Abstract

Thyroid transcription factor 1 (NKX2-1/TITF1) mutations cause brain–lung–thyroid syndrome, characterized by congenital hypothyroidism (CH), infant respiratory distress syndrome (IRDS) and benign hereditary chorea (BHC). The objectives of the present study were (i) detection of NKX2-1 mutations in patients with CH associated with pneumopathy and/or BHC, (ii) functional analysis of new mutations in vitro and (iii) description of the phenotypic spectrum of brain–lung–thyroid syndrome. We identified three new heterozygous missense mutations (L176V, P202L, Q210P), a splice site mutation (376-2A→G), and one deletion of NKX2-1 at 14q13. Functional analysis of the three missense mutations revealed loss of transactivation capacity on the human thyroglobulin enhancer/promoter. Interestingly, we showed that deficient transcriptional activity of NKX2-1-P202L was completely rescued by cotransfected PAX8-WT, whereas the synergistic effect was abolished by L176V and Q210P. The clinical spectrum of 6 own and 40 published patients with NKX2-1 mutations ranged from the complete triad of brain–lung–thyroid syndrome (50%), brain and thyroid disease (30%), to isolated BHC (13%). Thyroid morphology was normal (55%) and compensated hypothyroidism occurred in 61%. Lung disease occurred in 54% of patients (IRDS at term 76%; recurrent pulmonary infections 24%). On follow-up, 20% developed severe chronic interstitial lung disease, and 16% died. In conclusion, we describe five new NKX2.1 mutations with, for the first time, complete rescue by PAX8 of the deficient transactivating capacity in one case. Additionally, our review shows that the majority of affected patients display neurological and/or thyroidal problems and that, although less frequent, lung disease is responsible for a considerable mortality.

INTRODUCTION

Thyroid transcription factor 1 (NKX2-1, TITF1 or T/EBP) belongs to the homeodomain-containing transcription factor family and has been first identified as a nuclear protein able to bind specific sequences within the thyroglobulin gene (Tg) promoter (1,2). The human NKX2-1 gene is localized on 14q13 (2). During human development, NKX2-1 is first expressed at day 32 on mRNA level in the thyroid bud and in the prosencephalon, giving later rise to the hypothalamus. From 11 gestational weeks on, NKX2-1 mRNA expression is also found in the lung epithelium. Nuclear NKX2-1 staining is detected in thyrocyte precursors at 7 gestational weeks by immunohistochemistry (3,4).

In the thyrocyte, NKX2-1 binds not only to the promoter of TG, but also to the promoters of the thyroid peroxidase gene (TPO), and the pendrin gene (PDS) and results in up-regulation of their transcription (5). NKX2-1 activates transcription of Tg and Tpo in synergy with PAX8 (6,7). As concluded from transgenic animal models NKX2-1 is required for the survival of thyrocyte precursors, for expression of the TSH-receptor (Tshr) and Tg in the differentiating thyrocytes, as well as for the maintenance of ordered architecture and function in the adult thyroid (8–11). In the lung, NKX2-1 expression occurs in pneumocytes type II (12). These cells are producing the surfactant protein essential for reducing alveolar surface tension and local immune defense. NKX2-1 has been shown to bind to the promoters of the surfactant A, B, C and D genes (SFTPA, SFTPB, SFTPC, SFTPD) and regulate their expression positively (13–15). In the brain, NKX2-1 is specifically expressed in neurons of the hypothalamus (16,17) and is involved in interneuron specification and migration during forebrain development (18,19).

In accordance with the human expression pattern, NKX2-1 mutations have been associated with a triad of brain, lung and thyroid disease. However, the classical triad is not always present and severity of benign hereditary chorea (BHC), respiratory symptoms and congenital hypothyroidism (CH) vary largely (20,21).

The aim of this study was first to identify new mutations in the NKX2-1 gene, second, to study their consequence on molecular level and third, to characterize the general phenotypic spectrum of brain–lung–thyroid syndrome by comparing our six patients with the published cases worldwide.

RESULTS

Case reports

Patient 1 was born at term and presented with mild CH detected by neonatal screening and IRDS (Table 1). Thyroid function was compensated when tested after neonatal screening [TSH 11 mU/l, free T4 (fT4) 20 pmol/l]. The thyroid was of normal size on ultrasound and scintigraphy. For IRDS at term she needed ventilation for 8 days. After extubation she was dependent on supplementary oxygen during the first 3 months of life. On follow-up she developed recurrent pulmonary infections needing antibiotic treatment during several months. Atactic movements and psychomotor delay were observed during the first year of life. At the age of 6 years the child remains severely mentally retarded, the ataxia and saccadic movements are non-progressive. Cerebral MRI revealed agenesis of the corpus callosum.

Table 1.

Summary of patient phenotypes in our cohort

 del14q13 376-2A>G L176V P202L Q210P 
 Patient 1 Patient 2 and 3 Monozygotic twins Patient 4 Patient 5 Patient 6 
Thyroid 
CH detected by neonatal screening Yes Yes Yes No Yes Yes 
Thyroid function at diagnosis of CH 
 TSH (mU/l) 11 330 31 7.8 60 290 
 fT4 (pmol/l) 20 13 11 14 
Morphology Normal P2: athyreosis P3: hemiagenesis Hypoplasia Normal Hypoplasia  
Lungs 
IRDS Yes Yes Yes No No No 
Chronic infections Yes Yes No No No No 
CILD No No No No No No 
Neurology 
Hypotonia Yes Yes No Yes Yes Yes 
PM delay Yes Yes No Yes Yes Yes 
BHC 1 year 5 years No 2.5 years 1.5 years 2 years 
MRI Agenesis of corpus callosum Normal na Normal Normal Normal 
 del14q13 376-2A>G L176V P202L Q210P 
 Patient 1 Patient 2 and 3 Monozygotic twins Patient 4 Patient 5 Patient 6 
Thyroid 
CH detected by neonatal screening Yes Yes Yes No Yes Yes 
Thyroid function at diagnosis of CH 
 TSH (mU/l) 11 330 31 7.8 60 290 
 fT4 (pmol/l) 20 13 11 14 
Morphology Normal P2: athyreosis P3: hemiagenesis Hypoplasia Normal Hypoplasia  
Lungs 
IRDS Yes Yes Yes No No No 
Chronic infections Yes Yes No No No No 
CILD No No No No No No 
Neurology 
Hypotonia Yes Yes No Yes Yes Yes 
PM delay Yes Yes No Yes Yes Yes 
BHC 1 year 5 years No 2.5 years 1.5 years 2 years 
MRI Agenesis of corpus callosum Normal na Normal Normal Normal 

na, not applicable; P, patient; IRDS, infant respiratory distress syndrome; CILD, chronic interstitial lung disease; PM, psychomotor delay; BHC, benign hereditary chorea (years indicating onset of BHC).

Patients 2 and 3 are twins born prematurely at 33 gestational weeks. Both twins suffered from IRDS, and were diagnosed by neonatal screening as having CH (Table 1). Twin A had a TSH of 330 mU/l with a fT4 of 1.2 pmol/l. Twin B showed a TSH of 31 mU/l with a fT4 of 13 pmol/l. Twin A showed athyreosis on both, ultrasound and repeated thyroid scans, while twin B presented hemiagenesis. On follow-up, twin A developed recurrent pulmonary infections, while twin B did not suffer from pulmonary problems after the neonatal period. Further, twin A developed hypotonia and delay of gross motor landmarks, progressing to choreoathetosis at the age of 5 years. In contrast, twin B did not show any neurologic symptoms at repeated neurological evaluations until the age of 16 years. Monozygosity was strongly suspected clinically and based on identical results in blood group systems (AB0: A, Rhesus: D, c+, e+; Kell: K; Duffy: Fya; Kidd: Jka; MNSsU: M+, N+, S+, s+), and identical HLA haplotype DQB 0201/0302. Furthermore, analysis of 12 different microsatellite markers covering the locuses of NKX2-1 (ch14q13), PAX8 (ch2q12–14), FOXE1 (ch9q22) and 1 intragenic marker within the TSHR gene (ch14q13) as previously published confirmed allelic identity for the analyzed chromosomal regions (22).

