Objective: Defects in the human thyroid peroxidase (TPO) gene are reported to be one of the causes of congenital hypothyroidism (CH) due to a total iodide organification defect. The aim of the present study was to determine the nature and frequency of TPO gene mutations in patients with CH, characterised by elevated TSH levels and orthotopic thyroid gland, identified in the Portuguese National Neonatal Screening Programme.
Subjects and methods: The sample comprised 55 patients, from 53 unrelated families, with follow-up in the endocrinology clinics of the treatment centres of Porto and Lisbon. Mutation screening in the TPO gene (exons 1–17) was performed by single-strand conformational analysis followed by sequencing of fragments with abnormal migration patterns.
Results: Eight different mutations were detected in 13 patients (seven homozygotes and six compound heterozygotes). Novel mutations included three missense mutations, namely 391T > C (S131P), 1274A > G (N425S) and 2512T > A (C838S), as well as the predictable splice mutation 2748G > A (Q916Q/spl?). The undocumented polymorphism 180-47A > C was also detected.
Conclusion: The results are in accordance with previous observations confirming the genetic heterogeneity of TPO defects. The proportion of patients in which the aetiology was determined justifies the implementation of this molecular testing in our CH patients with dyshormonogenesis.
The latest annual report by the committee of the Portuguese National Neonatal Screening Programme indicates that the incidence of congenital hypothyroidism (CH) in Portugal is approximately 1/3238 newborns (1). Prior to the implementation of this screening programme (presently with a coverage rate of around 99.5%), CH was one of the most frequent causes of mental retardation in children. CH includes different clinical entities and, in the majority of cases, is a consequence of thyroid dysgenesis, in which the gland is either absent (thyroid agenesis) or located ectopically and/or severely reduced in size (hypoplasia). It is estimated that 15% of CH cases occur as a consequence of defects in the biochemical mechanisms responsible for thyroid hormone biosynthesis (thyroid dyshormonogenesis) in which human thyroid peroxidase (TPO) plays an essential role (2). Many patients with total iodide organification defect have been shown to have mutations in the TPO gene (3–15). TPO is a membrane-linked haemoprotein located at the apical membrane of the thyroid cells; it catalyses the iodination and subsequent coupling of tyrosine residues in thyroglobulin, resulting in the synthesis of the thyroid hormones tri-iodothyronine and thyroxine (T4). The human TPO gene is located on chromosome 2p25 (16) and consists of 17 exons that span 150 kb, while the respective full-length mRNA is about 3 kb long (17). Other genes thought to be implicated in this form of CH include the thyroglobulin (TG) gene (2), the sodium symporter (NIS) gene (18), the pendrin gene (PDS) (19) and, more recently, thyroid oxidase gene 2 (THOX2) (20). In the present study, we screened for TPO gene mutations in 55 CH patients presenting elevated thyroid-stimulating hormone (TSH) levels and orthotopic thyroid gland.
Subjects and methods
Since the implementation of the National Neonatal Screening Programme in Portugal, 723 cases of CH have been detected (1). Diagnosis of CH is based on elevated TSH levels (cut-off value 20μU/ml) and decreased T4 levels (normal > 6.5 μg/dl), ascertained in heel puncture blood samples collected on S&S 903 filter paper (Schleicher and Schuell GmbH, Dassel, Germany) between the 4th and 7th day of life. Mutation screening was performed on a group of CH patients with elevated TSH levels at the time of diagnosis and orthotopic thyroid gland. The perchlorate discharge test, which aids in the recognition of iodide organification defects, is not routinely performed; as such, elevated plasma Tg concentrations were also considered in the selection of patients, so as to exclude cases likely to result from defects in Tg synthesis or TSH receptor inactivation. The selected 55 patients were members of 53 apparently unrelated families, none of whom had any knowledge of consanguinity. Informed consent was obtained from the patients or, in the case of minors, from their parents.
