Abstract
Background: Congenital hypothyroidism (CH) is a common endocrine disorder in newborns. The cause of CH is thyroid dysgenesis in 80-85% of patients. Paired box gene 8 (PAX8) is a thyroid transcription factor that plays an important role in thyroid organogenesis and development. To date, 22 different PAX8 gene mutations have been reported. Methods: Four generations of a Hungarian Jewish family were affected, and in the 3 generations studied, 9 males and 4 females were affected and 3 first-degree relatives were unaffected. Six were diagnosed at birth [thyroid-stimulating hormone (TSH) level 59-442 mU/l] and 7 at 2-48 years of age (TSH level 6-223 mU/l). One affected patient had thyroid hemiagenesis on ultrasound. Results: Direct sequencing of the PAX8 gene revealed a novel single nucleotide substitution (c.162 A>T) in exon 2 that resulted in the substitution of the normal serine 54 with a cysteine (S54C), which segregated with elevated serum TSH levels. Other mutations of the same amino acid (S54G and S54R) have also been shown to produce functional impairment. Conclusion: We report a large family with a novel mutation in the PAX8 gene presenting with variable phenotype and with a high proportion of affected family members.
Established Facts
• PAX8 gene mutations are one of the genetic causes of thyroid dysgenesis.
• To date, 22 different PAX8 gene mutations have been reported in humans.
Novel Insights
• Identification of a novel single nucleotide substitution (c.162 A>T) in the PAX8 gene that resulted in the replacement of the normal serine 54 with a cysteine (S54C) in a large family with variable magnitude of hypothyroidism and with a high proportion of affected individuals.
Introduction
Congenital hypothyroidism (CH; OMIM: 218700) is a common endocrine disorder in newborns with an incidence of 1:4,000 to 1:1,500, depending on the assigned cutoff thyroid-stimulating hormone (TSH) value [1,2]. In 80-85% of cases, CH is secondary to thyroid dysgenesis, which presents with a thyroid gland that may be absent (athyreosis), hypoplastic (hypoplasia), or located in an unusual position (ectopy). In the remaining 15-20% of cases, CH results from inborn errors of thyroid hormone biosynthesis, secretion, or recycling (dyshormonogenesis) [3]. Thyroid dysgenesis occurs mostly as a sporadic disease; however, a genetic cause has been demonstrated in about 5% of the reported cases. To date, mutations in genes involved in thyroid organogenesis have been identified in the following genes: thyroid transcription factors 1 and 2 (TTF1 or NKX2.1 and TTF2 or FOXE1), NK2 homeobox 5 (NKX2.5), thyrotropin receptor (TSHR), transcription factor GLI similar 3 (GLIS3), and the paired box gene 8 (PAX8) [4] [PAX8 OMIM: 167415]. The latter, a paired domain-containing protein belonging to the Pax family of transcription factors, is expressed in the thyroid gland, kidney, and central nervous system [5]. The PAX8 gene is located on human chromosome 2q12-q14 and consists of 11 exons encoding a 128-amino-acid protein, which plays an important role in thyroid organogenesis [4]. In the adult thyroid, PAX8 is an essential regulator of thyroid-specific gene expression such as thyroid peroxidase (TPO), thyroglobulin (TG), and sodium/iodide symporter (NIS) [6]. PAX8 gene mutations are inherited in an autosomal dominant fashion [4], which contrasts with the recessive inheritance in Pax8 knockout mice [7]. To date, 22 different PAX8 gene mutations have been reported in humans [4,8]. Systematic PAX8 gene mutation screening was performed in 17 cohorts of patients with CH. Mutations occurred with a prevalence of 1.0%, ranging from 0 to 3.4% [8,9]. The clinical phenotype of individuals with identical PAX8 gene mutations can be variable, ranging from overt CH with severe thyroid hypoplasia to subclinical CH with a morphologically normal gland. The majority of PAX8 gene mutations are located within the DNA-binding paired domain and result in a severe reduction in DNA-binding affinity [3].