Patient 4 was born at term without signs of respiratory distress (Table 1). Neonatal screening for CH was normal. He was evaluated by a pediatric neurologist because of delayed walking, speech acquisition and choreic movements at the age of 32 months. Cerebral MRI was normal. Repeated TSH measurements during the next years were slightly elevated between 5.8–10.5 mU/l with normal fT4 values. L-thyroxine was started at 6 years of age after slow but progressive increase of TSH. Retrospectively, the TSH value at neonatal screening was 14 mU/l. The thyroid ultrasound showed a normal right, but a hypoplastic left thyroid lobe. TRH test at 3 years of age showed a basal TSH of 6.4 mU/l, a peak of 50 mU/l and a TSH value of 19 mU/l at 120 min. The child showed normal school performance, the choreoathetosis remained stable since the age of 5 years. Repeated episodes of bronchitis were documented until 7 years of age treated by inhalative steroids. Currently at the age of 14 years no respiratory problems are reported.

Patient 5 was detected by neonatal screening as having CH (Table 1). TSH and fT4 values at diagnosis were 60 mU/l and 14 pmol/l. The thyroid gland was described as eutopic with normal size by scintigraphy and ultrasound. Perchlorate discharge test and thyroglobulin were normal. The patient showed hypotonia, delay in acquisition of motor skills and language. At the age of 18 months, she presented with choreoathetotic movements. Cerebral MRI was normal. She shows normal school performance on follow-up. IRDS or repeated respiratory infections were not reported.

Patient 6 presented at birth overt congenital hypothyroidism with elevated TSH (290 mU/l) and low fT4 (6.2 mmol/l) (Table 1). Thyroid scintigraphy showed thyroid hypoplasia. She developed psychomotor delay during the first year of life that evolved to choreoathetosis. Cerebral MRI was normal. She has no history of IRDS at term or recurrent respiratory infections.

Molecular studies

Mutation analysis

Fluorescent in situ hybridization revealed a deletion of the whole NKX2-1 gene in patient 1. Direct sequencing of NKX2-1 revealed an intronic splice site mutation at 376-2A>G in monozygotic twins (patients 2 and 3) described previously (23) and three new missense mutations at L176V (patient 4), P202L (patient 5) and Q210P (patient 6). All three mutations are localized within the homeodomain of NKX2-1 (Fig. 1). All mutations occurred de novo, as none of the parents harbored the NKX2-1 mutations of their children.

Figure 1.

Fluorescent in situ hybridization (FISH) of del14q13 in one patient and NKX2-1 mutation analysis in the five remaining patients of our cohort. NKX2-1 gene: Position of the identified mutations within the NKX2-1 gene is indicated by arrows. HD, homeodomain. NKX2-1 gene defects: FISH showing complete deletion of one NKX2-1 allele on the left and chromatograms representing the already described intronic splice site mutation −2A>G, and the three new mutations L176V, P202L, Q210P on the right. Patients’ phenotypes summarize the combination of signs of brain–lung–thyroid syndrome in each of our six patients. The phenotype of the monozygotic twins with the -2A>G mutation is not identical. +, organ affected; (+), mild CH not detected by neonatal screening; —, organ not affected.

Figure 1.

Fluorescent in situ hybridization (FISH) of del14q13 in one patient and NKX2-1 mutation analysis in the five remaining patients of our cohort. NKX2-1 gene: Position of the identified mutations within the NKX2-1 gene is indicated by arrows. HD, homeodomain. NKX2-1 gene defects: FISH showing complete deletion of one NKX2-1 allele on the left and chromatograms representing the already described intronic splice site mutation −2A>G, and the three new mutations L176V, P202L, Q210P on the right. Patients’ phenotypes summarize the combination of signs of brain–lung–thyroid syndrome in each of our six patients. The phenotype of the monozygotic twins with the -2A>G mutation is not identical. +, organ affected; (+), mild CH not detected by neonatal screening; —, organ not affected.

Transactivation capacity of mutated and wild-type NKX2-1 proteins

To test the functional consequences of the NKX2-1 mutations L176V, P202L and Q210P, we built vectors expressing the mutant NKX2-1 proteins and compared their ability to activate the cotransfected reporter gene under the control of the human Tg-enhancer/promoter in HEK293 cells. Western blotting of whole-cell protein extracts from HEK293 cells, transfected with expression vectors for either NKX2-1-WT or the three mutations revealed reproducibly no difference in molecular weight (42 kDa), or electrophoretic mobility between WT and the mutant NKX2-1 proteins (Fig. 2D). These results indicate that the mutations do not cause destabilization of the mutant proteins.

Figure 2.

Functional activity of mutant versus. wild-type NKX2-1. HEK293 cells were transiently transfected with a reporter gene construct containing the human Tg enhancer/promoter upstream of the luciferase gene, an internal control plasmid BosβGalactosidase, and expression vectors for wild-type (NKX2-1-WT) or mutated NKX2-1 (NKX2-1-L176V, NKX2-1-P202L, NKX2-1-Q210P) alone or together in different ratios (1:1; 1:3) in the absence or in the presence of an expression vector for wild-type PAX8. Empty vector (pcDNA3+) was used as negative control. Promoter induction is expressed as fold induction, relative to the observed in the presence of empty expression vector represented as value 1. Luciferase activity is normalized to BosβGalactosidase activity to adjust for transfection efficiency. Results are mean±SEM of four independent experiments. **P < 0.01; *P < 0.05; n.s., no significant difference. (A) Ability to transactivate the human Tg-enhancer/promoter was tested for NKX2-1-WT and the three mutant NKX2-1 proteins in vitro. All three mutations exhibit abolished or significantly reduced transactivating capacity. (B) Transcriptional capacity of NKX2-1-WT alone and in the presence of mutant NKX2-1 proteins in the ratios 1:1 and 1:3 at constant amount of total DNA. Cotransfection of increasing doses of the three mutant proteins in the presence of NKX2-1-WT revealed a dominant negative effect of NKX2-1-L176V on NKX2-1-WT, but not for NKX2-1-P202L and NKX2-1-Q210P. (C) Transcription activity of NKX2-1-WT and NKX2-1-L176V, NKX2-1-P202L, NKX2-1-Q210P by interacting with cotransfected PAX8-WT on the human Tg-enhancer/promoter. NKX2-1-WT synergizes with PAX8-WT. This synergy is impaired by NKX2-1-L176V and NKX2-1-Q210P, but maintained by NKX2-1-P202L. (D) Equal expression, molecular weight and electrophoretic mobility of NKX2-1-WT or NKX2-1-mutants in all different experiments were assessed by western blot using anti-NKX2-1 and anti-actin antibody.

Figure 2.

Functional activity of mutant versus. wild-type NKX2-1. HEK293 cells were transiently transfected with a reporter gene construct containing the human Tg enhancer/promoter upstream of the luciferase gene, an internal control plasmid BosβGalactosidase, and expression vectors for wild-type (NKX2-1-WT) or mutated NKX2-1 (NKX2-1-L176V, NKX2-1-P202L, NKX2-1-Q210P) alone or together in different ratios (1:1; 1:3) in the absence or in the presence of an expression vector for wild-type PAX8. Empty vector (pcDNA3+) was used as negative control. Promoter induction is expressed as fold induction, relative to the observed in the presence of empty expression vector represented as value 1. Luciferase activity is normalized to BosβGalactosidase activity to adjust for transfection efficiency. Results are mean±SEM of four independent experiments. **P < 0.01; *P < 0.05; n.s., no significant difference. (A) Ability to transactivate the human Tg-enhancer/promoter was tested for NKX2-1-WT and the three mutant NKX2-1 proteins in vitro. All three mutations exhibit abolished or significantly reduced transactivating capacity. (B) Transcriptional capacity of NKX2-1-WT alone and in the presence of mutant NKX2-1 proteins in the ratios 1:1 and 1:3 at constant amount of total DNA. Cotransfection of increasing doses of the three mutant proteins in the presence of NKX2-1-WT revealed a dominant negative effect of NKX2-1-L176V on NKX2-1-WT, but not for NKX2-1-P202L and NKX2-1-Q210P. (C) Transcription activity of NKX2-1-WT and NKX2-1-L176V, NKX2-1-P202L, NKX2-1-Q210P by interacting with cotransfected PAX8-WT on the human Tg-enhancer/promoter. NKX2-1-WT synergizes with PAX8-WT. This synergy is impaired by NKX2-1-L176V and NKX2-1-Q210P, but maintained by NKX2-1-P202L. (D) Equal expression, molecular weight and electrophoretic mobility of NKX2-1-WT or NKX2-1-mutants in all different experiments were assessed by western blot using anti-NKX2-1 and anti-actin antibody.