DNA was isolated from peripheral blood according to the salting out method (21). The 17 exonic regions of the TPO gene were amplified by PCR with primers as described previously (5). The PCR reaction mixture contained 25 μl of a 2 × PCR master mix (Promega Corporation, Madison, WI, USA), 1 pmol each of forward and reverse primers, 1 μl genomic DNA (50–250 ng) and nuclease-free water to a final volume of 50 μl. The PCR reactions were performed in a 9600 thermal cycler (Applied Biosystems, Foster City, CA, USA) with an initial denaturation step of 10 min at 94 °C, followed by 35 cycles consisting of denaturation at 94 °C for 1 min, annealing at 57 °C for 1 min and extension at 72 °C for 1 min, and with a final extension at 72 °C for 10 min. The 18 amplicons (exon 8 subdivided into two fragments, 8A and 8B) were subjected to single-strand conformational analysis (SSCA) using both the PhastSystem (Pharmacia Biotech) in Phast-Gels, according to the manufacturer’s instructions, and a standard vertical electrophoresis system, on 0.5 × and 1.0 × MDE gels (FMC Bioproducts, Rock-land, ME, USA). Gels were stained by standard silver staining methods. Fragments presenting abnormal electrophoretic mobility were sequenced using the Big Dye Terminator Kit v2.0 (Applied Biosystems) and capillary electrophoresis. Because of poor electrophoretic separation on SSCA, some fragments (8A, 8B and 11) were also sequenced in all patients. A novel mutation identified in exon 8 was further characterised by restriction analysis. This mutation creates a new restriction site for DdeI. The PCR product of fragment 8B was incubated with DdeI (New England BioLabs, Beverly, MA, USA) overnight at 37 °C and separated on a 2% agarose gel.
To investigate whether an alteration was a causal mutation or a common polymorphic variant, population screens were carried out on 100 healthy controls, either by SSCA or by differential restriction analysis. Family co-segregation studies were performed whenever possible.
The novel predicted splice site mutation, 2748G > A, was run on the GENSCAN program which predicts the locations and exon–intron structures of genes (http://genes.mit.edu/GENSCAN.html). For the novel missense mutations, phylogenetic conservation of the amino acid sequences among the peroxidase superfamily was assessed with the aid of the CLUSTAL program (http://www.ebi.ac.uk/servicestmp/clustalw).
Results and discussion
Mutation analysis of the TPO gene
In all, eight different mutations and 15 polymorphisms were detected in this group of patients. Novel alterations included three missense and one putative splice mutation, as well as a silent A to C transversion in intron 3 (180-47A > C). In 13 of the 55 patients, deleterious mutations were identified in both alleles. Clinicopathologic and mutation data are summarised in Tables 1 and 2. Sequence analysis of exon 8 revealed a homozygous GGCC duplication at nucleotide position 1187 in two unrelated patients (1a and 2a). This mutation has been described previously (3) and leads to a frameshift with a termination signal in exon 9. The same duplication (1183_1186dupGGCC) was also found in the affected siblings of family 3, in heterozygosity with 1978C > G, a documented missense mutation in exon 11 which results in a glutamine to glutamic acid substitution at amino acid position 660 (Q660E) (10). Further investigation in other members of the family revealed that the father carried the 1183_1186dupGGCC mutation, while the mother and four unaffected sisters were carriers of the Q660E mutation. The patient in family 4 was found to be heterozygous for Q660E and a novel missense mutation in exon 8, where an A to G transition at nucleotide position 1274 results in an aspargine to serine change at codon 425 (Fig. 1). This alteration creates a new restriction site for DdeI, such that the 480 bp amplicon encompassing exon 8 is cut into fragments of 322 and 158 bp (Fig. 1C). Restriction analysis with DdeI was thus used to screen 100 healthy controls, none of whom presented this base change. Only the patient’s mother was available for sequence analysis, and she was found to carry the new N425S mutation. SSCA of exon 9 revealed an aberrant conformer in patient 5a. Both parents were heterozygous for this abnormal fragment, while an unaffected sister presented only the normal migration pattern. Sequence analysis showed a homozygous, known missense mutation at position 1477 with a nucleotide change from G to A, resulting in a glycine to serine substitution at codon 493. Patient 6a was homozygous for the documented mutation Q660E that is also present in compound heterozygosity in the affected siblings of families 3 and 10.
