoologica A. Dohrn (E.A., P.D.L., A.A., R.D.L., M.D.F.), Laboratorio di Genetica Animale at CEINGE, 80145 Naples, Italy
Dipartimento di Endocrinologia e Oncologia Molecolare e Clinica (P.E.M.) and Dipartimento di Biologia e Patologia Cellulare e Molecolare (D.T., V.M., R.D.L.), Universita Federico II
Biogem at CEINGE (A.R.), Naples, Italy
Instituto Nazionale Tumori Fondazione Pascale (G.C., C.A.), 80131 Naples, Italy
Laboratory of Metabolism (S.K.), National Cancer Institute, National Institutes of Health, Bethesda, Maryland 20892
Max Planck Institute für Biophysikalische Chemie (A.M.), Abteilung Molekulare Zellbiologie, 37077 Gttingen, Germany
Abstract
Congenital hypothyroidism with thyroid dysgenesis (TD) is a frequent human condition characterized by elevated levels of TSH in response to reduced thyroid hormone levels. Congenital hypothyroidism is a genetically heterogeneous disease. In the majority of cases studied, no causative mutations have been identified and very often the disease does not show a Mendelian transmission. However, in approximately 5% of cases, it can be a consequence of mutations in genes encoding the TSH receptor or the transcription factors TITF1, FOXE1, or PAX8. We report here that in mouse models, the combination of partial deficiencies in the Titf1 and Pax8 genes results in an overt TD phenotype that is absent in either of the singly deficient, heterozygous mice. The disease is characterized by a small thyroid gland, elevated levels of TSH, reduced thyroglobulin biosynthesis, and high occurrence of hemiagenesis. The observed phenotype is strain specific, and the pattern of transmission indicates that at least two other genes, in addition to Titf1 and Pax8, are necessary to generate the condition. These results show that TD can be of multigenic origin in mice and strongly suggest that a similar pathogenic mechanism may be observed in humans.
Introduction
CONGENITAL HYPOTHYROIDISM with thyroid dysgenesis (TD) (1) is a frequent human condition, the genetic basis of which has just begun to be described (2, 3), even though TD has been found discordant in 12 of 13 monozygotic twins studied (4). Disease-causing mutations were identified in the receptor for TSH (TSHR) (5, 6) and in the transcription factors TITF1 (7, 8, 9, 10), PAX8 (11, 12, 13, 14, 15), and FOXE1/TITF2 (16, 17). However, only 5% of the patients with TD examined presented mutations in these genes. This number could be an underestimate because the mutation analysis was limited to the coding regions of the genes examined, thus excluding potential disease-causing mutations in noncoding regions that might affect regulatory regions or result in aberrant splicing. Still, the low frequency of mutation in the genes studied thus far suggests that other genes (18) and/or other mechanisms could be involved in the pathogenesis of TD. In this study, we addressed whether TD could be a multigenic disorder. Such a pathogenetic mechanism would explain why most TD patients do not display a clear Mendelian transmission and would be in agreement with the observation that there is a higher incidence of thyroid condition in those families with at least one case of TD (19, 20, 21). Furthermore, the incomplete penetrance and the variable expressivity observed in familial cases of TD (11, 13) is also consistent with a multigenic mechanism.
To provide proof for a possible multigenic origin of TD, we used mouse models. The null mutations generated in the genes encoding the transcription factors Titf1 (23) and Pax8 (24) only display a TD phenotype when homozygous. In neither case do heterozygous animals show an overt TD phenotype. We report here that mice heterozygous for both Titf1- and Pax8-null alleles, denominated DHTP (double heterozygous for both Titf1- and Pax8-null mutations), present TD. An in-depth study of the phenotype reveals elevated TSH levels in DHTP mice associated with reduced thyroid hormone secretion and decreased thyroglobulin (Tg) biosynthesis. Furthermore, we show that the thyroid gland of DHTP mice is smaller during embryonic life, thus making these mice a bona fide TD model. Interestingly, adult DHTP mice also show a high incidence of thyroid hemiagenesis, suggesting also that this condition can be genetically determined.
Finally, we demonstrate that the observed TD condition is strain specific. TD is displayed when the DHTP mutations are in the C57BL/6 genetic background (DHTP/B6), and it is not observed in the 129/Sv strain (DHTP/Sv) or in the F1 generation of a C57BL/6 x 129/Sv cross (DHTP/B6/Sv). Furthermore, the pattern of transmission of the TD phenotype in the backcross DHTP/B6/Sv x C57BL/6 suggests that at least two additional C57BL/6-specific alleles, in addition to the null mutations in Titf1 and Pax8, are responsible for the emergence of TD.
This study is the first demonstration that TD can be of multigenic origin. The mouse model presented here is useful to begin understanding the molecular mechanism underlying TD in those patients for whom the genetic basis is unknown and might lead to the identification of additional genes required for proper thyroid gland organogenesis and efficient thyroid hormone biosynthesis.
Materials and Methods
Mice and breeding
Heterozygous Titf1-null mice (Titf1+/–) (23) and heterozygous Pax8-null mice (Pax8+/–) (24), originally generated in a mixed genetic background, were crossed to generate double heterozygote for both Titf1- and Pax8-null mutations (DHTP) mice. DHTP mice were backcrossed 10 generations in a C57BL/6J background (Charles River Laboratories, Calco, Italy) (Fig. 1). The DHTP progeny were estimated to have more than 99% of their genes derived from the C57BL/6J strain. Almost congenic DHTP mice were mated with 129/SvPasCrl wild-type (Charles River Laboratories) to generate F1 hybrid DHTP/B6/Sv mice.
Both male and female mice were used for breeding. Animals were housed in a conventional facility with a 12-h light, 12-h dark cycle and were supplied with standard rodent food. Genotypes were determined by PCR using genomic DNA isolated from the tail snips as described (25).
Histology and immunohistochemistry
Staged mouse embryos were obtained by dissection of pregnant females. The day in which the vaginal plug was detected was designated as embryonic d 0.5 (E0.5). Animals were killed by cervical dislocation. Adult thyroid glands were dissected from 3-month-old mice. Thyroids and embryos were fixed overnight at 4 C in 4% paraformaldehyde in PBS at pH 7.2, dehydrated through ethanol series, cleared in xylene, and embedded in paraffin, and 7-μm sections were cut.
For histological examinations, serial sections from dissected DHTP and wild-type mice were stained with Harry’s hematoxylin and eosin (BDH Laboratory Supplies, Poole, UK), according to the manufacturer’s instructions. For immunohistochemical analyses, sections from embryos and dissected thyroids were dewaxed by standard techniques. Heat treatment was performed to retrieve the antigen sites. Staining procedures and chromogenic reactions were carried out according to the Vectastain ABC kit protocol (Vector Laboratories, Burlingame, CA). The primary antibodies used antimouse Pax8 and antirat sodium/iodide symporter (NIS) have been described previously (26).
Hormone measurements
Blood samples were obtained from the tail vein, under light anesthesia, and collected in microtubes without anticoagulant. After coat formation, the samples were centrifuged, and the recovered serum was kept frozen at –20 C until assayed.
TSH levels were determined by specific RIA kit (rat TSH kit) from Amersham Biosciences Europe (Freiburg, Germany). Total T4 was measured with the Immulite analyzer using a commercial kit, as recommended by the manufacturer (Diagnostics Products Corp., Los Angeles, CA).
Real-time PCR
Total RNA was extracted using the guanidine isothiocyanate method (27) and further purified by the RNAeasy minikit (Qiagen, Hilden, Germany). Four micrograms of total RNA were used as template for the synthesis of the first-strand cDNA reverse starting from oligo dT primers using the Superscript II Reverse Transcriptase kit (Invitrogen Life Technologies, Carlsbad, CA) according to the instructions of the manufacturer.
Reactions for the quantification of mRNAs were performed in an ABI Prism 7300 Real Time PCR System (Applied Biosystems, Foster City, CA) using SYBR green as detector dye. The reaction mixtures contained 25 μl of SYBR Green PCR master mix (Applied Biosystems), primers at a concentration of 300 nM each, 10-ng template cDNA in a final volume of 50 μl. Reactions were carried out in triplicate.
Specific primer sets for each gene were designed using the program Primer Express (Applied Biosystems). The sequences of the primers used in the reactions were as follows: 1), Thyroglobulin: forward, 5'-CAT GGA ATC TAA TGC CAA GAA CTG-3'; reverse, 5'-TCC CTG TGA GCT TTT GGA ATG-3' . 2), Thyroid peroxidase: forward, 5'-CAA AGG CTG GAA CCC TAA TTT CT-3'; reverse, 5'-AAC TTG AAT GAG GTG CCT TGT CA-3'. 3) TSH receptor: forward, 5'-TCC CTG AAA ACG CAT TCC A-3'; reverse, 5'-GCA TCC AGC TTT GTT CCA TTG-3'. 4) Iodothyronine deiodinase I: forward, 5'-CCT CCA CAG CCG ATT TCC T-3'; reverse, 5'-GTT CTT CTT AAA AGC CCA GCC A-3'. 5) Abelson: forward, 5'-TCGGACGTGTGGGCATTT-3'; reverse, 5'-CGCATGAGCTCGTAGACCTTC-3'.
For each sample, the expression of the genes of interest was normalized for the expression of Abelson gene measured under the same condition. For each genotype, the data obtained represent the mean of three independent samples, each composed of the thyroids dissected from four E18 embryos.
RNA microarray
The thyroid glands were dissected from E18 embryos, and total RNA was extracted with guanidine isothiocyanate (27) and further purified by the RNeasy minikit (Qiagen). cRNA was generated by using the Affymetrix One-Cycle Target Labeling and Control Reagent kit (Affymetrix Inc., Santa Clara, CA) following the protocol of the manufacturer. The biotinylated cRNA was hybridized to the MOE430A Affymetrix DNA chips, containing over 22,000 genes and open-reading frames from Mus musculus Genome databases GenBank, dbEST, and RefSeq. Chips were washed and scanned on the Affymetrix Complete GeneChip Instrument System, generating digitized image data files. Reactions were carried out in triplicate.
Digitized image data files were analyzed by Micro Array Suite 5.0 for detection calls (Affymetrix Inc.) and Robust Multichip Average for expression values. The expression values obtained were analyzed using GeneSpring 7.1 (Silicon Genetics, Redwood City, CA). Results were filtered for flag (presence call) and then for fold change >1.5, obtaining a total of 445 probe sets differentially expressed in the B6 strain and 55 for the B6/Sv-F1. Statistical analysis was performed using the Welch t test, with P 0.05 as the cutoff value. The P values were adjusted by the Benjamini and Hochberg correction (28). Genes were classified by Gene Ontology for Biological Process and Molecular function. Hierarchical clustering were performed on gene lists by the Gene Tree algorithm using the Pearson correlation.