The three mutated proteins showed all significantly reduced (NKX2-1-P202L 75%, P < 0.01; NKX2-1-Q210P 65%, P < 0.05) or completely abolished (NKX2-1-L176V, P < 0.01) transactivating capacity of the human Tg-enhancer/promoter compared with NKX2-1-WT (Fig. 2A). Cotransfection of increasing doses of the three mutant proteins revealed a dominant negative effect of NKX2-1-L176V on NKX2-1-WT but not for NKX2-1-P202L and NKX2-1-Q210P (Fig. 2B).

To study the functional consequences of the three mutations on synergy of NKX2-1 with PAX8, we performed reconstitution experiments in HEK293 cells by cotransfection of a vector expressing either NKX2-1-WT or one of the three NKX2-1 mutant proteins together with a human PAX8-WT expressing vector. Activity of NKX2-1-WT was significantly higher in synergy with PAX8 than transfected alone (2.0-fold versus 5.6-fold, P < 0.01). In contrast, PAX8 was unable to enhance the activity of NKX2-1-L176V and NKX2-1-Q210P over empty vector baseline (P < 0.01 for both mutant proteins versus NKX2-1-WT). Interestingly, and in contrast to the mutations L176V and Q210P, transcriptional capacity of NKX2-1-P202L did not differ from NKX2-1-WT if cotransfected with PAX8 (6.2-fold versus 5.6-fold, P-value not significant) (Fig. 2C).

Three-dimensional model of the normal and the mutated NKX2-1 proteins

For a better understanding of the functional consequences of the mutations on protein/DNA interaction, we analyzed the three-dimensional structure the NK2 homeodomain in complex with a DNA fragment. The NK2 homeodomain sequence shares 75% identity and 88% homology with the free Nkx2-1 homeodomain. The two protein structures are largely identical, especially the core containing the mutations as well as the mutated residues themselves are conserved.

Figure 3 represents the corresponding mutations of our patients (L176V, P202L and Q210P) and one mutation reported in the literature located nearby (W208L) (24). Most of these mutations are part of helix 3 that crosses through the large DNA-binding groove. Residue Q210 forms a hydrogen bond with a cytosine. When replaced by a proline this hydrogen bond is lost. The loss of the Q210 hydrogen bond can lower the binding of the homeodomain with DNA (patient 6). Residues L176, and W208 constitute the core of the interaction between helix 1 and helix 3. Mutation of either L176 (patient 4) or W208 (24) might have same consequences as they are in close van der Waals contact. Their mutation in valine and leucine, respectively, lead to a clash that can disrupt the three-dimensional structure. It is more difficult to explain the consequence of the mutation P202 in leucine (patient 5). This residue would be in close contact with R191 and E192 and these two long amino acids could easily adapt the mutation. However, as prolines are known as bending amino acids, the proline could be necessary to orient properly helix 3 toward the DNA helix.

Figure 3.

Three-dimensional model of NKX2-1/DNA complex. (A) Three-dimensional model of the normal NK homeodomain/DNA complex with the amino acids from the NKX2-1 sequence. (B) The residues from the normal NKX2-1 protein have been replaced by the mutations described in this study (L176V, P202L and Q210P in red).

Figure 3.

Three-dimensional model of NKX2-1/DNA complex. (A) Three-dimensional model of the normal NK homeodomain/DNA complex with the amino acids from the NKX2-1 sequence. (B) The residues from the normal NKX2-1 protein have been replaced by the mutations described in this study (L176V, P202L and Q210P in red).

Phenotype analysis based on our cohort and all published cases in the literature

In addition to our six patients, we identified 40 published cases with documented genetic analysis, and thyroid function tests (Table 1). Thus, the presented review is based on 46 patients from 28 families. Overall, 7 complete deletions of the NKX2-1 locus and 20 mutations are reported. One mutation (376-2>G) was reported previously and was found in monozygotic twins in our cohort (23).

Inheritance

In 14 families the mutations occurred de novo, while in 12 families, an autosomal dominant inheritance was documented. In two papers, no information on other family members was available.

Combination of signs and symptoms

Twenty-three (50%) of 46 patients developed the complete triad of brain–lung–thyroid syndrome, 14/46 (30%) showed brain and thyroid disease, 6/43 (13%) were diagnosed having isolated benign hereditary chorea. So far 3/46 patients (7%), including patient 3 in our cohort did not develop neurologic signs (25,26). In summary, 43/46 (93%) of patients have neurological problems, 40/46 (87%) hypothyroidism and only 25/46 (54%) respiratory problems (Fig. 4A).

Figure 4.

Spectrum of phenotypes in 46 patients with NKX2-1 mutations. (A) Distribution of the combination of symptoms in 40 published cases (black dots) and our 6 patients (white dots). One patient in the literature died of IRDS at term and three patients of CILD on follow-up (crosses). (B) Relative frequency of different thyroid phenotypes in the general CH population according to the literature (52–54): ectopy 60%, athyreosis 20%, hypoplasia/hemiagenesis 5%, normal thyroid morphology 15%; pie chart), and distribution of thyroid morphology in 25 published patients including four with isolated BHC (black dots) and our six patients (white dots) with NKX2-1 mutations. (C) Correlation of TSH at diagnosis with thyroid morphology in 24 patients. Scatter plot with median TSH values of 13 patients with normal thyroid morphology, 9 patients with hypoplasia and 2 patients with athyreosis (P < 0.05 Kruskal–Wallis, P < 0.05 Mann–Whitney U between normal thyroid and hypoplasia).

Figure 4.

Spectrum of phenotypes in 46 patients with NKX2-1 mutations. (A) Distribution of the combination of symptoms in 40 published cases (black dots) and our 6 patients (white dots). One patient in the literature died of IRDS at term and three patients of CILD on follow-up (crosses). (B) Relative frequency of different thyroid phenotypes in the general CH population according to the literature (52–54): ectopy 60%, athyreosis 20%, hypoplasia/hemiagenesis 5%, normal thyroid morphology 15%; pie chart), and distribution of thyroid morphology in 25 published patients including four with isolated BHC (black dots) and our six patients (white dots) with NKX2-1 mutations. (C) Correlation of TSH at diagnosis with thyroid morphology in 24 patients. Scatter plot with median TSH values of 13 patients with normal thyroid morphology, 9 patients with hypoplasia and 2 patients with athyreosis (P < 0.05 Kruskal–Wallis, P < 0.05 Mann–Whitney U between normal thyroid and hypoplasia).

Thyroid phenotype

Hypothyroidism, overt or subclinical, was present in 40 of 46 patients (87%), while 6 patients (14%) had documented normal thyroid function tests at diagnosis of isolated BHC. CH was diagnosed at birth based on abnormal neonatal screening in 22/40 patients (55%), it was diagnosed after infancy in 18/40 patients (45%), however, the majority being born before neonatal screening program. Peripheral thyroid hormone values were documented only in 28/40 patients with CH. Compensated hypothyroidism at diagnosis was more prevalent (17/28 patients; 61%) than overt hypothyroidism (11/28 patients; 39%). Thyroid morphology was documented in 31/46 patients by ultrasound and/or scintigraphy, while no information was available for 15 patients. In 17/31 patients (55%), the thyroid gland was described as of normal size, including four patients with isolated BHC and normal thyroid function. Hypoplasia or hemiagenesis was reported in 11/31 patients (35%), and athyreosis in 3/31 patients (10%) (Fig. 4B). In patient 2 of our cohort scintigraphy and ultrasound studies failed to reveal any thyroid tissue. The TSH level at diagnosis of CH was available in 13/17 patients with normal thyroid gland (median TSH 16 mU/l), 9/11 patients with hypoplasia (median TSH 49 mU/l) and 2/3 patients with athyreosis (median TSH 582 mU/l) revealing a significant correlation between morphology and TSH at diagnosis (Kruskal–Wallis P < 0.05; Mann–Whitney UP < 0.05 normal thyroid versus hypoplasia) (Fig. 4C). At autopsy, thyroid histology was described as being normal in one patient with macroscopically normal thyroid gland (26).

