Forty-three (28 boys and 15 girls, male-female ratio 1.87:1.0, mean age 0.06 ± 0.03 years) cases of CH with goiter were identified in the neonatal screening at neonatal screening centers in Shandong province, China. The neonati were given full breast feeding 72 h after birth. Their heel blood was dropped in a dry blood spot on a filter paper to examine thyroid stimulating hormone (TSH) concentration for screening. Thyroxine (T4), triiodothyronine (T3), TSH, free thyroxine (FT4), and free triiodothyronine (FT3) in the serum were formally measured when the subjects had elevated TSH (≥20 μIU/ml) levels on the screening test. The subjects were diagnosed with CH if they had abnormally elevated TSH concentration (normal range 0.27–4.2 μIU/ml) and low FT4 concentration (normal range 12–22 pmol/L). This biochemical diagnosis of CH was followed by a 99mTc thyroid scan or thyroid ultrasound examination in the patients with eutopic thyroid gland. All the subjects from 43 unrelated families were free from other congenital diseases. Our study was approved by the medical ethics committee of the Affiliated Hospital of Qingdao University with informed consent of patient families.

Genomic DNA was isolated from peripheral blood leucocytes of patients and stored at −20°C up to the time of analysis. The complete sequences of DUOX1 (35 exons) and DUOXA1 (11 exons), including splice and flanking intronic regions, were amplified by polymerase chain reaction (PCR). The primer sequences of all exons in DUOX1 and DUOXA1 are listed in Tables 1, 2, respectively. Mutations in DUOX2 (10), DUOXA2 (9), TPO, TG, and NIS (18) for all subjects had been excluded in our previous study in these patients. The PCR reaction system (25 μL) consisted of 100 ng of genomic DNA, 1 U of Taq DNA polymerase, 500 μM dNTPs, 0.5 μM each primer and 1×reaction buffer (20 mM Tris-HCl, pH 8.0, 3 mM MgCl₂). Samples were pre-denatured at 95°C for 5 min, followed by PCR reaction for 35 cycles of 30 s at 95°C, 30 s at 55–61°C, and 30 s at 72°C, with a final extension at 72°C for 7 min. Purified PCR products were analyzed on 2% agarose gel electrophoresis and mutations were identified by analyzing the nucleotide sequence (NC_000015.10) after ABI 3730XL (Applied Biosystems) sequencing.

After extraction of RNA from human thyroid tissue, cDNA was synthesized by the primer oligo (dT) through the reverse transcription reaction. The consensus coding sequence of DUOX1 (4,656 bp) was amplified from cDNA (forward primer: 5′-ATGGGCTTCTGCCTGGCTCTA-3′, reverse primer: 5′-CTAGAAGTTCTCATAATGGTG-3′) and was cloned into pEASY-Blunt M2 expression vector. Similarly, the consensus coding sequence of DUOXA1 (1,452 bp) was amplified from cDNA (forward primer: 5′-ATGGCTACTTTGGGACACACA-3′, reverse primer: 5′-TCAGATTAGAGGTGTGTGGCG-3′) and was cloned into pEASY-Blunt M2 expression vector. DUOX1 p.R1307Q mutant was introduced by site-directed mutagenesis (Fast Mutagenesis System) into the expression vector using the mutagenic forward primer 5′-GAGCGGCCAGTGGGTCCAAATCGCTTGTC-3′ and the mutagenic reverse primer 5′-TGGACCCACTGGCCGCTCTTGTACTCGAA-3′. The expression vector of mutant DUOXA1 p.R56W was constructed on the basis of wild type expression vector using the mutagenic forward primer 5′-GGCTGTTCTGGCTGCTTTGGGTGGTGACC-3′ and the mutagenic reverse primer 5′-AAAGCAGCCAGAACAGCCTCGTCTTTCCC-3′ in the same way. All the constructed expression vectors were confirmed by DNA sequencing.