Results
Combination of recessive mutations in Titf1 and Pax8 genes induces an elevation in serum of TSH levels
Heterozygous null mice for either Titf1 (Titf1+/–) or Pax8 (Pax8+/–) were bred to generate double heterozygous mice (Titf1+/–, Pax8+/–), herein referred to as DHTP mice (Fig. 1). The analysis of genotype frequencies in the progenies of this cross revealed the expected proportion (25%) of DHTP mice. TSH serum levels were measured in 90-d-old mice to test for thyroid function. TSH levels in the single heterozygous mice resulted comparable to those of wild type. TSH measurements in DHTP mice showed exceedingly abnormal levels of this hormone, albeit in only a fraction of DHTP (Fig. 2A), suggesting impaired thyroid functionality.
Given that both the Titf1+/– and the Pax8+/– parents had a mixed genetic background, with contribution from both C57BL/6J (herein referred to as B6) and 129/SvPasCrl (herein referred to as Sv) strains, the incomplete penetrance of the high TSH phenotype observed in DHTP mice could be the consequence of the diverse genetic background among the mice examined. To test this hypothesis, DHTP mice were backcrossed 10 generations into the B6 strain to obtain both single heterozygous (Titf1+/– or Pax8+/–) and DHTP in a practically congenic B6 genetic background. The TSH levels at d 90 in these mice showed that serum TSH were invariably more elevated in DHTP than in wild-type and single heterozygous littermates, averaging an 8-fold increase (Fig. 2B). Backcrossing the DHTP mice on the Sv strain resulted in an F1 with DHTP mice presenting normal TSH levels (Fig. 2C). Here, these mice are referred to as DHTP/B6/Sv-F1.
These data demonstrate that the partial deficiency of Titf1 and Pax8 when associated with B6-specific allele(s) results in hyperthyrotropinemia when associated with B6-specific allele(s); the data also show that these B6 alleles are recessive to those in the Sv strain. Next, we investigated the thyroid function of DHTP mice in the B6 genetic background (herein called DHTP/B6). For the remainder of this study, the phenotypes of DHTP/B6 mice were compared with either that of wild-type littermates or of DHTP/B6/Sv-F1.
DHTP/B6 mice are hypothyroid
T4 levels measured in 90-d-old DHTP/B6 mice indicated an evident hypothyroidism. DHTP/B6 mice presented low T4 levels compared with their single heterozygous or wild-type littermates (Fig. 3A). Furthermore, in agreement with a reduced thyroid function, a significant reduction in body weight was observed in the DHTP/B6 (Fig. 3B). These data strongly suggest that a reduced secretion of T4 is the primary defect in DHTP/B6 mice and that the increased TSH levels and reduced body weight are secondary effects of thyroid hormone deficiency.
No significant difference in T4 serum levels (Fig. 3C) or body weight (Fig. 3D) was observed among wild-type, single heterozygous, and DHTP in the B6/Sv-F1 genetic background. The histological examination of the DHTP/B6 thyroids showed zones of irregular and diffuse hyperplasia, with follicles lined by tall columnar cells, typical of a chronic TSH stimulation (Fig. 4A), whereas wild-type thyroid follicles consisted of a rim of flattened or cuboidal cells (Fig. 4B). Moreover, immunochemistry assays revealed an increased expression of NIS in DHTP/B6 follicular cells, a sensitive parameter of TSH stimulation (Fig. 4, C vs. D). No morphological alterations were detected in DHTP/B6/Sv-F1 thyroids (data not shown).
Surprisingly, the thyroid gland was not enlarged in 90-d-old DHTP mice, despite an elevated serum TSH concentration. A more detailed analysis revealed that the thyroid of DHTP/B6 thyroid was smaller than the thyroid of control mice (Fig. 4, E vs. F), determined by evaluating the diameters of serial sections of glands (Fig. 4, G vs. H).
An additional remarkable feature of the DHTP/B6 mice was the hemiagenesis of the thyroid observed in about 30% of the animals examined. In these animals, the gland consisted of a single lobe, slightly hypertrophic, correctly located but on only one side of the trachea (Fig. 4I). Thyroid hemiagenesis was never observed in approximately 100 control mice.
Congenital thyroid dysgenesis in DHTP/B6 mice
To verify whether the thyroid hypoplasia and the frequent hemiagenesis observed in DHTP/B6 mice were congenital phenomena that were due to disturbances in thyroid organogenesis, we investigated the size and morphology of the gland during embryonic development. The thyroid bud originates from the primitive pharynx, where it migrates downward along the midline of the embryo. Shortly after detachment from the pharyngeal floor, the thyroid cell precursors proliferate and the bud increases in size and begins to expand laterally to form the two lobes of the thyroid gland that are symmetrically located on either side of the trachea. In the mouse, formation of the two lobes is completed by E15–16. All of the DHTP/B6 embryos examined presented embryonic thyroid lobes invariably smaller than those of cognate wild-type mice. Hypoplasia was already evident at E15 (Fig. 5, B vs. A) and persisted to the end of pregnancy (E18) (Fig. 5, E vs. D). Furthermore, in 30% of the DHTP/B6 embryos, only a single lobe of the developing thyroid was visible, consistent with the defect observed in adult DHTP/B6 mice (Fig. 5, C and F). At E18, the thyroids of DHTP/B6/Sv-F1 embryos did not appear different from those of cognate wild type (data not shown). Thus, thyroid dysgenesis with hypoplasia and increased incidence of hemiagenesis observed in adult DHTP/B6 mice was congenital and followed disturbances in organogenesis.
Reduced Tg biosynthesis in DHTP/B6
Several genes encoding proteins involved in the thyroid hormone biosynthetic machinery are expressed either exclusively, or at high levels, in thyroid follicular cells and have been proposed to be under the control of Titf1, Pax8, or both (29). We decided to investigate whether the decreased levels in circulating thyroid hormones was the consequence of defects in the expression of thyroid-enriched genes. This study was carried out in thyroid glands obtained from fetuses at E18. We chose this stage because during fetal life, maternal thyroid hormones largely make up for a reduced supply from the defective embryonic thyroid, thus preventing a compensatory TSH response that might complicate similar studies in the adult. At E18, thyroid follicular cells are capable of thyroid hormone biosynthesis and express all known genes required for hormone production and metabolism, such as Tg, thyroperoxidase (TPO), TSHR, NIS, and iodothyronine deiodinase type I (Dio1). We analyzed the mRNA levels of these genes by quantitative RT-PCR in RNA extracted from DHTP/B6 and wild-type E18 thyroids (Fig. 6). Only the levels of Tg and Dio1 mRNA were significantly reduced in DHTP/B6 compared with their cognate wild-type samples. These data suggest that a reduced Tg biosynthesis is primarily responsible for the hypothyroidism observed in adults. In agreement with this hypothesis, both Tg and Dio1 mRNA levels were similar in DHTP/B6/Sv-F1 and their cognate wild type (Fig. 6), thus strongly supporting their relevant role in the establishment of the disease and suggesting a related regulation of the two genes.
The morphological and molecular data obtained in the analysis of the DHTP/B6 mice indicate that these mice present congenital hypothyroidism as a consequence of defects in both the growth and differentiation of thyroid follicular cells.
Gene expression profiles in thyroids
To define the entire spectrum of changes in gene expression related to the hypothyroidism observed in DHTP/B6 mice, we analyzed the global gene expression profiles in E18 thyroids dissected from either DHTP/B6 or DHTP/B6/Sv-F1 embryos and their respective wild-type littermates using the Affymetrix mouse expression set, MOE430. For each genotype, three independent RNA samples were used, each representing RNA pooled from four embryonic thyroids. Statistically significant differences in the level of gene expression between DHTP and respective control mice were identified with values of P 0.05 and fold changes 1.5. Using these parameters, of the 18,000 transcripts with detectable expression in thyroid, the expression levels of 163 genes were different comparing thyroids from DHTP/B6 mice (showing TD) with their corresponding wild-type, whereas the expression of only 19 genes was different between the thyroids of DHTP/B6/Sv-F1 (not showing TD) and their wild-type littermates. These data show that the double heterozygosity for Titf1 and Pax8 triggers extremely different transcriptional consequences in the two strains, given that only seven genes were similarly regulated in the two different genetic backgrounds (Fig. 7, A and B). In Table 1, we list the genes that have modified expression in only DHTP/B6, with a fold change compared with the wild-type 2. A complete list of all genes that have modified expression is available on request.
The 156 genes modulated differently in DHTP/B6 compared with wild-type thyroids were classified according to their known or predicted function and assigned to a functional category according to the gene ontology database for the levels 2 and 3 biological process (Table 2). The distribution of the 156 genes among the various categories is no different from the total gene distribution in the thyroid used as the reference dataset. The only notable exceptions are the genes involved in cell adhesion that are significantly more represented (P = 5 x 10–5 by the binomial test).
One hypothesis to explain the resistance of the Sv strain to develop hypothyroidism when combined with heterozygous null mutations in Titf1 and Pax8 genes could be the reduced expression of thyroid-specific genes in this strain. We therefore compared the expression levels in the thyroids from wild-type B6 and DHTP/B6/Sv-F1 for those genes known to be relevant for thyroid development and differentiation. We did not observe significant changes between the two strains (Fig. 7C), except for TPO, which is expressed at a higher level in B6 than in DHTP/B6/Sv-F1 thyroid. Thus, the difference in susceptibility to develop hypothyroidism between these two strains is not related to significant differences in the expression of thyroid-enriched genes.
Identification of genes responsible for strain-specific hypothyroidism
To estimate the number of loci that contribute to the appearance of a hypothyroid phenotype in B6 and not in Sv, DHTP/B6/Sv-F1 hybrid mice were bred with B6 wild-type mice to generate a cohort of 540 first backcross mice. In these backcross progenies, DHTP mice were born at the expected Mendelian frequency (119/540, 22%) and in a similar proportion between the two sexes (61 male and 58 female). At d 30, 35 DHTP mice (16 male and 19 female) displayed elevated TSH levels (mean value, 22.6 ± 9.7), whereas 84 mice (45 male and 39 female) appeared euthyroid (mean value, 4.9 ± 1.6). Hence, the ratio of hypothyroid to euthyroid DHTP mice is almost 1:4. The sex ratio in the affected mice is normal, thus demonstrating that sex does not influence the appearance of the condition.