In the patient of family 7, a mobility shift in exon 14 led to the identification of homozygosity for the known single base pair deletion 2422delT (11). This frameshift mutation, which results in a premature stop codon, was found in heterozygosity in both parents and in one of two unaffected brothers. Patient 8a was also heterozygous for this frameshift mutation and the novel mutation in exon 5: a T to C transition at nucleotide 391 that replaces a serine at codon 131 with a proline (S131P). Both affected siblings of family 9 were found to be homozygous for a novel missense mutation in exon 14: a T to A transition at nucleotide 2512, resulting in a cysteine to serine substitution at codon 838 (C838S). The parents were heterozygous for this mutation. Finally, two other affected siblings (family 10) were found to be compound heterozygotes for Q660E and a novel point mutation, 2748G > A, in exon 16 (Fig. 2). This novel mutation, silent at the amino acid level (Q916Q), was located at position −1 of the 5′ (donor) splice site consensus sequence in exon 16 and thus putatively affects splicing. The father is a carrier of this novel mutation in exon 16 and the mother carries the known mutation in exon 11.
Deleterious effect of the novel identified mutations
No thyroid tissue was available for the functional studies of the peroxidases with the four novel mutations. The deleterious effect of the missense mutations was therefore evaluated by assessing the degree of evolutionary conservation of the respective amino acids, among several human and other animal wild-type peroxidases (Table 3). This approach of multiple sequence alignment indicated that N425 and some neighbouring residues are well conserved among the peroxidase superfamily. C838 is similarly well conserved among the TPOs of different species but has no counterpart in human mieloperoxidase (MPO), lactoperoxidase (LPO) or eosinofil peroxidase (EPO), or in bovine lactoperoxidase. The region encoded by exons 8, 9 and 10 is thought to be the catalytic centre of the TPO protein and several deleterious mutations in these exons have been reported (5–7, 11, 12, 15). The missense mutation N425S described here also falls within this domain. An acidic to neutral amino acid change is likely to influence the electron transfer environment and consequently the enzymatic activity of the protein. Besides the membrane spanning region encoded by exon 15 (7), little is known about the carboxyl terminus of TPO, where the relationship between function and structure is poorly understood. Mutations in this area have been reported in several patients with severe thyroid function (3, 5, 7, 9, 12–15). The region encoded by exon 14 bears significant similarities to the epidermal growth factor (EGF)-like potential calcium binding domain (15), where there are three disulphide bonds, one of which is formed between residues C825 and C838. The latter is precisely the residue which is altered in the novel mutation described here (C838S). The substitution of cysteine by serine disables the formation of the disulphide bond and may thus disrupt the tertiary structure of the EGF-like potential calcium-binding domain. The novel mutation S131P, identified in exon 5, is likely to disrupt the potential glycosylation site at N129 (29), since the acceptor sequence required for N-glycosylation, namely N-X-S/T is modified to N-X-P. Studies have shown that N-glycans play an essential role in the correct folding, intracellular trafficking and activity of TPO (30). The novel mutation identified in exon 16 (2748G > A) is located at position −1 of the 5′ (donor) consensus splice site. Krawczak et al. (28) have reported a 97% consensus value for this residue in a similar sequence context. The GENSCAN program (31), that is designed to predict complete gene structures in genomic sequences, attributes a log-odds score of −20 for splicing with this base change, providing further evidence that the mutation abolishes splice site recognition. In all four cases, co-segregation analyses were consistent with a causal nature of the new mutations. Moreover, they were not detected in 100 healthy controls (200 alleles), as opposed to the hitherto undocumented polymorphism 180-47A > C, which was found in 19% of the normal control alleles.