Lung phenotype

In 25/46 patients (54%) any kind of lung disease (IRDS and/or pulmonary infections) was documented. Initial presentation of lung disease was IRDS at term in 19/25 (76%) patients, most of them needed mechanical ventilation for up to several weeks. In 6/25 patients (24%) frequent mild-to-severe pulmonary infections were the first sign of associated lung disease. On follow-up, 12/19 patients with IRDS also developed recurrent pulmonary infections. Chronic interstitial lung disease (CILD) is documented in 5/18 patients with recurrent pulmonary infections after the neonatal period, 3/5 died of respiratory failure at the age of 8 months, 3 years and 23 years (9,21,27). One patient died at 40th day post-natally of IRDS (26). Mortality of patients with NKX2-1 mutations suffering from lung disease is as high as 16%. No deaths are reported in patients without lung disease. The 23-year-old patient had metastatic lung carcinoma diagnosed on autopsy (21). One of the five CILD patients developed severe ARDS at the age of 10 years (27). One CILD patient’s father with the same NKX2-1 deletion as his son developed CILD at the age of 43 years, which completely resolved after 1 year of corticoids (28). One patient without CILD needed resection of a lung sequester within the first year of life because of repeated severe respiratory infections in infancy. The post-operative course is not documented (20). The first patient with evidence of developmental defects of the lung was recently published. He died of IRDS at 40th day and the autopsy showed pulmonary tissue with low alveolar counts, simplification of the pulmonary architecture with impaired pulmonary branching and morphology suggestive of surfactant deficiency (26).

Neurological phenotype

Thirty-eight (83%) of forty-six patients developed BHC at a median age of 2 years, while five additional patients were reported with hypotonia and developmental delay at an age of 8 months (2 patients), 3 years, 12 years without documented BHC (12%). Chorea was non-progressive, in accordance with the diagnosis of BHC. One patient in the literature (25) and patient 3 in our cohort did not develop any neurologic problems or BHC until the age of 52 years and 16 years respectively. Our patient was repeatedly evaluated by a pediatric neurologist together with his monozygotic twin, who developed relevant BHC. A third patient died at day 40 after birth without documented neurologic signs (26). MRI investigations showed pathological results in 7/39 (18%) patients ranging from dysgenetic basal ganglia (2), cystic mass in the sella (2), cerebral atrophy (3) to hyperdensity of cerebellum (1) and agenesis of corpus callosum (patient 1).

Additional clinical symptoms or malformations in patient with NKX2-1 deletions

In six families, deletions ranging from 0.9 to 17.9 MB on chromosome 14 containing the NKX2-1 gene were documented. Additional non-classical symptoms of brain–lung–thyroid disease were reported in three of six families: hypo or oligodontia (28,29), microcephaly, failure to thrive, growth retardation, unspecific dysmorphic features (27), hypoparathyroidism and malabsorption (28).

DISCUSSION

The association of NKX2-1 mutations and a syndromic form of CH with BHC and/or with neonatal or chronic lung diseases is well established. The term brain–lung–thyroid syndrome has been introduced by Willemsen et al. (21). Therefore, although defects in the NKX2-1 gene are rare, the specific triad of endocrine, neurological and respiratory signs is highly suggestive for NKX2-1 anomalies (20,27,30–34).

By systematic screening for NKX2-1 mutations in 41 patients with the combination CH, neurological and/or respiratory problems, we identified three new missense mutations (L176V, P202L and Q210P), one splice site mutation already described and one complete deletion of the NKX2-1 gene. All NKX2-1 mutations were present in the heterozygous state, in accordance with NKX2-1 haploinsufficiency, and occurred de novo. The three new missense mutations are all located within the DNA-binding domain.

Our functional studies show that NKX2-1-L176V, NKX2-1-P202L and NKX2-1-Q210P proteins cause complete or partial loss of functional activity on the human Tg-enhancer/promoter in non-thyroidal cells in vitro. Additionally, a dominant negative effect on transcriptional activity of NKX2-1-WT was seen for NKX2-1-L176V. These in vitro results strongly support the role of these loss-of-function mutations for the disease in our patients.

Thyrocyte specific target genes of NKX2-1 as Tg, and Tpo are transactivated in synergy with PAX8 (6,7). Further, NKX2-1 as well as PAX8 use p300, an ubiquitious transcriptional cofactor, as coactivator, and p300 may play an important role for NKX2-1/PAX8 synergy (35,36). NKX2-1 and PAX8 are only co-expressed in thyrocytes, but neither in type II pneumocytes nor in neurons of the hypothalamus. Thus, it has been hypothesized, that the generally mild form of CH and thyroid dysgenesis due to NKX2-1 mutations could be a consequence of partial compensation of NKX2-1 haploinsufficiency by PAX8 synergy. However until now, no functional data were available supporting this hypothesis. In reconstitution experiments in the non-thyroidal HEK293 cell line, we found that synergy with cotransfected PAX8-WT was completely abolished by NKX2-1-L176V and NKX2-1-Q210P. In contrast, NKX2-1-P202L retained its ability to synergize with PAX8 on the Tg-enhancer/promoter. Thus, on one hand, our results confirmed previous reports on complete or partial loss of NKX2-1/PAX8 synergy for mutations of NKX2-1 (25) and PAX8 (36,37). On the other hand, our data provide, for the first time, experimental evidence that synergy with PAX8 can completely overcome the functional defect of a specific NKX2-1 mutation on the Tg promoter. As HEK293 cells are deficient in p300 (38), it remains to be shown, whether cotransfected p300 could rescue NKX2-1/PAX8 synergism of NKX2-1-L176V and NKX2-1-Q210P mutations. However, the retained synergy of NKX2-1-P202L with PAX8 seems to be independent of p300 in in vitro conditions very similar as used previously (36). Three-dimensional modeling of the WT and mutant NKX2-1 protein in silico provide structural arguments in support of our in vitro data: the mutations L176V and Q210P are expected either to cause disrupted three-dimensional structure of the NKX2-1 protein or impaired binding capacity of the NKX2-1 homeodomain, while the effect of the P202L mutation on protein structure is expected to be compensated by residues in close vicinity to the mutation.

Are these functional data compatible with the patient’s thyroid function and morphology? Several methodological and biological arguments speak against a direct correlation of in vitro data with patient’s phenotypes. First, HEK293 cells do not represent a thyrocyte cell context, and lack p300. This model allows specific analysis of the NKX2-1/PAX8 synergy in the absence of p300 and thyrocyte specific cofactors. However, this methodological advantage in vitro is unable to reproduce the complexity of possible interactions of NKX2-1 with multiple target genes in differentiating and mature thyrocytes. Second, it has been shown in Nkx2-1+/− mice that Nkx2-1 haploinsufficiency is mainly caused by reduced expression of Tshr and probably to a lesser degree due to reduced Tg expression (9). Thus, functional experiments on the Tg-promoter give important insights into synergy of mutant NKX2-1 with PAX8, but cover only one known aspect of hypothyroidism due to NKX2-1 defects. Third, it is assumed that patients with NKX2-1 mutations and hypoplasia have in addition to disturbed function some degree of impaired thyroid development and differentiation.

In summary, our in vitro and in silico data provide (i) strong arguments that the identified mutations cause the disease in the patients and (ii) experimental evidence supporting the hypothesis that the functional defect of specific NKX2-1 mutations could completely be compensated by synergy with PAX8.