This observed pattern of segregation is most easily explained by a model in which the DHTP-induced hypothyroidism is a recessive trait contributed by two autosomal B6-specific alleles (30). Other hypotheses, involving a number of loci other than two, have a probability value of P < 10–5. Importantly, the observed proportion of hypothyroid vs. euthyroid animals among the DHTP mice in this backcross experiment is not easily compatible with a linkage to either the Titf1 or the Pax8 locus, thus indicating that the difference between the B6 and the Sv strain is unlikely to be due to differences in expression levels of Titf1 and Pax8. Preliminary genetic mapping in this cohort of mice support this model, indicating that two B6-specific modifiers are linked to the hypothyroid condition (Amendola, E., M. De Felice, T. Dragani, and R. Di Lauro, manuscript in preparation).
Discussion
Here we present evidence for the multigenic origin of congenital hypothyroidism with thyroid dysgenesis. TD is a frequent human condition characterized by elevated TSH levels, reduced secretion of thyroid hormones, and altered development of the thyroid gland. We generated a mouse model with partial deficiencies in Titf1 and Pax8, two transcription factors known to be essential for organogenesis and considered to be important players in functional differentiation of the thyroid gland. Single heterozygous null mutations for either gene have been reported and do not present a TD phenotype (8, 9). In this study, we have confirmed these observations in a congenic genetic background and have demonstrated that the combination of two heterozygous null mutations for Titf1 and Pax8 results in a severe hypothyroidism characterized by elevated TSH, reduced thyroid hormones, decreased body weight accompanied by thyroid hypoplasia, and increased incidence of thyroid hemiagenesis. Interestingly, such a phenotype is only present in the B6 genetic background, whereas it is completely absent in both Sv and in DHTP/B6/Sv-F1 hybrids. Genetic data strongly suggest that two modifier genes are responsible for the observed strain specificity and that the B6 alleles of such modifiers, which predispose to hypothyroidism, are recessive to those of Sv.
Hence, this mouse model suggests that human congenital hypothyroidism can be of multigenic origin and indicates that at least four genes could be involved: Titf1, Pax8, and the two modifiers suggested by this study. This hypothesis is consistent with the observation in humans that mutations in either TITF1 or PAX8 display variable expressivity (7, 11) or incomplete penetrance (13). The relevance of the genetic background in the development of the TD phenotype reported here might also explain the observation that null mutations in either Titf1 (23) or Pax8 (24) genes are recessive in pure mouse strains but dominant in humans (7, 8, 11, 12, 13).
It is noteworthy that Titf1 and Pax8 are transcription factors that have been demonstrated to exert at least two different roles in the thyroid gland. The first during organogenesis, where both factors are required for proper development of the gland (23, 24, 25), and the second in the fully differentiated thyroid gland, where both factors have been implicated in controlling expression of genes important for thyroid hormone biosynthesis, such as Tg (31), thyroperoxidase (32) and NIS (33). Interestingly, it has been demonstrated that Titf1 and Pax8 synergize only on the Tg gene promoter and that such a synergism correlates with a physical interaction between them (34). Consistent with this observation, the DHTP/B6 mice show reduced levels of Tg mRNA, whereas both TPO and NIS mRNA levels were normal. The simplest model to explain these observations predicts that for Titf1 and Pax8 to synergize, the participation of at least two other genes is essential. We suggest that when at least one of these genes is an Sv allele, cooperation also occurs at the diminished concentration of Titf1 and Pax8 present in DHTP mice, whereas if both alleles are of B6 origin, cooperation occurs only at the full dosage of the two transcription factors. Hence, according to this model, the modifiers indicated by this study would be transcriptional cofactors that participate in the assembly of the complex containing Titf1 and Pax8. If this were the case, it would be the first example of modifiers involving component(s) of the transcriptional apparatus (35).
The observation that the size of the embryonic thyroid gland is reduced in DHTP/B6 embryos is consistent with a model that implicates Titf1 and Pax8 in the control of thyroid-cell precursor survival and/or proliferation (25) and suggests that these two transcription factors need to cooperate with each other to exert this function.
Interestingly, adult DHTP/B6 mice, even in presence of very high serum TSH levels, display low levels of T4 and possess a thyroid gland that is smaller than that of control mice. One possible explanation for this reduced size of the thyroid gland, even in the presence of high levels of circulating TSH, is that Tshr expression might be down-regulated in our mutants, as already reported in the case of Titf1+/– (9). In DHTP/B6 the amount of Tshr mRNA is, on average, 50% lower than that in wild-type mice, although this difference does not reach statistical significance. It is unlikely that the reduced expression of Tshr we observed in the DHTP/B6 mice is responsible for the thyroid phenotype. Heterozygous loss-of-function Tshr mutations in mice (26) and humans (36) do not display overt hypothyroidism. In addition, the strong expression of NIS (Fig. 3) confirms that the TSH/Tshr pathway is active in DHTP thyroid follicular cells. It is possible that the inability of TSH to rescue thyroid gland size is due to the impaired expression of gene(s), controlled by Titf1 or Pax8, involved in thyroid cell proliferation. It is worth noting that a hypoplastic thyroid in the presence of elevated TSH levels has been reported in a number of patients carrying heterozygous mutations for both TITF1 (7, 8) and PAX8 (11). Interestingly, a patient with a PAX8 mutation presented a normal-sized thyroid (15), underlining the relevance of the genetic background in the establishment of the phenotype consequent to mutations in this gene. Alternatively, the reduced thyroid gland size observed in adult DHTP/B6 mice could be the result of reduced gland growth in embryogenesis and of the consequent reduced pool of follicular cells.
Another important finding of our study is the observation that 30% of DHTP mice displayed hemiagenesis of the thyroid. The process controlling the formation of two symmetrical thyroid lobes, which fails in the case of hemiagenesis, is still unknown. In humans, thyroid hemiagenesis has a prevalence of 0.05% in healthy children (37). The occurrence of hemiagenesis in monozygotic twins (38), among members of the same family (39), or together with other thyroid malformations within one family (40, 41, 42) suggests that genetic factors could be involved in this anomaly. Candidate genes responsible for the hemiagenesis have not yet been described; however, this condition has been reported in some cases of Williams syndrome caused by a deletion at 7q11.23 (43, 44, 45) and in a patient carrying a deletion encompassing the TITF1 locus. Hemiagenesis of the thyroid has been reported in both Hoxa–/– (46, 47) and Shh–/– mouse embryos (48). It is worth noting that in both Hoxa3- and Shh-null mice, hemiagenesis is most likely secondary to defects in the Hoxa3- or Shh-dependent patterning of other tissues, because these two genes are not expressed in the thyroid (48, 49). On the contrary, hemiagenesis in DHTP mice must be the consequence of impairment of thyroid-specific process(es) because Titf1 and Pax8 are expressed in thyroid follicular cells but are absent in neighboring cells. This observation gives additional relevance to the DHTP mice as a model to further our understanding of human congenital dysgenesis. Taken together, the results in Hoxa3–/– and Shh–/– mice with those reported in this study strongly suggest that the determination of the final shape of the thyroid gland requires interactions between autonomous events restricted to the thyroid follicular cells that must interact with an appropriate cellular environment. In humans, hemiagenesis is more frequently associated with subclinical hypothyroidism (50) than with overt hypothyroidism (22, 51); furthermore, hemiagenesis is reported among euthyroid first-degree relatives of patients with congenital hypothyroidism (21). These data suggest that in the absence of other defects, the reduction of 50% of thyroid tissue per se is not sufficient to cause hypothyroidism. It is possible that the frank hypothyroidism displayed by DHTP/B6 mice could result from the decreased number of follicular cells (hypoplasia or hemiagenesis as well) and the marked reduction in Tg biosynthesis with a consequent decrease in thyroid hormone secretion.
Global gene expression was studied in E18 embryonic thyroids from wild-type and DHTP mice in either the B6 or the B6/Sv-F1 genetic backgrounds. The results of such studies showed that the expression of many more genes was affected by the DHTP mutations in the disease-prone B6 background than in the healthy B6/Sv hybrid background. More importantly, this study indicated a group of genes that, in addition to the decreased Tg level, might contribute to the observed phenotype. We suggest that among these genes, there might be those responsible for hypoplasia and hemiagenesis; however, for these aspects of the DHTP phenotype, condition studies of gene expression earlier in organogenesis could be more informative.
In conclusion, the DHTP mouse model demonstrates that TD can be of multigenic origin in mice and strongly suggests that a similar pathogenetic mechanism might be operating in some human patients. Furthermore, the strain specificity observed opens the way to the identification of modifier genes that, in conjunction with Titf1 and Pax8, control morphogenesis, growth, and differentiation of the thyroid gland. Finally, this study demonstrates that, at least in mice, hemiagenesis of the thyroid gland can be explained by genetic mechanisms.
Footnotes
This work was supported, in part, by Telethon, Congenital hypothyroidism with thyroid dysgenesis: candidate genes, animal models and molecular mechanisms; Ministero dell’Universita e della Ricerca Scientifica e Tecnologica, I geni dell’uomo, cluster 01; and Biogem societa coperativa a responsabilita limitata (to A.R. and A.A.).
First Published Online September 8, 2005
Abbreviations: Dio1, Iodothyronine deiodinase type I; E0.5, embryonic d 0.5; NIS, sodium/iodide symporter; TD, thyroid dysgenesis; Tg, thyroglobulin; TPO, thyroperoxidase; TSHR, TSH receptor.
Accepted for publication September 1, 2005.