Patients and Methods
Case History
Four generations of a Jewish family of Hungarian origin were affected. Of the three generations studied, 9 males and 4 females were affected and 3 were unaffected first-degree relatives (fig. 1). Of the 6 miscarriages, 5 occurred in 1 woman (III-8). Clinical features of all family members are shown in table 1. Affected individuals had hypothyroidism of variable severity diagnosed at different ages. None of them had autoantibodies to thyroperoxidase or thyroglobulin. Six (patient IV-3, IV-4, IV-6, IV-9, IV-10, and IV-11) were diagnosed at birth with TSH values ranging from 59 to 442 mU/l. The remaining 7 affected individuals were diagnosed at 2-48 years of age, with serum TSH values from 6 to 223 mU/l. Note that in Israel the screening program for CH is based on the measurement of total T4 (TT4) followed by a confirmatory TSH test on samples with T4 values below the 10th percentile. Thus, it is possible that individuals with T4 within the normal range (for example IV-7) were missed in the neonatal screening. Of note is a striking variability in the initial clinical presentation within the family. For example, in the first nuclear family, patient IV-3 was identified by neonatal screening while his affected brother (IV-2) had a normal TSH value upon neonatal screening and was found to have subclinical hypothyroidism at the age of 6 years. His father (III-1) was found to have overt hypothyroidism at the age of 48 years when he was admitted to hospital for severe weakness. He also showed Parkinson's and chronic kidney disease of unknown cause with normal renal ultrasound. Renal function tests (serum urea and creatinine) were normal in all affected individuals. Ultrasound of the kidney in 4 other affected family members (IV-6, IV-7, IV-10, and II-3) was normal. In the second nuclear family, patient IV-4 was found to be hypothyroid following neonatal study, while her mother was diagnosed at the age of 14 years after presenting with delayed puberty. No test results prior to treatment could be found. In the third nuclear family, patient IV-6 was found to be hypothyroid following neonatal screening while his younger brother (IV-7) was diagnosed at the age of 3 years with a very high TSH level. His father (III-5) was found to have high TSH values in adulthood, after presenting with dizziness without hypothyroidism-related manifestations. In the fourth nuclear family, patient IV-9 was found to have CH with thyroid hemiagenesis, while in her brother (IV-11), hypothyroidism was diagnosed at the age of 3 years. He had normal thyroid gland imaging, and his father (III-7) had asymptomatic subclinical hypothyroidism identified only while testing for the present study. Six affected individuals had neurologic and cognitive abnormalities, developmental delay, and/or attention deficit hyperactive disorder.
Thyroid Function Tests
Blood was collected locally and shipped for analysis to the Chicago laboratory. TSH, TT4 and TT3 were measured on the Elecsys Automated System (Roche Molecular Biochemicals GmbH and Hitachi, Ltd., Indianapolis, Ind., USA) platform, total reverse T3 by ZenTech (Liege, Belgium), TG by in-house radioimmunoassay, and antibodies to TG and TPO by the Kronus (Star, Idaho, USA). The free T4 index was calculated from the TT4 and the resin T4 uptake ratio.
DNA Analysis
The clinical and genetic studies were approved by the institutional review boards. After written informed consent was obtained from all participating family members, genomic DNA from peripheral mononuclear blood cells was isolated using the QIAamp DNA Mini Kit (Qiagen) followed by amplification of genomic DNA by polymerase chain reaction and direct sequencing of the PAX8 gene, exons 0 through exon 11. All polymerase chain reaction samples were sequenced using automated fluorescence-based sequencing (373OXL 96 capillary; Applied Biosystem Carlsbad, Calif., USA). Primer sequences are available upon request.