The aim of our review of all published cases with NKX2-1 mutations was to better characterize the diversity as well as the main diagnostic features of brain–lung–thyroid syndrome in infancy, childhood and in adults. The combination of neurological problems and hypothyroidism are the leading signs present in 80% of patients, while only 50% develop the complete triad of the syndrome. Isolated BHC has been reliably shown to present the mildest expression of the syndrome present in only 13% of patients. However, a significant number (n = 33) of patients with isolated BHC were excluded from our review due to lacking information on thyroid and lung disease (24). Thus, isolated BHC may be underestimated by our review. Extended phenotypes with additional non-specific malformations have been found in patients with large deletions on chromosome 14 containing further contiguous genes as, e.g. PAX9 (27,29).

We confirm the notion that NKX2-1 anomalies result in a generally mild form of thyroid dysgenesis and CH. Patients exhibit normal thyroid morphology in 55% and hypoplasia or hemiagenesis in 35%. CH was compensated in as much as 61% of patients, while only 39% had overt CH. Normal thyroid morphology was associated with significantly milder CH than hypoplasia. It is noteworthy, that some patients were not detected by neonatal screening but presented with hypothyroidism later in life. These results are in accordance with data from transgenic murine models: besides its crucial role for thyroid development, Nkx2-1 in the adult murine thyroid is required for maintenance of ordered follicular architecture and function of the differentiated thyroid (11).

Non-progressive BHC is the most common and most specific sign for NKX2-1 gene anomalies. However, neurological symptoms might not be present at birth, and BHC is preceded by hypotonia and psychomotor delay during the first year of life. Cerebral anomalies are not a common feature of the syndrome.

IRDS at term and recurrent respiratory infections are the less frequent hallmark of the triad. However, our review revealed for the first time a considerable mortality rate of these patients at any age. The majority of NKX2-1 mutations result in congenital surfactant deficiency syndrome, as NKX2-1 regulates the expression of the surfactant proteins A, B, C and D in the lung epithelium (13–15). A minority of patients developed recurrent pulmonary infections of mild-to-severe degree without IRDS at term. An unexpectedly high proportion of patients died either of IRDS or of CILD after recurrent pulmonary infections during infancy, childhood or as young adults. NKX2-1 also is involved in lung organogenesis (12,16). NKX2-1 haploinsufficiency was associated with general developmental defects of the lung in one infant (26), and one patient was described with a lung sequester (20). Recent data propose NKX2-1 as a new lineage specific oncogene of lung cancer and showed amplification of NKX2-1 expression in lung cancers (39). In our review one patient who died from CILD presented at autopsy large cell carcinoma of the lung. However, no other patient was reported with lung cancer, even in affected multigeneration families.

The diagnostic value and prognostic consequence of disorders of the three different organs affected can be summarized as follows: as CH is screened systematically in newborns it is an important element to raise suspicion of NKX2-1 defects in the context of neurological or respiratory problems. At birth, affected newborns will present only hypotonia. However, on follow-up BHC is the most specific and most frequent sign of brain–lung–thyroid syndrome and responsible for life-long morbidity of variable degree. Finally, lung disease as IRDS at term or recurrent infection is the less frequent sign of brain–lung–thyroid syndrome but is responsible for a considerable mortality of patients with NKX2-1 anomalies if evolving to CILD.

Finally, disease phenotype and severity varies considerably even within families with the same NKX2-1 mutation. In our cohort, monozygotic twins (patients 2 and 3) are concordant for CH, but discordant for the type of thyroid dysgenesis, as well as for the occurrence of brain disease and recurrent pulmonary infections. It is further of interest that most of monozygotic twin pairs were shown to be discordant for CH (40).

In summary, our data add the following aspects to current knowledge: first, we identified a complete deletion of NKX2-1, an intronic splice site mutation and three new NKX2-1 missense mutations within the homeodomain. The missense mutations result in impaired transactivating activity of mutated NKX2-1. Second, our reconstitution experiments in vitro suggest for the first time experimental evidence that the synergistic effect of PAX8 can completely overcome the functional defect of a specific NKX2-1 mutation. Whether such compensation by PAX8 is associated with mild CH phenotype in other patients with NKX2-1 anomalies remains to be shown for further mutations. Third, our review provides a detailed analysis of the clinical diversity of NKX2-1 defects at birth, and the evolution of the disease until adulthood. These data should improve early diagnosis of affected patients and counseling of their families.

MATERIALS AND METHODS

Patients

Forty-one patients with CH in combination with either neurological symptoms (hypotonia, psychomotor delay, choreic movements) or respiratory symptoms [infant respiratory distress syndrome (IRDS) at term, recurrent respiratory infections or CILD] were included in the study. The study was approved by the institutional review board.

Mutational screening

After obtaining informed consent, DNA was extracted from peripheral blood leukocytes according to the manufacturer’s instructions (FlexiGene DNA Kit, Qiagen, Hilden, Germany). Genomic DNA was amplified by PCR (Invitrogen, Carlsbad, CA). The following primers were used to sequence the homeodomain: NKX2-1 F4 5′-GCGCGGAAAACAGGGGTGGC-3′ and NKX2-1 R4 5′-GCTGCGCCGCCTTGTCCTTG-3′. The amplicons (373 base pairs) were purified, directly sequenced with the ‘Big Dye Terminator v1.1 Cycle Sequencing’ kit (Applied Biosystems, Foster City, CA) and an ABI Prism 377 automatic sequencer (Perkin Elmer, Wellesleg, USA). The rest of the NKX2-1 gene was sequenced in the same manner (primers available on request).

Mutagenesis, constructs and transient transfections

cDNA sequence of NKX2-1 contained in pSG5-hNKX2-1 (a gift of S. Refetoff) was subcloned in pcDNA3+ (Invitrogen, Carlsbad, CA). The mutations were introduced by site-directed mutagenesis (QuickChange, Stratagene, La Jolla, CA). Human kidney epithelial cells, HEK293 (ATCC, CRL-1573) were grown in DMEM (Dulbecco’s modified Eagle medium) supplemented with 10% fetal bovine serum (Biowest, Nuaillé, France) and 1% penicillin and streptomycin (Invitrogen, Carlsbad, CA). For luciferase and beta-galactosidase assays, a reporter gene construct containing NKX2-1 and PAX8 binding sites of the human Tg enhancer/promoter (given by S. Refetoff) upstream of the luciferase gene was used. We plated the HEK293 cells at a density of 2 × 105 per well in 24-well plate 24 h before transfection.

We transiently cotransfected cells with 520 ng reporter gene, 130 ng hNKX2-1-pcDNA3, 130 ng hPAX8-pcDNA3 (given by G. Vassart) and 150 ng plasmid BosβGalactosidase (given by K. Chatterjee). To test for dominant negative effect of mutant NKX2-1 proteins, WT and mutant hNKX2-1-pcDNA3 were cotransfected in ratios 1:1 (65 ng each) and 1:3 (32 ng WT and 98 ng mutant hNKX2-1pcDNA3). Cotransfection was carried out by a 6-h exposure to 3 µl of lipofectamine in Opti-Mem medium (Invitrogen). After 48 h, cells were harvested and luciferase and beta-galactosidase assays were performed. Luciferase values were normalized to beta-galactosidase activity from the internal control plasmid BosβGalactosidase (41) and represent the mean±SEM of four independent experiments, each performed in triplicate.

Western blotting

For western blot, cells (5 × 106) were plated on 100-mm diameter culture dish 24 h before transfection. Cells were transfected with 4 µg hPAX8-pcDNA3 and 4 µg hNKX2-1-pcDNA3 with 20 µl lipofectamine or cotransfected with 4 µg of each plasmid with 30 µl lipofectamine.