References
Fisher DA, Klein, A 1981 Thyroid development and disorders of thyroid function in the newborn. N Engl J Med 304:702–712
De Felice M, Di Lauro R 2004 Thyroid development and its disorders: genetics and molecular mechanisms. Endocr Rev 25:722–746
Park SM, Chatterjee VK 2005 Genetics of congenital hypothyroidism. J Med Genet 42:379–389
Perry R, Heinrichs C, Bourdoux P, Khoury K, Szots F, Dussault JH, Vassart G, Van Vliet G 2002 Discordance of monozygotic twins for thyroid dysgenesis: implications for screening and for molecular pathophysiology. J Clin Endocrinol Metab 87:4072–4077
Sunthornthepvarakui T, Gottschalk ME, Hayashi Y, Refetoff S 1995 Brief report: resistance to thyrotropin caused by mutations in the thyrotropin-receptor gene. N Engl J Med 332:155–160
Refetoff S 2003 Resistance to thyrotropin. J Endocrinol Invest 26:770–779
Krude H, Schutz B, Biebermann H, von Moers A, Schnabel D, Neitzel H, Tonnies H, Weise D, Lafferty A, Schwarz S, DeFelice M, von Deimling A, van Landeghem F, DiLauro R, Gruters A 2002 Choreoathetosis, hypothyroidism, and pulmonary alterations due to human NKX2–1 haploinsufficiency. J Clin Invest 109:475–480
Pohlenz J, Dumitrescu A, Zundel D, Martine U, Schonberger W, Koo E, Weiss RE, Cohen RN, Kimura S, Refetoff S 2002 Partial deficiency of thyroid transcription factor 1 produces predominantly neurological defects in humans and mice. J Clin Invest 109:469–473
Moeller LC, Kimura S, Kusakabe T, Liao XH, Van Sande J, Refetoff S 2003 Hypothyroidism in thyroid transcription factor 1 haploinsufficiency is caused by reduced expression of the thyroid stimulating hormone receptor. Mol Endocrinol 17:2295–2302
Doyle DA, Gonzalez I, Thomas B, Scavina M 2004 Autosomal dominant transmission of congenital hypothyroidism, neonatal respiratory distress, and ataxia caused by a mutation of NKX2–1. J Pediatr 145:190–193
Macchia PE, Lapi P, Krude H, Pirro MT, Missero C, Chiovato L, Souabni A, Baserga M, Tassi V, Pinchera A, Fenzi G, Gruters A, Busslinger M, Di Lauro R 1998 PAX8 mutations associated with congenital hypothyroidism caused by thyroid dysgenesis. Nat Genet 19:83–86
Vilain C, Rydlewski C, Duprez L, Heinrichs C, Abramowicz M, Malvaux P, Renneboog B, Parma J, Costagliola S, Vassart G 2001 Autosomal dominant transmission of congenital thyroid hypoplasia due to loss-of-function mutation of PAX8. J Clin Endocrinol Metab 86:234–238
Congdon T, Nguyen LQ, Nogueira CR, Habiby RL, Medeiros-Neto G, Kopp P 2001 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 86:3962–3967
Komatsu M, Takahashi T, Takahashi I, Nakamura M, Takahashi I, Takada G 2001 Thyroid dysgenesis caused by PAX8 mutation: the hypermutability with CpG dinucleotides at codon 31. J Pediatr 139:597–599
Meeus L, Gilbert B, Rydlewski C, Parma J, Roussie AL, Abramowicz M, Vilain C, Christophe D, Costagliola S, Vassart G 2004 Characterization of a novel loss of function mutation of PAX8 in a familial case of congenital hypothyroidism with in-place, normal-sized thyroid. J Clin Endocrinol Metab 89:4285–4291
Clifton-Bligh RJ, Wentworth JM, Heinz P, Crisp MS, John R, LJ, Ludgate M, Chatterjee V 1998 Mutation of the gene encoding human TTF-2 associated with thyroid agenesis, cleft palate and choanal atresia. Nat Genet 19:399–401
Castanet M, Park SM, Smith A, Bost M, Leger J, Lyonnet S, Pelet A, Czernichow P, Chatterjee K, Polak M 2002 A novel loss-of-function mutation in TTF-2 is associated with congenital hypothyroidism, thyroid agenesis and cleft palate. Hum Mol Genet 11:2051–2059
Castanet M, Sura Trueba S, Chauty A, Carre A, Heath, S, Leger J, Lyonnet S, Czernichow P, Polak M 2005 Linkage and mutational analysis of familial thyroid dysgenesis suggest genetic heterogeneity. Eur J Hum Genet 13:232–239
Castanet M, Polak M, Bonaiti-Pellie C, Lyonnet S, Czernichow P, Leger J 2001 Nineteen years of national screening for congenital hypothyroidism: familial cases with thyroid dysgenesis suggest the involvement of genetic factors. J Clin Endocrinol Metab 86:2009–2014
Castanet M, Lyonnet S, Bonaiti-Pellie C, Polak M, Czernichow P, Leger J 2000 Familial forms of thyroid dysgenesis among infants with congenital hypothyroidism. N Engl J Med 343:441–442
Leger J, Marinovic D, Garel C, Bonaiti-Pellie C, Polak M, Czernichow P 2002 Thyroid developmental anomalies in first degree relatives of children with congenital hypothyroidism. J Clin Endocrinol Metab 87:575–580
Mikosch P, Gallowitsch HJ, Kresnik E, Molnar M, Gomez I, Lind P 1999 Thyroid hemiagenesis in an endemic goiter area diagnosed by ultrasonography: report of sixteen patients. Thyroid 9:1075–1084
Kimura S, Hara Y, Pineau T, Fernandez-Salguero P, Fox C, Ward J, Gonzalez F 1996 The T/ebp null mouse: thyroid-specific enhancer-binding protein is essential for the organogenesis of the thyroid, lung, ventral forebrain, and pituitary. Genes Dev 10:60–69
Mansouri A, Chowdhury K, Gruss P 1998 Follicular cells of the thyroid gland require Pax8 gene function. Nat Genet 19:87–90
Parlato R, Rosica A, Rodriguez-Mallon A, Affuso A, Postiglione MP, Arra C, Mansouri A, Kimura S, Di Lauro R, De Felice M 2004 An integrated regulatory network controlling survival and migration in thyroid organogenesis. Dev Biol 276:464–475
Postiglione MP, Parlato R, Rodriguez-Mallon A, Rosica A, Mithbaokar P, Maresca M, Marians RC, Davies TF, Zannini MS, De Felice M, Di Lauro R 2002 Role of the thyroid-stimulating hormone receptor signaling in development and differentiation of the thyroid gland. Proc Natl Acad Sci USA 99:15462–15467
Chomczynski P, Sacchi N 1987 Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 162:156–159
Benjamini Y, Hochberg Y 1995 Controlling the false discovery rate: a practical and powerful approach to multiple testing. J R Stat Soc B 57:289–300
Damante G, Tell G, Di Lauro R 2001 A unique combination of transcription factors controls differentiation of thyroid cells. Prog Nucleic Acid Res Mol Biol 66:307–356
Li W, Reich J 2000 A complete enumeration and classification of two-locus disease models. Hum Hered 50:334–349
De Felice M, Damante G, Zannini MS, Francis-Lang H, Di Lauro R 1995 Redundant domains contribute to the transcriptional activity of Thyroid Transcription Factor 1 (TTF-1). J Biol Chem 270:26649–26656
Zannini MS, Francis-Lang H, Plachov D, Di Lauro R 1992 Pax-8, a paired domain-containing protein, binds to a sequence overlapping the recognition site of a homeodomain and activates transcription from two thyroid-specific promoters. Mol Cell Biol 12:4230–4241
Ohno M, Zannini MS, Levy O, Carrasco N, Di Lauro R 1999 The paired-domain transcription factor Pax8 binds to the upstream enhancer of the rat sodium/iodide symporter gene and participates in both thyroid-specific and cyclic-AMP-dependent transcription. Mol Cell Biol 19:2051–2060
Di Palma T, Nitsch R, Mascia A, Nitsch L, Di Lauro R, Zannini M 2003 The paired domain-containing factor Pax8 and the homeodomain-containing factor TTF-1 directly interact and synergistically activate transcription. J Biol Chem 278:3395–3402
Flint J, Valdar W, Shifman S, Mott R 2005 Strategies for mapping and cloning quantitative trait genes in rodents. Nat Rev Genet 6:271–286
Alberti L, Proverbio MC, Costagliola S, Romoli R, Boldrighini B, Vigone MC, Weber G, Chiumello G, Beck-Peccoz P, Persani L 2002 Germline mutations of TSH receptor gene as cause of nonautoimmune subclinical hypothyroidism. J Clin Endocrinol Metab 2002 87:2549–2555
Maiorana R, Carta A, Floriddia G, Leonardi D, Buscema M, Sava L, Calaciura F, Vigneri R 2003 Thyroid hemiagenesis: prevalence in normal children and effect on thyroid function. J Clin Endocrinol Metab 88:1534–1536
Greig WR, Henderson AS, Boyle JA, McGirr EM, Hutchison JH 1966 Thyroid dysgenesis in two pairs of monozygotic twins and in a mother and child. J Clin Endocrinol Metab 26:1309–1316
Rajmil HO, Rodriguez-Espinosa J, Soldevila J, Ordonez-Llanos J 1984 Thyroid hemiagenesis in two sisters. J Endocrinol Invest 7:393–394
Rosenberg T, Gilboa Y 1980 Familial thyroid ectopy and hemiagenesis. Arch Dis Child 55:639–641
McLean R, Howard N, Murray IP 1985 Thyroid dysgenesis in monozygotic twins: variants identified by scintigraphy. Eur J Nucl Med 10:346–348
Castanet M, Leenhardt L, Leger J, Simon-Carre A, Lyonnet S, Pelet A, Czernichow P, Polak M 2005 Thyroid hemiagenesis is a rare variant of thyroid dysgenesis with a familial component but without Pax8 mutations in a cohort of 22 cases. Pediatr Res 57:908–913
Bini R, Pela I 2004 New case of thyroid dysgenesis and clinical signs of hypothyroidism in Williams syndrome. Am J Med Genet A 127:183–185
Stagi S, Bindi G, Neri AS, Giovannucci-Uzielli ML, Lapi E, Galluzzi F, Salti R 2003 Thyroid hypoplasia of the left lobe in two girls affected by Williams syndrome. Clin Dysmorphol 12:267–268
Cammareri V, Vignati G, Nocera G, Beck-Peccoz P, Persani L 1999 Thyroid hemiagenesis and elevated thyrotropin levels in a child with Williams syndrome. Am J Med Genet 85:491–494
Manley NR, Capecchi M 1995 The role of Hoxa-3 in mouse thymus and thyroid development. Development 121:1989–2003
Manley NR, Capecchi M 1998 Hox group 3 paralogs regulate the development and migration of the thymus, thyroid, and parathyroid glands. Dev Biol 195:1–15
Fagman H, Grande M, Gritli-Linde A, Nilsson M 2004 Genetic deletion of sonic hedgehog causes hemiagenesis and ectopic development of the thyroid in mouse. Am J Pathol 164:1865–1872
Meunier D, Aubin J, Jeannotte L 2003 Perturbed thyroid morphology and transient hypothyroidism symptoms in Hoxa5 mutant mice. Dev Dyn 227:367–378
Calaciura F, Motta RM, Miscio G, Fichera G, Leonardi D, Carta A, Trischitta V, Tassi V, Sava L, Vigneri R 2002 Subclinical hypothyroidism in early childhood: a frequent outcome of transient neonatal hyperthyrotropinemia. J Clin Endocrinol Metab 87:3209–3214
Karabay N, Comlekci A, Canda MS, Bayraktar F, Degirmenci B 2003 Thyroid hemiagenesis with multinodular goiter: a case report and review of the literature. Endocr J 50:409–413(Elena Amendola, Pasquale De Luca, Paolo )
Dipartimento di Endocrinologia e Oncologia Molecolare e Clinica (P.E.M.) and Dipartimento di Biologia e Patologia Cellulare e Molecolare (D.T., V.M., R.D.L.), Universita Federico II
Biogem at CEINGE (A.R.), Naples, Italy
Instituto Nazionale Tumori Fondazione Pascale (G.C., C.A.), 80131 Naples, Italy
Laboratory of Metabolism (S.K.), National Cancer Institute, National Institutes of Health, Bethesda, Maryland 20892
Max Planck Institute für Biophysikalische Chemie (A.M.), Abteilung Molekulare Zellbiologie, 37077 Gttingen, Germany
Abstract
Congenital hypothyroidism with thyroid dysgenesis (TD) is a frequent human condition characterized by elevated levels of TSH in response to reduced thyroid hormone levels. Congenital hypothyroidism is a genetically heterogeneous disease. In the majority of cases studied, no causative mutations have been identified and very often the disease does not show a Mendelian transmission. However, in approximately 5% of cases, it can be a consequence of mutations in genes encoding the TSH receptor or the transcription factors TITF1, FOXE1, or PAX8. We report here that in mouse models, the combination of partial deficiencies in the Titf1 and Pax8 genes results in an overt TD phenotype that is absent in either of the singly deficient, heterozygous mice. The disease is characterized by a small thyroid gland, elevated levels of TSH, reduced thyroglobulin biosynthesis, and high occurrence of hemiagenesis. The observed phenotype is strain specific, and the pattern of transmission indicates that at least two other genes, in addition to Titf1 and Pax8, are necessary to generate the condition. These results show that TD can be of multigenic origin in mice and strongly suggest that a similar pathogenic mechanism may be observed in humans.