In the 13 patients with TPO mutations, the most prevalent mutation was Q660E, previously reported in a Brazilian patient (10). It was found in a homozygous state in patient 6 and in compound heterozygosity in patients 3a, 3b, 4a, 10a and 10b. A similar prevalence was noted for the 1183_1186dupGGCC mutation (six alleles of the 13 patients with TPO mutations), also reported to be the most prevalent among Dutch patients (11). The severity of these two mutations is evident in family 3 where the older sibling 3a, who had not received l-thyroxine therapy, was severely affected with mental retardation. In the same line of reasoning, one may speculate that the novel predictive splice mutation described here (2748G > A) is somewhat milder, since, in contrast with the other patients, neither affected sibling in family 10 presented goiter. Analysis of human TPO mRNA in these patients would help to elucidate the consequence of the splicing error (exon skipping and/or resort to cryptic splice sites, and resulting reading frame). This is the first molecular characterisation ever performed in a cohort of Portuguese CH patients in order to establish the aetiology of CH due to a dyshormonogenic defect. The exceptionally large proportion of patients found to have TPO mutations (approximately 24% of our sample) justifies the implementation of routine molecular testing in our CH neonates, with immediate benefits in terms of counselling and monitoring of future pregnancies, and with the foreseeable future benefit of aetiology-based differential treatment.
The authors would like to thank the families who gave their consent and collaborated in this study and the staff of the Neonatal Screening Laboratory for providing the original screening data of the patients. This work was supported by the Comissão de Fomento da Investigação em Cuidados da Saú de of the Portuguese Ministry of Health (project no. 163/99, Edital no. 898/98).
Clinicopathological characteristics of the CH patients (ten families) with TPO mutations identified in this study.
|Family/patient||Sex||Current age(years)||Thyroid gland||TSH* (μU/ml)||T4* (μg/dl)||Tg (ng/ml)||Identified mutations|
|* The levels of T4 and TSH at the 4th–7th day of life. ND, not determined.|
|1/a||F||7||Goiter||215||0.7||299||[1183_1186dupGGCC] + [1183_1186dupGGCC]|
|2/a||F||2||Goiter||195||2.9||955||[1183_1186dupGGCC] + [1183_1186dupGGCC]|
|3/a||F||32||Goiter||ND||ND||ND||[1183_1186dupGGCC] + [1978C > G]|
|3/b||F||13||Goiter||235||1.9||79.7||[1183_1186dupGGCC] + [1978C > G]|
|4/a||F||19||Goiter||81||3.6||ND||[1274A > G] + [1978C > G]|
|5/a||M||11||Goiter||109.6||1.2||1344||[1477G > A] + [1477G > A]|
|6/a||F||5||Goiter||285||2.8||731||[1978C > G] + [1978C > G]|
|7/a||F||18||Goiter||192||0.6||ND||[2422delT] + [2422delT]|
|8/a||M||3||Goiter||258||0.5||416||[2422delT] + [391T > C]|
|9/a||M||5||Goiter||328||2.9||2356||[2512T > A] + [2512T > A]|
|9/b||M||3||Goiter||378||1.8||1758||[2512T > A] + [2512T > A]|
|10/a||F||7||Normal||288||0.7||640||[1978C > G] + [2748G > A]|
|10/b||F||3||Normal||35.8||3.7||ND||[1978C > G] + [2748G > A]|
Description of TPO mutations in the 13 CH patients.
|Exon||Mutationa||Effect of the mutation on protein synthesis||Reference||Frequency of mutant allelesb|
|a The first A in the TPO start codon considered as position 1.|
|b Frequency amongst the 55 patients (110 alleles).|
|5||391T > C||S131P||Present study||1/110|
|8||1274A > G||N425S||Present study||1/110|
|9||1477G > A||G493S||15||2/110|
|11||1978C > G||Q660E||10||7/110|
|14||2512T > A||C838S||Present study||4/110|
|16||2748G > A||Spl?||Present study||1/110|
Comparison of amino acid sequences among various peroxidases coincident with mutations N425S and C838S.