Forty-eight hours after transfection, proteins were quantified on the supernatant with the DC Protein Assay (BioRad, Marne la Coquette, France). Fifteen micrograms of total proteins were loaded on a 10% SDS–PAGE and electroblotted onto a Hybond ECL membrane (Amersham Bioscience, Fairfield, USA). Membranes were probed with anti-NKX2-1 antibody diluted 1:1000 (Dako-Cytomation, Glostrup, Danemark), anti-PAX8 antibody diluted 1:2000 (given by G. Vassart) and anti-beta-actin diluted 1:5000 (Sigma-Aldrich Corp., St Louis, MO). Horseradish peroxidase-conjugated swine anti-rabbit antibody (Dako-Cytomation, Glostrup, Denmark) and sheep anti-mouse antibody were used as secondary antibodies, respectively. Bound antibodies were revealed with a chemiluminescence Kit (Amersham Bioscience, Fairfield, USA).

Three-dimensional modeling of nkx2-1

The sequence of nkx2-1 homeodomain (161–227) was submitted to a Blast search (http://blast.wustl.edu) using the Brookhaven Protein Data Bank (http://www.rcsb.org/index.html) as database. This search resulted in several NMR structures. Two of them were used in this study: the structure of the homeodomain of nkx2-1 itself (PDB code 1FTT) (42) and the structure of the homeodomain of vnd/NK2 in complex with a DNA fragment of 16 nucleic acids on each strand (PDB code 1NK2) (43). The modeling of the mutants and the analysis were performed using the PyMOL Molecular Graphics System (DeLano Scientific: http://pymol.sourceforge.net/).

Review of the literature

We performed a systematic review of the literature by a MEDLINE search using the key words ‘congenital hypothyroidism’, ‘benign hereditary chorea’, ‘thyroid transcription factor 1’, ‘NKX2-1’, ‘TITF1’, ‘TTF1’ and ‘mutation’. Inclusion criteria for the analysis were (i) patients with documented mutational analysis and (ii) individual clinical data on thyroid, lung and neurologic disorders. All patients with NKX2-1 gene defects reported in the literature were reviewed. Clinical features (age at diagnosis of CH, IRDS, CILD and BHC), data on laboratory and imaging investigations (result of neonatal screening, TSH, T4, T3, thyroid ultrasound or scintigraphy, cerebral MRI) were entered in a database. All patients with isolated BHC without documented thyroid function were excluded (24). One family was reported in two papers but included only once (17,43). By this approach we included 40 patients from 23 families reported in 19 papers until 31 October 2008 (9,20,21,23,25–29,32,33,44–51).

Statistical analysis

Mann–Whitney U and Kruskal–Wallis tests (SAS software, version 9.1 SAS Institute, Cary, NC) were used to analyze differences of transactivation activities and TSH values at diagnosis of CH. Bar graphs represent mean±SEM.

FUNDING

A.C. and S.S.-T. were supported by Convention Industrielle de Formation par la Recherche in collaboration with HRA Pharma directed by Dr André Ulmann and the Ministère de l’Education Nationale et de la Recherche. G.S. was supported by grants from the Swiss National Research Foundation (PBBSB-101027), the Margarete and Walter Lichtenstein Stiftung, Basel, Switzerland and by an awarded grant from the Fondation Endocrinologie Genève, Geneva, Switzerland. M.C. was supported by the ‘leem research’.

ACKNOWLEDGEMENTS

We thank all members of the affected families for their collaborative participation in this study. We are grateful to Dr Samuel Refetoff (University of Chicago, Chicago, IL) for human pSG5TTF1, pTGenh/prom-Luc, to Dr Gilbert Vassart (Université Libre de Bruxelles, Brussels, Belgium) for human PAX8 expression vector and anti-PAX8 antibody and to Dr Krishna Chatterjee (University of Cambridge, Cambridge, UK) for BosβGalactosidase plasmid. We thank Marie-Ange Delrue (University of Bordeaux, France) for helpful discussions. We thank Pr Paul Czernichow (Hôpital Necker Enfants-Malades, Paris, France) for his longstanding and continuous support of our projects on congenital hypothyroidism.

Conflict of Interest statement. None declared.