Introduction
CONGENITAL HYPOTHYROIDISM with thyroid dysgenesis (TD) (1) is a frequent human condition, the genetic basis of which has just begun to be described (2, 3), even though TD has been found discordant in 12 of 13 monozygotic twins studied (4). Disease-causing mutations were identified in the receptor for TSH (TSHR) (5, 6) and in the transcription factors TITF1 (7, 8, 9, 10), PAX8 (11, 12, 13, 14, 15), and FOXE1/TITF2 (16, 17). However, only 5% of the patients with TD examined presented mutations in these genes. This number could be an underestimate because the mutation analysis was limited to the coding regions of the genes examined, thus excluding potential disease-causing mutations in noncoding regions that might affect regulatory regions or result in aberrant splicing. Still, the low frequency of mutation in the genes studied thus far suggests that other genes (18) and/or other mechanisms could be involved in the pathogenesis of TD. In this study, we addressed whether TD could be a multigenic disorder. Such a pathogenetic mechanism would explain why most TD patients do not display a clear Mendelian transmission and would be in agreement with the observation that there is a higher incidence of thyroid condition in those families with at least one case of TD (19, 20, 21). Furthermore, the incomplete penetrance and the variable expressivity observed in familial cases of TD (11, 13) is also consistent with a multigenic mechanism.
To provide proof for a possible multigenic origin of TD, we used mouse models. The null mutations generated in the genes encoding the transcription factors Titf1 (23) and Pax8 (24) only display a TD phenotype when homozygous. In neither case do heterozygous animals show an overt TD phenotype. We report here that mice heterozygous for both Titf1- and Pax8-null alleles, denominated DHTP (double heterozygous for both Titf1- and Pax8-null mutations), present TD. An in-depth study of the phenotype reveals elevated TSH levels in DHTP mice associated with reduced thyroid hormone secretion and decreased thyroglobulin (Tg) biosynthesis. Furthermore, we show that the thyroid gland of DHTP mice is smaller during embryonic life, thus making these mice a bona fide TD model. Interestingly, adult DHTP mice also show a high incidence of thyroid hemiagenesis, suggesting also that this condition can be genetically determined.
Finally, we demonstrate that the observed TD condition is strain specific. TD is displayed when the DHTP mutations are in the C57BL/6 genetic background (DHTP/B6), and it is not observed in the 129/Sv strain (DHTP/Sv) or in the F1 generation of a C57BL/6 x 129/Sv cross (DHTP/B6/Sv). Furthermore, the pattern of transmission of the TD phenotype in the backcross DHTP/B6/Sv x C57BL/6 suggests that at least two additional C57BL/6-specific alleles, in addition to the null mutations in Titf1 and Pax8, are responsible for the emergence of TD.
This study is the first demonstration that TD can be of multigenic origin. The mouse model presented here is useful to begin understanding the molecular mechanism underlying TD in those patients for whom the genetic basis is unknown and might lead to the identification of additional genes required for proper thyroid gland organogenesis and efficient thyroid hormone biosynthesis.
Materials and Methods
Mice and breeding
Heterozygous Titf1-null mice (Titf1+/–) (23) and heterozygous Pax8-null mice (Pax8+/–) (24), originally generated in a mixed genetic background, were crossed to generate double heterozygote for both Titf1- and Pax8-null mutations (DHTP) mice. DHTP mice were backcrossed 10 generations in a C57BL/6J background (Charles River Laboratories, Calco, Italy) (Fig. 1). The DHTP progeny were estimated to have more than 99% of their genes derived from the C57BL/6J strain. Almost congenic DHTP mice were mated with 129/SvPasCrl wild-type (Charles River Laboratories) to generate F1 hybrid DHTP/B6/Sv mice.
Both male and female mice were used for breeding. Animals were housed in a conventional facility with a 12-h light, 12-h dark cycle and were supplied with standard rodent food. Genotypes were determined by PCR using genomic DNA isolated from the tail snips as described (25).
Histology and immunohistochemistry
Staged mouse embryos were obtained by dissection of pregnant females. The day in which the vaginal plug was detected was designated as embryonic d 0.5 (E0.5). Animals were killed by cervical dislocation. Adult thyroid glands were dissected from 3-month-old mice. Thyroids and embryos were fixed overnight at 4 C in 4% paraformaldehyde in PBS at pH 7.2, dehydrated through ethanol series, cleared in xylene, and embedded in paraffin, and 7-μm sections were cut.
For histological examinations, serial sections from dissected DHTP and wild-type mice were stained with Harry’s hematoxylin and eosin (BDH Laboratory Supplies, Poole, UK), according to the manufacturer’s instructions. For immunohistochemical analyses, sections from embryos and dissected thyroids were dewaxed by standard techniques. Heat treatment was performed to retrieve the antigen sites. Staining procedures and chromogenic reactions were carried out according to the Vectastain ABC kit protocol (Vector Laboratories, Burlingame, CA). The primary antibodies used antimouse Pax8 and antirat sodium/iodide symporter (NIS) have been described previously (26).
Hormone measurements
Blood samples were obtained from the tail vein, under light anesthesia, and collected in microtubes without anticoagulant. After coat formation, the samples were centrifuged, and the recovered serum was kept frozen at –20 C until assayed.
TSH levels were determined by specific RIA kit (rat TSH kit) from Amersham Biosciences Europe (Freiburg, Germany). Total T4 was measured with the Immulite analyzer using a commercial kit, as recommended by the manufacturer (Diagnostics Products Corp., Los Angeles, CA).
Real-time PCR
Total RNA was extracted using the guanidine isothiocyanate method (27) and further purified by the RNAeasy minikit (Qiagen, Hilden, Germany). Four micrograms of total RNA were used as template for the synthesis of the first-strand cDNA reverse starting from oligo dT primers using the Superscript II Reverse Transcriptase kit (Invitrogen Life Technologies, Carlsbad, CA) according to the instructions of the manufacturer.
Reactions for the quantification of mRNAs were performed in an ABI Prism 7300 Real Time PCR System (Applied Biosystems, Foster City, CA) using SYBR green as detector dye. The reaction mixtures contained 25 μl of SYBR Green PCR master mix (Applied Biosystems), primers at a concentration of 300 nM each, 10-ng template cDNA in a final volume of 50 μl. Reactions were carried out in triplicate.
Specific primer sets for each gene were designed using the program Primer Express (Applied Biosystems). The sequences of the primers used in the reactions were as follows: 1), Thyroglobulin: forward, 5'-CAT GGA ATC TAA TGC CAA GAA CTG-3'; reverse, 5'-TCC CTG TGA GCT TTT GGA ATG-3' . 2), Thyroid peroxidase: forward, 5'-CAA AGG CTG GAA CCC TAA TTT CT-3'; reverse, 5'-AAC TTG AAT GAG GTG CCT TGT CA-3'. 3) TSH receptor: forward, 5'-TCC CTG AAA ACG CAT TCC A-3'; reverse, 5'-GCA TCC AGC TTT GTT CCA TTG-3'. 4) Iodothyronine deiodinase I: forward, 5'-CCT CCA CAG CCG ATT TCC T-3'; reverse, 5'-GTT CTT CTT AAA AGC CCA GCC A-3'. 5) Abelson: forward, 5'-TCGGACGTGTGGGCATTT-3'; reverse, 5'-CGCATGAGCTCGTAGACCTTC-3'.
For each sample, the expression of the genes of interest was normalized for the expression of Abelson gene measured under the same condition. For each genotype, the data obtained represent the mean of three independent samples, each composed of the thyroids dissected from four E18 embryos.
RNA microarray
The thyroid glands were dissected from E18 embryos, and total RNA was extracted with guanidine isothiocyanate (27) and further purified by the RNeasy minikit (Qiagen). cRNA was generated by using the Affymetrix One-Cycle Target Labeling and Control Reagent kit (Affymetrix Inc., Santa Clara, CA) following the protocol of the manufacturer. The biotinylated cRNA was hybridized to the MOE430A Affymetrix DNA chips, containing over 22,000 genes and open-reading frames from Mus musculus Genome databases GenBank, dbEST, and RefSeq. Chips were washed and scanned on the Affymetrix Complete GeneChip Instrument System, generating digitized image data files. Reactions were carried out in triplicate.
Digitized image data files were analyzed by Micro Array Suite 5.0 for detection calls (Affymetrix Inc.) and Robust Multichip Average for expression values. The expression values obtained were analyzed using GeneSpring 7.1 (Silicon Genetics, Redwood City, CA). Results were filtered for flag (presence call) and then for fold change >1.5, obtaining a total of 445 probe sets differentially expressed in the B6 strain and 55 for the B6/Sv-F1. Statistical analysis was performed using the Welch t test, with P 0.05 as the cutoff value. The P values were adjusted by the Benjamini and Hochberg correction (28). Genes were classified by Gene Ontology for Biological Process and Molecular function. Hierarchical clustering were performed on gene lists by the Gene Tree algorithm using the Pearson correlation.