Results
Direct sequencing of exon 0-11 of the PAX8 gene revealed a novel single nucleotide substitution in exon 2 of the PAX8 gene (c.162 A>T) that resulted in the substitution of the normal serine 54 with a cysteine (S54C) (fig. 2a). This S54C mutation cosegregated with a biochemical hypothyroid phenotype in all three generations tested (fig. 1). Evaluation of the new sequence alterations using ‘PolyPhen-2' indicated that this variant is ‘probably damaging', with a score of 0.997.
Discussion
We identified a novel mutation in the PAX8 gene, present in all 13 affected individuals of the family. The thyroid phenotype was considerably variable, with respect to (1) thyroid function (overt to subclinical hypothyroidism), (2) onset of disease (at birth to late adulthood), and (3) thyroid gland anatomy (hemiagenesis to normal). Variable phenotypic expression in PAX8 mutations has been reported in several families. Patients can be euthyroid or severely hypothyroid, and the thyroid development can range from athyreosis to normal-sized orthotropic gland [4,10,11,12,13,14,15,16]. Possible explanations for interfamilial variability include a polygenic etiology, epigenetic mechanisms that cause stochastic variations of gene expression at multiple loci, variation in the timing of PAX8 expression in embryonic life, or somatic mutations with a dominant effect in a thyroid development gene [17].
The S54C Pax8 mutation affects a highly conserved amino acid in the paired domain which lies between the second and the third helical region of the N-terminal homeodomain-like motif (fig. 2b). Other mutations of the same amino acid (S54G and S54R) have also been shown to exhibit functional impairment. Meeus et al. [18] identified an S54G mutation in a French family with CH. In addition, one of the affected siblings displayed unilateral kidney disease. Functional analysis of the mutant PAX8 demonstrated that it is unable to bind a specific cis-acting element of the promoter and has almost lost the ability to control together with TTF1 TG gene transcription. Hermanns et al. [19] reported an S54R mutation in 2 members of a Turkish family. In vitro studies showed that the mutant protein had an impaired binding to the TPO and TG gene promoter-binding sites and exerted a dominant negative effect on the wild-type PAX8. Both reports support a functional impairment of a mutant amino acid at this location.
The PAX8 gene is expressed in the kidney and plays an important role in its development. Urogenital malformations (horseshoe kidney, undescended testes, hydrocele, ureterocele, and kidney agenesis) associated with PAX8 gene mutations have been previously reported [18,20]. One of the affected individuals of this family had chronic kidney disease, but kidney ultrasound showed no abnormal urogenital malformation. Four other affected members of the family were found to have normal kidney ultrasounds. Interestingly, 6 affected members of the family had neurologic or cognitive abnormalities including Parkinsonism, developmental delay, and attention deficit hyperactive disorder. This suggests a possible association between the PAX8 gene mutation and neurocognitive impairment, which has not been previously reported. The PAX8 gene is transiently expressed during development in the myencephalon and in the entire length of the neural tube, but no expression is detected in the brain at later stages or in adults [5]. Its role in these tissues has not been as well demonstrated as in the kidney and thyroid, and no neurological dysfunction is evidenced in Pax8 knockout mice [7]. Thus, it is currently not possible to determine whether the neurological manifestations could be attributed to the mutation. In this respect, it should be noted that 2 of the 4 family members with attention deficit hyperactive disorder did not carry the PAX8 gene mutation (IV-5 and IV-8). Another unusual feature is the high proportion of individuals harboring the mutation (13 of 16) when in autosomal dominant inheritance one would have expected an equal number of affected and unaffected cases.
In conclusion, we report a large family with a novel PAX8 gene mutation (S54C) causing autosomal dominant CH of variable expressivity and with a high proportion of affected individuals.
Acknowledgements
This work is supported in part by grant R37DK15070 from the National Institutes of Health and the Seymour J. Abrams fund for thyroid research (to S.R.). We thank all family members for their participation in this study. We are grateful to the following colleagues for advice in the course of investigation: Dr. Roy E. Weiss, Dr. Alexandra M. Dumitrescu, and Dr. Theodora Pappa.
Disclosure Statement
The authors declare that they have no conflicts of interest.