|Mutant TPO||K A L S425 A H W||G R T S838 V D S|
|* No C838 counterpart found in these peroxidases. Highly conserved amino acids are expressed in bold.|
|Human TPO (17)||K A L N425 A H W||G R T C838 V D S|
|Pig TPO (22)||K A L N424 A H W||G R T C837 V D A|
|Mouse TPO (23)||K A I N413 K H W||G K T C826 I D S|
|Rat TPO (24)||K A I N413 T H W||G K T C826 I D S|
|Human MPO (25)||K S L N434P R W||*|
|Human LPO (26)||K R L N401P Q W||*|
|Bovine LPO (26)||K K L N401P H W||*|
|Human EPO (27)||R R L N406P R W||*|
Abramowicz MJ, Targovnik HM, Varela V, Cochaux P, Krawiec L, Pisarev MA, Propato FV, Juvenal G, Chester HA & Vassart G. Identification of a mutation in the coding sequence of the human thyroid peroxidase gene causing congenital goiter. Journal of Clinical Investigation 1992 90 1200–1204.
Bikker H, den Hartog MT, Baas F, Gons MH, Vulsma T & de Vijlder JJ. A 20 basepair duplication in the human thyroid peroxidase gene results in a total iodide organification defect and congenital hypothyroidism. Journal of Clinical Endocrinology and Metabolism 1994 79 248–252.
Bikker H, Vulsma T, Baas F & de Vijlder JJM. Identification of five novel inactivating mutations in the human thyroid peroxidase gene by denaturing gradient gel electrophoresis. Human Mutation 1995 6 9–16.
Bikker H, Waelkens JJ, Bravenboer B & de Vijlder JJ. Congenital hypothyroidism caused by a premature termination signal in exon 10 of the human thyroid peroxidase gene. Journal of Clinical Endocrinology and Metabolism 1996 81 2076–2079.
Bikker H, Baas F & de Vijlder JJM. Molecular analysis of mutated thyroid peroxidase detected in patients with total iodide organification defects. Journal of Clinical Endocrinology and Metabolism 1997 82 649–653.
de Vijlder JT, Ris-Stalpers C & Vulsma T. Inborn errors of thyroid hormone biosynthesis. Experimental Clinical Endocrinology and Diabetes 1997 105 32–37.
Kotani T, Umeki K, Yamamoto I, Maesaka H, Tachibana K & Ohtaki S. A novel mutation in human thyroid peroxidase gene resulting in total organification defect. Journal of Endocrinology 1999 160 267–273.
Santos CL, Bikker H, Rego KGM, Nascimento AC, Tambascia M, de Vijlder JJM & Medeiros-Neto G. A novel mutation in TPO gene in goitrous hypothyroid patients with iodide organification defect. Clinical Endocrinology 1999 51 165–172.
Bakker B, Bikker H, Vulsma T, de Randamie JS, Wiedijk BM & de Vijlder JJM. Two decades of screening for congenital hypothyroidism in the Netherlands: TPO gene mutations in total iodide organification defects (an update). Journal of Clinical Endocrinology and Metabolism 2000 85 3708–3712.
Ambrugger P, Stoeva I, Biebermann H, Torresani T, Leitner C & Gruters A. Novel mutations of thyroid peroxidase gene in patients with permanent congenital hypothyroidism. European Journal of Endocrinology 2001 145 19–24.
Umeki K, Kotani T, Kawano J, Suganuma T, Yamamoto I, Aratake Y, Furujo M & Ichiba Y. Two novel missense mutations in thyroid peroxidase gene, R665W and G771R, result in localization defect and cause congenital hypothyroidism. European Journal of Endocrinology 2002 146 491–498.
Niu DM, Hwang B, Chu YK, Liao CJ, Wang PL & Lin CY. High prevalence of a novel mutation (2268 insT) of the thyroid peroxidase gene in Taiwanese patients with total iodide organification defect, and evidence for a founder effect. Journal of Clinical Endocrinology and Metabolism 2002 87 4208–4212.
Wu JY, Shu SG, Yang CF, Lee CC & Tsai FJ. Mutation analysis of thyroid peroxidase gene in Chinese patients with total iodide organification defect: identification of five novel mutations. Journal of Endocrinology 2002 172 627–635.