Results
Combination of recessive mutations in Titf1 and Pax8 genes induces an elevation in serum of TSH levels
Heterozygous null mice for either Titf1 (Titf1+/–) or Pax8 (Pax8+/–) were bred to generate double heterozygous mice (Titf1+/–, Pax8+/–), herein referred to as DHTP mice (Fig. 1). The analysis of genotype frequencies in the progenies of this cross revealed the expected proportion (25%) of DHTP mice. TSH serum levels were measured in 90-d-old mice to test for thyroid function. TSH levels in the single heterozygous mice resulted comparable to those of wild type. TSH measurements in DHTP mice showed exceedingly abnormal levels of this hormone, albeit in only a fraction of DHTP (Fig. 2A), suggesting impaired thyroid functionality.
Given that both the Titf1+/– and the Pax8+/– parents had a mixed genetic background, with contribution from both C57BL/6J (herein referred to as B6) and 129/SvPasCrl (herein referred to as Sv) strains, the incomplete penetrance of the high TSH phenotype observed in DHTP mice could be the consequence of the diverse genetic background among the mice examined. To test this hypothesis, DHTP mice were backcrossed 10 generations into the B6 strain to obtain both single heterozygous (Titf1+/– or Pax8+/–) and DHTP in a practically congenic B6 genetic background. The TSH levels at d 90 in these mice showed that serum TSH were invariably more elevated in DHTP than in wild-type and single heterozygous littermates, averaging an 8-fold increase (Fig. 2B). Backcrossing the DHTP mice on the Sv strain resulted in an F1 with DHTP mice presenting normal TSH levels (Fig. 2C). Here, these mice are referred to as DHTP/B6/Sv-F1.
These data demonstrate that the partial deficiency of Titf1 and Pax8 when associated with B6-specific allele(s) results in hyperthyrotropinemia when associated with B6-specific allele(s); the data also show that these B6 alleles are recessive to those in the Sv strain. Next, we investigated the thyroid function of DHTP mice in the B6 genetic background (herein called DHTP/B6). For the remainder of this study, the phenotypes of DHTP/B6 mice were compared with either that of wild-type littermates or of DHTP/B6/Sv-F1.
DHTP/B6 mice are hypothyroid
T4 levels measured in 90-d-old DHTP/B6 mice indicated an evident hypothyroidism. DHTP/B6 mice presented low T4 levels compared with their single heterozygous or wild-type littermates (Fig. 3A). Furthermore, in agreement with a reduced thyroid function, a significant reduction in body weight was observed in the DHTP/B6 (Fig. 3B). These data strongly suggest that a reduced secretion of T4 is the primary defect in DHTP/B6 mice and that the increased TSH levels and reduced body weight are secondary effects of thyroid hormone deficiency.
No significant difference in T4 serum levels (Fig. 3C) or body weight (Fig. 3D) was observed among wild-type, single heterozygous, and DHTP in the B6/Sv-F1 genetic background. The histological examination of the DHTP/B6 thyroids showed zones of irregular and diffuse hyperplasia, with follicles lined by tall columnar cells, typical of a chronic TSH stimulation (Fig. 4A), whereas wild-type thyroid follicles consisted of a rim of flattened or cuboidal cells (Fig. 4B). Moreover, immunochemistry assays revealed an increased expression of NIS in DHTP/B6 follicular cells, a sensitive parameter of TSH stimulation (Fig. 4, C vs. D). No morphological alterations were detected in DHTP/B6/Sv-F1 thyroids (data not shown).
Surprisingly, the thyroid gland was not enlarged in 90-d-old DHTP mice, despite an elevated serum TSH concentration. A more detailed analysis revealed that the thyroid of DHTP/B6 thyroid was smaller than the thyroid of control mice (Fig. 4, E vs. F), determined by evaluating the diameters of serial sections of glands (Fig. 4, G vs. H).
An additional remarkable feature of the DHTP/B6 mice was the hemiagenesis of the thyroid observed in about 30% of the animals examined. In these animals, the gland consisted of a single lobe, slightly hypertrophic, correctly located but on only one side of the trachea (Fig. 4I). Thyroid hemiagenesis was never observed in approximately 100 control mice.
Congenital thyroid dysgenesis in DHTP/B6 mice
To verify whether the thyroid hypoplasia and the frequent hemiagenesis observed in DHTP/B6 mice were congenital phenomena that were due to disturbances in thyroid organogenesis, we investigated the size and morphology of the gland during embryonic development. The thyroid bud originates from the primitive pharynx, where it migrates downward along the midline of the embryo. Shortly after detachment from the pharyngeal floor, the thyroid cell precursors proliferate and the bud increases in size and begins to expand laterally to form the two lobes of the thyroid gland that are symmetrically located on either side of the trachea. In the mouse, formation of the two lobes is completed by E15–16. All of the DHTP/B6 embryos examined presented embryonic thyroid lobes invariably smaller than those of cognate wild-type mice. Hypoplasia was already evident at E15 (Fig. 5, B vs. A) and persisted to the end of pregnancy (E18) (Fig. 5, E vs. D). Furthermore, in 30% of the DHTP/B6 embryos, only a single lobe of the developing thyroid was visible, consistent with the defect observed in adult DHTP/B6 mice (Fig. 5, C and F). At E18, the thyroids of DHTP/B6/Sv-F1 embryos did not appear different from those of cognate wild type (data not shown). Thus, thyroid dysgenesis with hypoplasia and increased incidence of hemiagenesis observed in adult DHTP/B6 mice was congenital and followed disturbances in organogenesis.
Reduced Tg biosynthesis in DHTP/B6
Several genes encoding proteins involved in the thyroid hormone biosynthetic machinery are expressed either exclusively, or at high levels, in thyroid follicular cells and have been proposed to be under the control of Titf1, Pax8, or both (29). We decided to investigate whether the decreased levels in circulating thyroid hormones was the consequence of defects in the expression of thyroid-enriched genes. This study was carried out in thyroid glands obtained from fetuses at E18. We chose this stage because during fetal life, maternal thyroid hormones largely make up for a reduced supply from the defective embryonic thyroid, thus preventing a compensatory TSH response that might complicate similar studies in the adult. At E18, thyroid follicular cells are capable of thyroid hormone biosynthesis and express all known genes required for hormone production and metabolism, such as Tg, thyroperoxidase (TPO), TSHR, NIS, and iodothyronine deiodinase type I (Dio1). We analyzed the mRNA levels of these genes by quantitative RT-PCR in RNA extracted from DHTP/B6 and wild-type E18 thyroids (Fig. 6). Only the levels of Tg and Dio1 mRNA were significantly reduced in DHTP/B6 compared with their cognate wild-type samples. These data suggest that a reduced Tg biosynthesis is primarily responsible for the hypothyroidism observed in adults. In agreement with this hypothesis, both Tg and Dio1 mRNA levels were similar in DHTP/B6/Sv-F1 and their cognate wild type (Fig. 6), thus strongly supporting their relevant role in the establishment of the disease and suggesting a related regulation of the two genes.
The morphological and molecular data obtained in the analysis of the DHTP/B6 mice indicate that these mice present congenital hypothyroidism as a consequence of defects in both the growth and differentiation of thyroid follicular cells.
Gene expression profiles in thyroids
To define the entire spectrum of changes in gene expression related to the hypothyroidism observed in DHTP/B6 mice, we analyzed the global gene expression profiles in E18 thyroids dissected from either DHTP/B6 or DHTP/B6/Sv-F1 embryos and their respective wild-type littermates using the Affymetrix mouse expression set, MOE430. For each genotype, three independent RNA samples were used, each representing RNA pooled from four embryonic thyroids. Statistically significant differences in the level of gene expression between DHTP and respective control mice were identified with values of P 0.05 and fold changes 1.5. Using these parameters, of the 18,000 transcripts with detectable expression in thyroid, the expression levels of 163 genes were different comparing thyroids from DHTP/B6 mice (showing TD) with their corresponding wild-type, whereas the expression of only 19 genes was different between the thyroids of DHTP/B6/Sv-F1 (not showing TD) and their wild-type littermates. These data show that the double heterozygosity for Titf1 and Pax8 triggers extremely different transcriptional consequences in the two strains, given that only seven genes were similarly regulated in the two different genetic backgrounds (Fig. 7, A and B). In Table 1, we list the genes that have modified expression in only DHTP/B6, with a fold change compared with the wild-type 2. A complete list of all genes that have modified expression is available on request.
The 156 genes modulated differently in DHTP/B6 compared with wild-type thyroids were classified according to their known or predicted function and assigned to a functional category according to the gene ontology database for the levels 2 and 3 biological process (Table 2). The distribution of the 156 genes among the various categories is no different from the total gene distribution in the thyroid used as the reference dataset. The only notable exceptions are the genes involved in cell adhesion that are significantly more represented (P = 5 x 10–5 by the binomial test).
One hypothesis to explain the resistance of the Sv strain to develop hypothyroidism when combined with heterozygous null mutations in Titf1 and Pax8 genes could be the reduced expression of thyroid-specific genes in this strain. We therefore compared the expression levels in the thyroids from wild-type B6 and DHTP/B6/Sv-F1 for those genes known to be relevant for thyroid development and differentiation. We did not observe significant changes between the two strains (Fig. 7C), except for TPO, which is expressed at a higher level in B6 than in DHTP/B6/Sv-F1 thyroid. Thus, the difference in susceptibility to develop hypothyroidism between these two strains is not related to significant differences in the expression of thyroid-enriched genes.
Identification of genes responsible for strain-specific hypothyroidism
To estimate the number of loci that contribute to the appearance of a hypothyroid phenotype in B6 and not in Sv, DHTP/B6/Sv-F1 hybrid mice were bred with B6 wild-type mice to generate a cohort of 540 first backcross mice. In these backcross progenies, DHTP mice were born at the expected Mendelian frequency (119/540, 22%) and in a similar proportion between the two sexes (61 male and 58 female). At d 30, 35 DHTP mice (16 male and 19 female) displayed elevated TSH levels (mean value, 22.6 ± 9.7), whereas 84 mice (45 male and 39 female) appeared euthyroid (mean value, 4.9 ± 1.6). Hence, the ratio of hypothyroid to euthyroid DHTP mice is almost 1:4. The sex ratio in the affected mice is normal, thus demonstrating that sex does not influence the appearance of the condition.