REFERENCES

1
Civitareale
D.
Lonigro
R.
Sinclair
A.J.
Di Lauro
R.
A thyroid-specific nuclear protein essential for tissue-specific expression of the thyroglobulin promoter
EMBO J.
 , 
1989
, vol. 
8
 (pg. 
2537
-
2542
)
2
Guazzi
S.
Price
M.
De Felice
M.
Damante
G.
Mattei
M.G.
Di Lauro
R.
Thyroid nuclear factor 1 (TTF-1) contains a homeodomain and displays a novel DNA binding specificity
EMBO J.
 , 
1990
, vol. 
9
 (pg. 
3631
-
3639
)
3
Trueba
S.
Auge
J.
Mattei
G.
Etchevers
H.
Martinovic
J.
Czernichow
P.
Vekemans
M.
Polak
M.
Attie-Bitach
T.
PAX8, TITF1 and FOXE1 gene expression patterns during human development: new insights into human thyroid development and thyroid dysgenesis-associated malformations
J. Clin. Endocrinol. Metab.
 , 
2005
, vol. 
90
 (pg. 
455
-
462
)
4
Szinnai
G.
Lacroix
L.
Carre
A.
Guimiot
F.
Talbot
M.
Martinovic
J.
Delezoide
A.L.
Vekemans
M.
Michiels
S.
Caillou
B.
, et al.  . 
Sodium/iodide symporter (NIS) gene expression is the limiting step for the onset of thyroid function in the human fetus
J. Clin. Endocrinol. Metab.
 , 
2007
, vol. 
92
 (pg. 
70
-
76
)
5
Dentice
M.
Luongo
C.
Elefante
A.
Ambrosio
R.
Salzano
S.
Zannini
M.
Nitsch
R.
Di Lauro
R.
Rossi
G.
Fenzi
G.
, et al.  . 
Pendrin is a novel in vivo downstream target gene of the TTF-1/Nkx-2.1 homeodomain transcription factor in differentiated thyroid cells
Mol. Cell. Biol.
 , 
2005
, vol. 
25
 (pg. 
10171
-
10182
)
6
Di Palma
T.
Nitsch
R.
Mascia
A.
Nitsch
L.
Di Lauro
R.
Zannini
M.
The paired domain-containing factor Pax8 and the homeodomain-containing factor TTF-1 directly interact and synergistically activate transcription
J. Biol. Chem.
 , 
2003
, vol. 
278
 (pg. 
3395
-
3402
)
7
Miccadei
S.
De Leo
R.
Zammarachi
E.
Natali
P.G.
Civitareale
D.
The synergistic activity of thyroidtranscription factor 1 and Pax 8 relies on the promotor/enhancer interplay
Mol. Endocrinol.
 , 
2002
, vol. 
16
 (pg. 
837
-
846
)
8
Kimura
S.
Hara
Y.
Pineau
T.
Fernandez-Salguero
P.
Fox
C.H.
Ward
J.M.
Gonzalez
F.J.
The T/ebp null mouse: Thyroid-specific enhancer-binding protein is essential for the organogenesis of the thyroid, lung, ventral forebrain
Genes Dev.
 , 
1996
, vol. 
10
 (pg. 
60
-
69
)
9
Moeller
L.C.
Kimura
S.
Kusakabe
T.
Liao
X.H.
Van Sande
J.
Refetoff
S.
Hypothyroidism in thyroid transcription factor 1 haploinsufficiency is caused by reduced expression of the thyroid-stimulating hormone receptor
Mol. Endocrinol.
 , 
2003
, vol. 
17
 (pg. 
2295
-
2302
)
10
De Felice
M.
Di Lauro
R.
Thyroid development and its disorders: genetics and molecular mechanisms
Endocr. Rev.
 , 
2004
, vol. 
25
 (pg. 
722
-
746
)
11
Kusakabe
T.
Kawaguchi
A.
Hoshi
N.
Kawaguchi
R.
Hoshi
S.
Kimura
S.
Thyroid-specific enhancer-binding protein/NKX2.1 is required for the maintenance of ordered architecture and function of the differentiated thyroid
Mol. Endocrinol.
 , 
2006
, vol. 
20
 (pg. 
1796
-
1809
)
12
Maeda
Y.
Dave
V.
Whitsett
J.A.
Transcriptional control of lung morphogenesis
Physiol. Rev.
 , 
2007
, vol. 
87
 (pg. 
219
-
244
)
13
Bohinski
R.J.
Di Lauro
R.
Whitsett
J.A.
The lung-specific surfactant protein B gene promoter is a target for thyroid transcription factor 1 and hepatocyte nuclear factor 3, indicating common factors for organ-specific gene expression along the foregut axis
Mol. Cell. Biol.
 , 
1994
, vol. 
14
 (pg. 
5671
-
5681
)
14
Yi
M.
Tong
G.X.
Murry
B.
Mendelson
C.R.
Role of CBP/p300 and SRC-1 in transcriptional regulation of the pulmonary surfactant protein-A (SP-A) gene by thyroid transcription factor-1 (TTF-1)
J. Biol. Chem.
 , 
2002
, vol. 
277
 (pg. 
2997
-
3005
)
15
Liu
C.
Glasser
S.W.
Wan
H.
Whitsett
J.A.
GATA-6 and thyroid transcription factor-1 directly interact and regulate surfactant protein-C gene expression
J. Biol. Chem.
 , 
2002
, vol. 
277
 (pg. 
4519
-
4525
)
16
Lazzaro
D.
Price
M.
de Felice
M.
Di Lauro
R.
The transcription factor TTF-1 is expressed at the onset of thyroid and lung morphogenesis and in restricted regions of the foetal brain
Development
 , 
1991
, vol. 
113
 (pg. 
1093
-
1104
)
17
Kleiner-Fisman
G.
Calingasan
N.Y.
Putt
M.
Chen
J.
Beal
M.F.
Lang
A.E.
Alterations of striatal neurons in benign hereditary chorea
Mov. Disord.
 , 
2005
, vol. 
20
 (pg. 
1353
-
1357
)
18
Butt
S.J.
Sousa
V.H.
Fuccillo
M.V.
Hjerling-Leffler
J.
Miyoshi
G.
Kimura
S.
Fishell
G.
The requirement of Nkx2–1 in the temporal specification of cortical interneuron subtypes
Neuron
 , 
2008
, vol. 
59
 (pg. 
722
-
732
)
19
Nobrega-Pereira
S.
Kessaris
N.
Du
T.
Kimura
S.
Anderson
S.A.
Marin
O.
Postmitotic Nkx2–1 controls the migration of telencephalic interneurons by direct repression of guidance receptors
Neuron
 , 
2008
, vol. 
59
 (pg. 
733
-
745
)
20
Krude
H.
Schuetz
B.
Biebermann
H.
von Moers
A.
Schnabel
D.
Neitzel
H.
Tönnies
H.
Weise
D.
Lafferty
A.
Schwarz
S.
, et al.  . 
Choreoathetosis, Hypothyroidism, and pulmonary alterations due to human NKX2.1 haploinsufficiency
J. Clin. Invest.
 , 
2002
, vol. 
109
 (pg. 
475
-
480
)
21
Willemsen
M.A.
Breedveld
G.J.
Wouda
S.
Otten
B.J.
Yntema
J.L.
Lammens
M.
de Vries
B.B.
Brain-Thyroid-Lung syndrome: a patient with a severe multi-system disorder due to a de novo mutation in the thyroid transcription factor 1 gene
Eur. J. Pediatr.
 , 
2005
, vol. 
164
 (pg. 
28
-
30
)
22
Castanet
M.
Sura-Trueba
S.
Chauty
A.
Carre
A.
de Roux
N.
Heath
S.
Leger
J.
Lyonnet
S.
Czernichow
P.
Polak
M.
Linkage and mutational analysis of familial thyroid dysgenesis demonstrate genetic heterogeneity implicating novel genes
Eur. J. Hum. Genet.
 , 
2005
, vol. 
13
 (pg. 
232
-
239
)
23
Doyle
D.A.
Gonzalez
I.
Thomas
B.
Scavina
M.
Autosomal dominant transmission of congenital hypothyroidism, neonatal respiratory distress, and ataxia caused by a mutation of NKX2–1
J. Pediatr.
 , 
2004
, vol. 
145
 (pg. 
190
-
193
)
24
Breedveld
G.J.
van Dongen
J.W.
Danesino
C.
Guala
A.
Percy
A.K.
Dure
L.S.
Harper
P.
Lazarou
L.P.
van der Linde
H.
Joosse
M.
, et al.  . 
Mutations in TITF-1 are associated with benign hereditary chorea
Hum. Mol. Genet.
 , 
2002
, vol. 
11
 (pg. 
971
-
979
)
25
Moya
C.M.
Perez de Nanclares
G.
Castano
L.
Potau
N.
Bilbao
J.R.
Carrascosa
A.
Bargada
M.
Coya
R.
Martul
P.
Vicens-Calvet
E.
, et al.  . 
Functional study of a novel single deletion in the TITF1/NKX2.1 homeobox gene that produces congenital hypothyroidism and benign chorea but not pulmonary distress
J. Clin. Endocrinol. Metab.
 , 
2006
, vol. 
91
 (pg. 
1832
-
1841
)
26
Maquet
E.
Costagliola
S.
Parma
J.
Christophe-Hobertus
C.
Oligny
L.L.
Fournet
J.C.
Robitaille
Y.
Vuissoz
J.M.
Payot
A.
Laberge
S.
, et al.  . 
Lethal respiratory failure and mild primary hypothyroidism in a term girl with a de novo heterozygous mutation in the TITF1/NKX2.1 gene
J. Clin. Endocrinol. Metab.
 , 
2009
, vol. 
94
 (pg. 
197
-
203
)
27
Iwatani
N.
Mabe
H.
Devriendt
K.
Kodama
M.
Mike
T.
Deletion of NKX2.1 gene encoding thyroid transcription factor-1 in two siblings with hypothyroidism and respiratory failure
J. Pediatr.
 , 
2000
, vol. 
137
 (pg. 
272
-
276
)
28
Devos
D.
Vuillaume
I.
de Becdelievre
A.
de Martinville
B.
Dhaenens
C.M.
Cuvellier
J.C.
Cuisset