Kimura S, Kotani T, Mcbride OW, Umeki K, Hirai K, Nakayama T & Ohtaki S. Human thyroid peroxidase: complete cDNA and protein sequence, chromosome mapping, and identification of two alternative spliced mRNAs. PNAS 1987 84 5555–5559.
Endo Y, Onogi S, Umeki K, Yamamoto I, Kotani T, Ohtaki S & Fujita T. Regional localization of the gene for thyroid peroxidase to human chromosome 2p25 and mouse chromosome 12C. Genomics 1995 25 760–761.
Fujiwara H, Tatsumi K, Miki K, Harada T, Miyai K, Takai S & Amino N. Congenital hypothyroidism caused by a mutation in the Na +/I − symporter. Nature Genetics 1997 16 124–125.
Coyle B, Reardon W, Herbrik JA, Tsui LC, Gausden E, Lee J, Coffey R, Grueters A, Grossman A, Phelps PD, Luxon L, Kendall-Taylor P, Scherer SW & Trembath RC. Molecular analysis of PDS in Pendred syndrome. Human Molecular Genetics 1998 7 1105–1112.
Moreno JC, Bikker H, Kempers MJ, Trotsenburg P, Baas F, de Vijlder JJM, Vulsma T & Ris-Stalpers C. Inactivation mutations in gene for thyroid oxidase 2 (THOX2) and congenital hypothyroidism. New England Journal of Medicine 2002 347 95–102.
Miller SA, Dykes DD & Polesky HF. A simple salting out procedure for extracting DNA from human nucleated cells. Nucleic Acids Research 1988 16 1215.
Magnusson R, Gestautas J, Taurog A & Rapoport B. Molecular cloning of the structural gene for porcine thyroid peroxidase. Journal of Biological Chemistry 1987 262 13885–13888.
Kotani T, Umeki K, Yamamoto I, Takeuchi M, Takeuchi S, Nakayama T & Ohtaki S. Nucleotide sequence of the cDNA encoding mouse thyroid peroxidase. Gene 1993 123 289–290.
Derwahl M, Seto P & Rapoport B. Complete nucleotide sequence of the cDNA for thyroid peroxidase in FRTL5 rat thyroid cells. Nucleic Acids Research 1989 17 8380.
Johnson KR, Nauseef WM, Care A, Wheelock MJ, Shane S, Hudson S, Koeffler HP, Selsted M, Miller C & Rovera G. Characterization of cDNA clones of human myeloperoxidase: predicted amino acid sequence and evidence for multiple mRNA species. Nucleic Acids Research 1987 15 2013–2028.
Dull T, Oyeda C, Strosberg D, Nedwin G & Seilhamer J. Molecular cloning of cDNAs encoding bovine and human lactoperoxidase. DNA and Cell Biology 1990 1 499–509.
Sakamaki K, Tomonaga M, Tsukui K & Nagata S. Molecular cloning and characterization of chromosomal gene for human eosinophil peroxidase. Journal of Biological Chemistry 1989 264 16828–16836.
Krawczak M, Reiss J & Cooper D. The mutational spectrum of single base-pair substitutions in mRNA splice junctions of human genes: causes and consequences. Human Genetics 1992 90 41–54.
Taurog A. Hormone synthesis: thyroid iodine metabolism. In Werner and Ingbar’s The Thyroid, edn 8, part 1, ch. 4, pp 63–64. Eds LE Braverman & RD Utiger. Philadelphia: Lippincott Williams & Wilkins, 2000.
Fayadat L, Niccoli-Sire P, Lanet J & Franc JL. Human thyroperoxidase is largely retained and rapidly degraded in the endoplasmic reticulum. Its N-glycans are required for folding and intracellular trafficking. Endocrinology 1998 139 4277–4285.
Burge CB. Modeling dependencies in pre-mRNA splicing signals. In Computational Methods in Molecular Biology, pp 127–163. Eds S Salzberg, D Searls & S Kasif. Amsterdam: Elsevier Science, 1998.