This observed pattern of segregation is most easily explained by a model in which the DHTP-induced hypothyroidism is a recessive trait contributed by two autosomal B6-specific alleles (30). Other hypotheses, involving a number of loci other than two, have a probability value of P < 10–5. Importantly, the observed proportion of hypothyroid vs. euthyroid animals among the DHTP mice in this backcross experiment is not easily compatible with a linkage to either the Titf1 or the Pax8 locus, thus indicating that the difference between the B6 and the Sv strain is unlikely to be due to differences in expression levels of Titf1 and Pax8. Preliminary genetic mapping in this cohort of mice support this model, indicating that two B6-specific modifiers are linked to the hypothyroid condition (Amendola, E., M. De Felice, T. Dragani, and R. Di Lauro, manuscript in preparation).
Discussion
Here we present evidence for the multigenic origin of congenital hypothyroidism with thyroid dysgenesis. TD is a frequent human condition characterized by elevated TSH levels, reduced secretion of thyroid hormones, and altered development of the thyroid gland. We generated a mouse model with partial deficiencies in Titf1 and Pax8, two transcription factors known to be essential for organogenesis and considered to be important players in functional differentiation of the thyroid gland. Single heterozygous null mutations for either gene have been reported and do not present a TD phenotype (8, 9). In this study, we have confirmed these observations in a congenic genetic background and have demonstrated that the combination of two heterozygous null mutations for Titf1 and Pax8 results in a severe hypothyroidism characterized by elevated TSH, reduced thyroid hormones, decreased body weight accompanied by thyroid hypoplasia, and increased incidence of thyroid hemiagenesis. Interestingly, such a phenotype is only present in the B6 genetic background, whereas it is completely absent in both Sv and in DHTP/B6/Sv-F1 hybrids. Genetic data strongly suggest that two modifier genes are responsible for the observed strain specificity and that the B6 alleles of such modifiers, which predispose to hypothyroidism, are recessive to those of Sv.
Hence, this mouse model suggests that human congenital hypothyroidism can be of multigenic origin and indicates that at least four genes could be involved: Titf1, Pax8, and the two modifiers suggested by this study. This hypothesis is consistent with the observation in humans that mutations in either TITF1 or PAX8 display variable expressivity (7, 11) or incomplete penetrance (13). The relevance of the genetic background in the development of the TD phenotype reported here might also explain the observation that null mutations in either Titf1 (23) or Pax8 (24) genes are recessive in pure mouse strains but dominant in humans (7, 8, 11, 12, 13).
It is noteworthy that Titf1 and Pax8 are transcription factors that have been demonstrated to exert at least two different roles in the thyroid gland. The first during organogenesis, where both factors are required for proper development of the gland (23, 24, 25), and the second in the fully differentiated thyroid gland, where both factors have been implicated in controlling expression of genes important for thyroid hormone biosynthesis, such as Tg (31), thyroperoxidase (32) and NIS (33). Interestingly, it has been demonstrated that Titf1 and Pax8 synergize only on the Tg gene promoter and that such a synergism correlates with a physical interaction between them (34). Consistent with this observation, the DHTP/B6 mice show reduced levels of Tg mRNA, whereas both TPO and NIS mRNA levels were normal. The simplest model to explain these observations predicts that for Titf1 and Pax8 to synergize, the participation of at least two other genes is essential. We suggest that when at least one of these genes is an Sv allele, cooperation also occurs at the diminished concentration of Titf1 and Pax8 present in DHTP mice, whereas if both alleles are of B6 origin, cooperation occurs only at the full dosage of the two transcription factors. Hence, according to this model, the modifiers indicated by this study would be transcriptional cofactors that participate in the assembly of the complex containing Titf1 and Pax8. If this were the case, it would be the first example of modifiers involving component(s) of the transcriptional apparatus (35).
The observation that the size of the embryonic thyroid gland is reduced in DHTP/B6 embryos is consistent with a model that implicates Titf1 and Pax8 in the control of thyroid-cell precursor survival and/or proliferation (25) and suggests that these two transcription factors need to cooperate with each other to exert this function.
Interestingly, adult DHTP/B6 mice, even in presence of very high serum TSH levels, display low levels of T4 and possess a thyroid gland that is smaller than that of control mice. One possible explanation for this reduced size of the thyroid gland, even in the presence of high levels of circulating TSH, is that Tshr expression might be down-regulated in our mutants, as already reported in the case of Titf1+/– (9). In DHTP/B6 the amount of Tshr mRNA is, on average, 50% lower than that in wild-type mice, although this difference does not reach statistical significance. It is unlikely that the reduced expression of Tshr we observed in the DHTP/B6 mice is responsible for the thyroid phenotype. Heterozygous loss-of-function Tshr mutations in mice (26) and humans (36) do not display overt hypothyroidism. In addition, the strong expression of NIS (Fig. 3) confirms that the TSH/Tshr pathway is active in DHTP thyroid follicular cells. It is possible that the inability of TSH to rescue thyroid gland size is due to the impaired expression of gene(s), controlled by Titf1 or Pax8, involved in thyroid cell proliferation. It is worth noting that a hypoplastic thyroid in the presence of elevated TSH levels has been reported in a number of patients carrying heterozygous mutations for both TITF1 (7, 8) and PAX8 (11). Interestingly, a patient with a PAX8 mutation presented a normal-sized thyroid (15), underlining the relevance of the genetic background in the establishment of the phenotype consequent to mutations in this gene. Alternatively, the reduced thyroid gland size observed in adult DHTP/B6 mice could be the result of reduced gland growth in embryogenesis and of the consequent reduced pool of follicular cells.
Another important finding of our study is the observation that 30% of DHTP mice displayed hemiagenesis of the thyroid. The process controlling the formation of two symmetrical thyroid lobes, which fails in the case of hemiagenesis, is still unknown. In humans, thyroid hemiagenesis has a prevalence of 0.05% in healthy children (37). The occurrence of hemiagenesis in monozygotic twins (38), among members of the same family (39), or together with other thyroid malformations within one family (40, 41, 42) suggests that genetic factors could be involved in this anomaly. Candidate genes responsible for the hemiagenesis have not yet been described; however, this condition has been reported in some cases of Williams syndrome caused by a deletion at 7q11.23 (43, 44, 45) and in a patient carrying a deletion encompassing the TITF1 locus. Hemiagenesis of the thyroid has been reported in both Hoxa–/– (46, 47) and Shh–/– mouse embryos (48). It is worth noting that in both Hoxa3- and Shh-null mice, hemiagenesis is most likely secondary to defects in the Hoxa3- or Shh-dependent patterning of other tissues, because these two genes are not expressed in the thyroid (48, 49). On the contrary, hemiagenesis in DHTP mice must be the consequence of impairment of thyroid-specific process(es) because Titf1 and Pax8 are expressed in thyroid follicular cells but are absent in neighboring cells. This observation gives additional relevance to the DHTP mice as a model to further our understanding of human congenital dysgenesis. Taken together, the results in Hoxa3–/– and Shh–/– mice with those reported in this study strongly suggest that the determination of the final shape of the thyroid gland requires interactions between autonomous events restricted to the thyroid follicular cells that must interact with an appropriate cellular environment. In humans, hemiagenesis is more frequently associated with subclinical hypothyroidism (50) than with overt hypothyroidism (22, 51); furthermore, hemiagenesis is reported among euthyroid first-degree relatives of patients with congenital hypothyroidism (21). These data suggest that in the absence of other defects, the reduction of 50% of thyroid tissue per se is not sufficient to cause hypothyroidism. It is possible that the frank hypothyroidism displayed by DHTP/B6 mice could result from the decreased number of follicular cells (hypoplasia or hemiagenesis as well) and the marked reduction in Tg biosynthesis with a consequent decrease in thyroid hormone secretion.
Global gene expression was studied in E18 embryonic thyroids from wild-type and DHTP mice in either the B6 or the B6/Sv-F1 genetic backgrounds. The results of such studies showed that the expression of many more genes was affected by the DHTP mutations in the disease-prone B6 background than in the healthy B6/Sv hybrid background. More importantly, this study indicated a group of genes that, in addition to the decreased Tg level, might contribute to the observed phenotype. We suggest that among these genes, there might be those responsible for hypoplasia and hemiagenesis; however, for these aspects of the DHTP phenotype, condition studies of gene expression earlier in organogenesis could be more informative.
In conclusion, the DHTP mouse model demonstrates that TD can be of multigenic origin in mice and strongly suggests that a similar pathogenetic mechanism might be operating in some human patients. Furthermore, the strain specificity observed opens the way to the identification of modifier genes that, in conjunction with Titf1 and Pax8, control morphogenesis, growth, and differentiation of the thyroid gland. Finally, this study demonstrates that, at least in mice, hemiagenesis of the thyroid gland can be explained by genetic mechanisms.
Footnotes
This work was supported, in part, by Telethon, Congenital hypothyroidism with thyroid dysgenesis: candidate genes, animal models and molecular mechanisms; Ministero dell’Universita e della Ricerca Scientifica e Tecnologica, I geni dell’uomo, cluster 01; and Biogem societa coperativa a responsabilita limitata (to A.R. and A.A.).
First Published Online September 8, 2005
Abbreviations: Dio1, Iodothyronine deiodinase type I; E0.5, embryonic d 0.5; NIS, sodium/iodide symporter; TD, thyroid dysgenesis; Tg, thyroglobulin; TPO, thyroperoxidase; TSHR, TSH receptor.
Accepted for publication September 1, 2005.