J.M.
Vallee
L.
Lemaitre
M.P.
Bourteel
H.
, et al.  . 
New syndromic form of benign hereditary chorea is associated with a deletion of TITF-1 and PAX-9 contiguous genes
Mov. Disord.
 , 
2006
, vol. 
21
 (pg. 
2237
-
2240
)
29
Guala
A.
Falco
V.
Breedveld
G.
De Filippi
P.
Danesino
C.
Deletion of PAX9 and oligodontia: a third family and review of the literature
Int. J. Paediatr. Dent.
 , 
2008
, vol. 
18
 (pg. 
441
-
445
)
30
Perna
M.G.
Civitareale
D.
De Filippis
V.
Sacco
M.
Cisternino
C.
Tassi
V.
Absence of mutations in the gene encoding thyroid transcription factor-1 (TTF-1) in patients with thyroid dysgenesis
Thyroid
 , 
1997
, vol. 
7
 (pg. 
377
-
381
)
31
Lapi
P.
Macchia
P.E.
Chiovato
L.
Biffali
E.
Moschini
L.
Larizza
D.
Baserga
M.
Pinchera
A.
Fenzi
G.
Di Lauro
R.
Mutations in the gene encoding Thyroid Transcription Factor-1 (TTF-1) are not a frequent cause of congenital hypothyroidism (CH) with thyroid dysgenesis
Thyroid
 , 
1997
, vol. 
7
 (pg. 
383
-
387
)
32
Devriendt
K.
Vanhole
C.
Matthis
G.
de Zegher
F.
Deletion of Thyroid Transcription Factor 1 gene in an infant with neonatal thyroid dysfunction and respiratory failure
N. Engl. J. Med.
 , 
1998
, vol. 
338
 (pg. 
1317
-
1318
)
33
Pohlenz
J.
Dumitrescu
A.
Zundel
D.
Martiné
U.
Schönberger
W.
Koo
E.
Weiss
R.E.
Cohen
R.N.
Kimura
S.
Refetoff
S.
Partial deficiency of thyroid transcription factor 1 produces predominantly neurological defects in humans and mice
J. Clin. Invest.
 , 
2002
, vol. 
109
 (pg. 
469
-
473
)
34
Al Taji
E.
Biebermann
H.
Limanova
Z.
Hnikova
O.
Zikmund
J.
Dame
C.
Gruters
A.
Lebl
J.
Krude
H.
Screening for mutations in transcription factors in a Czech cohort of 170 patients with congenital and early-onset hypothyroidism: identification of a novel PAX8 mutation in dominantly inherited early-onset non-autoimmune hypothyroidism
Eur. J. Endocrinol.
 , 
2007
, vol. 
156
 (pg. 
521
-
529
)
35
De Leo
R.
Miccadei
S.
Zammarchi
E.
Civitareale
D.
Role for p300 in Pax 8 induction of thyroperoxidase gene expression
J. Biol. Chem.
 , 
2000
, vol. 
275
 (pg. 
34100
-
34105
)
36
Grasberger
H.
Ringkananont
U.
Lefrancois
P.
Abramowicz
M.
Vassart
G.
Refetoff
S.
Thyroid transcription factor 1 rescues PAX8/p300 synergism impaired by a natural PAX8 paired domain mutation with dominant negative activity
Mol. Endocrinol.
 , 
2005
, vol. 
19
 (pg. 
1779
-
1791
)
37
Meeus
L.
Gilbert
B.
Rydlewski
C.
Parma
J.
Roussie
A.L.
Abramowicz
M.
Vilain
C.
Christophe
D.
Costagliola
S.
Vassart
G.
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.
 , 
2004
, vol. 
89
 (pg. 
4285
-
4291
)
38
Svensson
C.
Akusjarvi
G.
Adenovirus 2 early region 1A stimulates expression of both viral and cellular genes
EMBO J.
 , 
1984
, vol. 
3
 (pg. 
789
-
794
)
39
Kwei
K.A.
Kim
Y.H.
Girard
L.
Kao
J.
Pacyna-Gengelbach
M.
Salari
K.
Lee
J.
Choi
Y.L.
Sato
M.
Wang
P.
, et al.  . 
Genomic profiling identifies TITF1 as a lineage-specific oncogene amplified in lung cancer
Oncogene
 , 
2008
, vol. 
27
 (pg. 
3635
-
3640
)
40
Perry
R.
Heinrichs
C.
Bourdoux
P.
Khoury
K.
Szots
F.
Dussault
J.H.
Vassart
G.
Van Vliet
G.
Discordance of monozygotic twins for thyroid dysgenesis: implications for screening and for molecular pathophysiology
J. Clin. Endocrinol. Metab.
 , 
2002
, vol. 
87
 (pg. 
4072
-
4077
)
41
Collingwood
T.N.
Adams
M.
Tone
Y.
Chatterjee
V.K.
Spectrum of transcriptional, dimerization, and dominant negative properties of twenty different mutant thyroid hormone beta-receptors in thyroid hormone resistance syndrome
Mol. Endocrinol.
 , 
1994
, vol. 
8
 (pg. 
1262
-
1277
)
42
Esposito
G.
Fogolari
F.
Damante
G.
Formisano
S.
Tell
G.
Leonardi
A.
Di Lauro
R.
Viglino
P.
Analysis of the solution structure of the homeodomain of rat thyroid transcription factor 1 by 1H-NMR spectroscopy and restrained molecular mechanics
Eur. J. Biochem.
 , 
1996
, vol. 
241
 (pg. 
101
-
113
)
43
Gruschus
J.M.
Tsao
D.H.
Wang
L.H.
Nirenberg
M.
Ferretti
J.A.
Interactions of the vnd/NK-2 homeodomain with DNA by nuclear magnetic resonance spectroscopy: basis of binding specificity
Biochemistry
 , 
1997
, vol. 
36
 (pg. 
5372
-
5380
)
44
Kleiner-Fisman
G.
Rogaeva
E.
Halliday
W.
Houle
S.
Kawarai
T.
Sato
C.
Medeiros
H.
St George-Hyslop
P.H.
Lang
A.E.
Benign hereditary chorea: clinical, genetic, and pathological findings
Ann. Neurol.
 , 
2003
, vol. 
54
 (pg. 
244
-
247
)
45
Asmus
F.
Horber
V.
Pohlenz
J.
Schwabe
D.
Zimprich
A.
Munz
M.
Schoning
M.
Gasser
T.
A novel TITF-1 mutation causes benign hereditary chorea with response to levodopa
Neurology
 , 
2005
, vol. 
64
 (pg. 
1952
-
1954
)
46
do Carmo Costa
M.
Costa
C.
Silva
A.P.
Evangelista
P.
Santos
L.
Ferro
A.
Sequeiros
J.
Maciel
P.
Nonsense mutation in TITF1 in a Portuguese family with benign hereditary chorea
Neurogenetics
 , 
2005
, vol. 
6
 (pg. 
209
-
215
)
47
Asmus
F.
Devlin
A.
Munz
M.
Zimprich
A.
Gasser
T.
Chinnery
P.F.
Clinical differentiation of genetically proven benign hereditary chorea and myoclonus-dystonia
Mov. Disord.
 , 
2007
, vol. 
22
 (pg. 
2104
-
2109
)
48
Provenzano
C.
Veneziano
L.
Appleton
R.
Frontali
M.
Civitareale
D.
Functional characterization of a novel mutation in TITF-1 in a patient with benign hereditary chorea
J. Neurol. Sci.
 , 
2008
, vol. 
264
 (pg. 
56
-
62
)
49
Nagasaki
K.
Narumi
S.
Asami
T.
Kikuchi
T.
Hasegawa
T.
Uchiyama
M.
Mutation of a gene for thyroid transcription factor-1 (TITF1) in a patient with clinical features of resistance to thyrotropin
Endocr. J.
 , 
2008
, vol. 
55
 (pg. 
875
-
878
)
50
Ferrara
A.M.
De Michele
G.
Salvatore
E.
Di Maio
L.
Zampella
E.
Capuano
S.
Del Prete
G.
Rossi
G.
Fenzi
G.
Filla
A.
, et al.  . 
A novel NKX2.1 mutation in a family with hypothyroidism and benign hereditary chorea
Thyroid
 , 
2008
, vol. 
18
 (pg. 
1005
-
1009
)
51
Glik
A.
Vuillaume
I.
Devos
D.
Inzelberg
R.
Psychosis, short stature in benign hereditary chorea: a novel thyroid transcription factor-1 mutation
Mov. Disord.
 , 
2008
, vol. 
23
 (pg. 
1744
-
1747
)
52
Devos
H.
Rodd
C.
Gagné
N.
Laframboise
R.
Van Vliet
G.
A search for possible molecular mechanisms of thyroid dysgenesis: sex ratios and associated malformations
J. Clin. Endocrinol. Metab.
 , 
1999
, vol. 
84
 (pg. 
2502
-
2506
)
53
Castanet
M.
Polak
M.
Bonaïti-Pellié
C.
Lyonnet
S.
Czernichow
P.
Léger
J.
19 years of national screening for congenital hypothyroidism: familial cases with thyroid dysgenesis suggest the involvement of genetic factors
J. Clin. Endocrinol. Metab.
 , 
2001
, vol. 
86
 (pg. 
2009
-
2012
)
54
Olivieri
A.
Stazi
M.A.
Mastroiacovo
P.
Fazzini
C.
Medda
E.
Spagnolo
A.
De Angelis
S.
Garndolfo
M.E.
Taruscio
D.
Cordeddu
V.
, et al.  . 
A population based study on the frequency of additional congenital malformations in infants with congenital hypothyroidism: data from the Italian registry for congenital hypothyroidism (1991–1998)
J. Clin. Endocrinol. Metab.
 , 
2002
, vol. 
87
 (pg. 
557
-
662
)

Author notes

The authors wish it to be known that, in their opinion, the first two authors should be regarded as joint First Authors.
Present address: Pediatric Endocrinology, University Children’s Hospital Basel, Basel, Switzerland.