References
Fisher DA, Klein, A 1981 Thyroid development and disorders of thyroid function in the newborn. N Engl J Med 304:702–712
De Felice M, Di Lauro R 2004 Thyroid development and its disorders: genetics and molecular mechanisms. Endocr Rev 25:722–746
Park SM, Chatterjee VK 2005 Genetics of congenital hypothyroidism. J Med Genet 42:379–389
Perry R, Heinrichs C, Bourdoux P, Khoury K, Szots F, Dussault JH, Vassart G, Van Vliet G 2002 Discordance of monozygotic twins for thyroid dysgenesis: implications for screening and for molecular pathophysiology. J Clin Endocrinol Metab 87:4072–4077
Sunthornthepvarakui T, Gottschalk ME, Hayashi Y, Refetoff S 1995 Brief report: resistance to thyrotropin caused by mutations in the thyrotropin-receptor gene. N Engl J Med 332:155–160
Refetoff S 2003 Resistance to thyrotropin. J Endocrinol Invest 26:770–779
Krude H, Schutz B, Biebermann H, von Moers A, Schnabel D, Neitzel H, Tonnies H, Weise D, Lafferty A, Schwarz S, DeFelice M, von Deimling A, van Landeghem F, DiLauro R, Gruters A 2002 Choreoathetosis, hypothyroidism, and pulmonary alterations due to human NKX2–1 haploinsufficiency. J Clin Invest 109:475–480
Pohlenz J, Dumitrescu A, Zundel D, Martine U, Schonberger W, Koo E, Weiss RE, Cohen RN, Kimura S, Refetoff S 2002 Partial deficiency of thyroid transcription factor 1 produces predominantly neurological defects in humans and mice. J Clin Invest 109:469–473
Moeller LC, Kimura S, Kusakabe T, Liao XH, Van Sande J, Refetoff S 2003 Hypothyroidism in thyroid transcription factor 1 haploinsufficiency is caused by reduced expression of the thyroid stimulating hormone receptor. Mol Endocrinol 17:2295–2302
Doyle DA, Gonzalez I, Thomas B, Scavina M 2004 Autosomal dominant transmission of congenital hypothyroidism, neonatal respiratory distress, and ataxia caused by a mutation of NKX2–1. J Pediatr 145:190–193
Macchia PE, Lapi P, Krude H, Pirro MT, Missero C, Chiovato L, Souabni A, Baserga M, Tassi V, Pinchera A, Fenzi G, Gruters A, Busslinger M, Di Lauro R 1998 PAX8 mutations associated with congenital hypothyroidism caused by thyroid dysgenesis. Nat Genet 19:83–86
Vilain C, Rydlewski C, Duprez L, Heinrichs C, Abramowicz M, Malvaux P, Renneboog B, Parma J, Costagliola S, Vassart G 2001 Autosomal dominant transmission of congenital thyroid hypoplasia due to loss-of-function mutation of PAX8. J Clin Endocrinol Metab 86:234–238
Congdon T, Nguyen LQ, Nogueira CR, Habiby RL, Medeiros-Neto G, Kopp P 2001 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 86:3962–3967
Komatsu M, Takahashi T, Takahashi I, Nakamura M, Takahashi I, Takada G 2001 Thyroid dysgenesis caused by PAX8 mutation: the hypermutability with CpG dinucleotides at codon 31. J Pediatr 139:597–599
Meeus L, Gilbert B, Rydlewski C, Parma J, Roussie AL, Abramowicz M, Vilain C, Christophe D, Costagliola S, Vassart G 2004 Characterization of a novel loss of function mutation of PAX8 in a familial case of congenital hypothyroidism with in-place, normal-sized thyroid. J Clin Endocrinol Metab 89:4285–4291
Clifton-Bligh RJ, Wentworth JM, Heinz P, Crisp MS, John R, LJ, Ludgate M, Chatterjee V 1998 Mutation of the gene encoding human TTF-2 associated with thyroid agenesis, cleft palate and choanal atresia. Nat Genet 19:399–401
Castanet M, Park SM, Smith A, Bost M, Leger J, Lyonnet S, Pelet A, Czernichow P, Chatterjee K, Polak M 2002 A novel loss-of-function mutation in TTF-2 is associated with congenital hypothyroidism, thyroid agenesis and cleft palate. Hum Mol Genet 11:2051–2059
Castanet M, Sura Trueba S, Chauty A, Carre A, Heath, S, Leger J, Lyonnet S, Czernichow P, Polak M 2005 Linkage and mutational analysis of familial thyroid dysgenesis suggest genetic heterogeneity. Eur J Hum Genet 13:232–239
Castanet M, Polak M, Bonaiti-Pellie C, Lyonnet S, Czernichow P, Leger J 2001 Nineteen years of national screening for congenital hypothyroidism: familial cases with thyroid dysgenesis suggest the involvement of genetic factors. J Clin Endocrinol Metab 86:2009–2014
Castanet M, Lyonnet S, Bonaiti-Pellie C, Polak M, Czernichow P, Leger J 2000 Familial forms of thyroid dysgenesis among infants with congenital hypothyroidism. N Engl J Med 343:441–442
Leger J, Marinovic D, Garel C, Bonaiti-Pellie C, Polak M, Czernichow P 2002 Thyroid developmental anomalies in first degree relatives of children with congenital hypothyroidism. J Clin Endocrinol Metab 87:575–580
Mikosch P, Gallowitsch HJ, Kresnik E, Molnar M, Gomez I, Lind P 1999 Thyroid hemiagenesis in an endemic goiter area diagnosed by ultrasonography: report of sixteen patients. Thyroid 9:1075–1084
Kimura S, Hara Y, Pineau T, Fernandez-Salguero P, Fox C, Ward J, Gonzalez F 1996 The T/ebp null mouse: thyroid-specific enhancer-binding protein is essential for the organogenesis of the thyroid, lung, ventral forebrain, and pituitary. Genes Dev 10:60–69
Mansouri A, Chowdhury K, Gruss P 1998 Follicular cells of the thyroid gland require Pax8 gene function. Nat Genet 19:87–90
Parlato R, Rosica A, Rodriguez-Mallon A, Affuso A, Postiglione MP, Arra C, Mansouri A, Kimura S, Di Lauro R, De Felice M 2004 An integrated regulatory network controlling survival and migration in thyroid organogenesis. Dev Biol 276:464–475
Postiglione MP, Parlato R, Rodriguez-Mallon A, Rosica A, Mithbaokar P, Maresca M, Marians RC, Davies TF, Zannini MS, De Felice M, Di Lauro R 2002 Role of the thyroid-stimulating hormone receptor signaling in development and differentiation of the thyroid gland. Proc Natl Acad Sci USA 99:15462–15467
Chomczynski P, Sacchi N 1987 Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 162:156–159
Benjamini Y, Hochberg Y 1995 Controlling the false discovery rate: a practical and powerful approach to multiple testing. J R Stat Soc B 57:289–300
Damante G, Tell G, Di Lauro R 2001 A unique combination of transcription factors controls differentiation of thyroid cells. Prog Nucleic Acid Res Mol Biol 66:307–356
Li W, Reich J 2000 A complete enumeration and classification of two-locus disease models. Hum Hered 50:334–349
De Felice M, Damante G, Zannini MS, Francis-Lang H, Di Lauro R 1995 Redundant domains contribute to the transcriptional activity of Thyroid Transcription Factor 1 (TTF-1). J Biol Chem 270:26649–26656
Zannini MS, Francis-Lang H, Plachov D, Di Lauro R 1992 Pax-8, a paired domain-containing protein, binds to a sequence overlapping the recognition site of a homeodomain and activates transcription from two thyroid-specific promoters. Mol Cell Biol 12:4230–4241
Ohno M, Zannini MS, Levy O, Carrasco N, Di Lauro R 1999 The paired-domain transcription factor Pax8 binds to the upstream enhancer of the rat sodium/iodide symporter gene and participates in both thyroid-specific and cyclic-AMP-dependent transcription. Mol Cell Biol 19:2051–2060
Di Palma T, Nitsch R, Mascia A, Nitsch L, Di Lauro R, Zannini M 2003 The paired domain-containing factor Pax8 and the homeodomain-containing factor TTF-1 directly interact and synergistically activate transcription. J Biol Chem 278:3395–3402
Flint J, Valdar W, Shifman S, Mott R 2005 Strategies for mapping and cloning quantitative trait genes in rodents. Nat Rev Genet 6:271–286
Alberti L, Proverbio MC, Costagliola S, Romoli R, Boldrighini B, Vigone MC, Weber G, Chiumello G, Beck-Peccoz P, Persani L 2002 Germline mutations of TSH receptor gene as cause of nonautoimmune subclinical hypothyroidism. J Clin Endocrinol Metab 2002 87:2549–2555
Maiorana R, Carta A, Floriddia G, Leonardi D, Buscema M, Sava L, Calaciura F, Vigneri R 2003 Thyroid hemiagenesis: prevalence in normal children and effect on thyroid function. J Clin Endocrinol Metab 88:1534–1536
Greig WR, Henderson AS, Boyle JA, McGirr EM, Hutchison JH 1966 Thyroid dysgenesis in two pairs of monozygotic twins and in a mother and child. J Clin Endocrinol Metab 26:1309–1316
Rajmil HO, Rodriguez-Espinosa J, Soldevila J, Ordonez-Llanos J 1984 Thyroid hemiagenesis in two sisters. J Endocrinol Invest 7:393–394
Rosenberg T, Gilboa Y 1980 Familial thyroid ectopy and hemiagenesis. Arch Dis Child 55:639–641
McLean R, Howard N, Murray IP 1985 Thyroid dysgenesis in monozygotic twins: variants identified by scintigraphy. Eur J Nucl Med 10:346–348
Castanet M, Leenhardt L, Leger J, Simon-Carre A, Lyonnet S, Pelet A, Czernichow P, Polak M 2005 Thyroid hemiagenesis is a rare variant of thyroid dysgenesis with a familial component but without Pax8 mutations in a cohort of 22 cases. Pediatr Res 57:908–913
Bini R, Pela I 2004 New case of thyroid dysgenesis and clinical signs of hypothyroidism in Williams syndrome. Am J Med Genet A 127:183–185
Stagi S, Bindi G, Neri AS, Giovannucci-Uzielli ML, Lapi E, Galluzzi F, Salti R 2003 Thyroid hypoplasia of the left lobe in two girls affected by Williams syndrome. Clin Dysmorphol 12:267–268
Cammareri V, Vignati G, Nocera G, Beck-Peccoz P, Persani L 1999 Thyroid hemiagenesis and elevated thyrotropin levels in a child with Williams syndrome. Am J Med Genet 85:491–494
Manley NR, Capecchi M 1995 The role of Hoxa-3 in mouse thymus and thyroid development. Development 121:1989–2003
Manley NR, Capecchi M 1998 Hox group 3 paralogs regulate the development and migration of the thymus, thyroid, and parathyroid glands. Dev Biol 195:1–15
Fagman H, Grande M, Gritli-Linde A, Nilsson M 2004 Genetic deletion of sonic hedgehog causes hemiagenesis and ectopic development of the thyroid in mouse. Am J Pathol 164:1865–1872
Meunier D, Aubin J, Jeannotte L 2003 Perturbed thyroid morphology and transient hypothyroidism symptoms in Hoxa5 mutant mice. Dev Dyn 227:367–378
Calaciura F, Motta RM, Miscio G, Fichera G, Leonardi D, Carta A, Trischitta V, Tassi V, Sava L, Vigneri R 2002 Subclinical hypothyroidism in early childhood: a frequent outcome of transient neonatal hyperthyrotropinemia. J Clin Endocrinol Metab 87:3209–3214
Karabay N, Comlekci A, Canda MS, Bayraktar F, Degirmenci B 2003 Thyroid hemiagenesis with multinodular goiter: a case report and review of the literature. Endocr J 50:409–413(Elena Amendola, Pasquale De Luca, Paolo )