HeLa cells were grown in six-well plates in 1640 culture medium supplemented with 10% fetal calf serum in a humidified condition with 5% CO₂ at 37°C. HeLa cells were divided into eight groups according to the transfected expression vector type before transfection, including DUOX1 (containing DUOX1 wild type [WT] expression vector only), DUOXA1 (containing DUOXA1 WT expression vector only), DUOX1 and DUOXA1 (containing both DUOX1 WT and DUOXA1 WT expression vector), p.R1307Q and DUOXA1 (containing DUOX1 p.R1307Q mutant expression vector and DUOXA1 WT expression vector), DUOX1 and p.R56W (containing DUOX1 WT expression vector and DUOXA1 p.R56W mutant expression vector), p.R1307Q and p.R56W (containing both DUOX1 p.R1307Q and DUOXA1 p.R56W mutant expression vectors), pEASY-Blunt M2 (containing an empty vector), and parental non-transfected cells (containing no expression vector). When HeLa cells reached 70–80% confluence, they were transfected with 4 μg of plasmid DNA (WT, mutant, and control) in six-well plates using TransIn EL Transfection Reagent (TransGen, Beijing, China). To ensure quantitative consistency of plasmids, empty plasmid was used as a reference in individual transfected groups and the DUOX1/DUOXA1 ratio was 7:1. The culture medium was replaced at 6 h after transfection and cells were harvested after 48 h.

HeLa cells cultured on six-well plates were suspended in Trizol reagent (Invitrogen, Carlsbad, USA) and total RNA was purified 48 h after transfection. Reverse transcription was performed using a Prime Script (™) RT Enzyme Mix I (TaKaRa, Shiga, Japan) on 1 μg of total RNA in a reaction volume of 20 μL. One-fifth of the cDNA product was used for real time PCR amplification using TIANGEN Taq DNA polymerase (TIANGEN, Beijing, China) and the amplified PCR products were detected according to the fluorescence intensity of SYBR Green I. For specific amplification of human DUOX1 cDNA, forward primer was 5′-TCTGGGACTGCTCTGGTTC-3′ and reverse primer 5′-CGATGTTCTGGTAGGTGGC-3′; for human DUOXA1, forward primer was 5′-AAAGGCTCTGGAGAAGGG-3′ and reverse primer 5′-GCATGGCTGAGGTGTAGTG-3′; for GAPDH (loading control), forward primer was 5′-AGAAGGCTGGGGCTCATTTG-3′ and reverse primer 5′-AGGGGCCATCCACAGTCTTC-3′. Samples were heated to 95°C for 15 min, followed by 40 cycles of 10 s at 95°C, 32 s with annealing/elongation at 63°C. The amplified level of the target gene was normalized against that of GAPDH.

After transfection for 48 h, monolayer cells were suspended in cell lysis solution containing protease inhibitors (Phenyl methane sulfonyl fluoride, PMSF) on ice for 30 min after washing with physiological saline. Total protein was collected from the supernatant of cell lysates after centrifugation for 20 min at 4°C. After measurement of total protein concentration using a BCA Protein Quantitative Kit (TransGen, Beijing, China), protein samples (30 μg) were denatured by heating water at 95°C for 10 min in the loading buffer (TransGen, Beijing, China), subjected to SDS-PAGE using a 8% acrylamide mini-gel, and transferred to a PVDF membrane. The PVDF membrane was blocked for 2 h at room temperature with 5% nonfat dry milk in PBS (pH7.4) and 0.1% Tween 20 and was then incubated overnight at 4°C with primary antibodies against DUOX1 (Bioss, Beijing, China) at a 1:250 dilution and against beta-Actin (Loading Control) at a 1:2,000 dilution in PBS-Tween 20 containing 5% nonfat dry milk. Membranes were washed, incubated for 1 h with peroxidase-labeled secondary antibody anti-rabbit, washed again, and visualized with enhanced chemiluminescence on CL-XPosure films.

Extracellular H₂O₂ generation was quantified by the Amplex red/horseradish peroxidase assay (Molecular Probes, Invitrogen), which detected the accumulation of a fluorescent oxidized product. Amplex Red reagent (10-acetyl-3,7-dihydroxyphenoxazine, 50 μM) and horseradish (0.1 U/ml) peroxidase (HRP) was added into each well after transfection for 48 h and the six-well plate was placed in an incubator with dark environment at 37°C for 1 h. The fluorescence intensity was measured by Synergy (™) multi-functional enzyme marking instrument using excitation at 535 nm and emission at 595 nm. H₂O₂ release of experiment groups and control groups were quantified using standard calibration curves (as shown in Supplementary Figure 1) which could change the fluorescence intensity into absolute nanomoles of H₂O₂.

Cycloheximide (CHX) at 100 μg/ml was added to block the synthesis of intracellular protein and total protein was extracted after cells were disassociated in four different time periods within 24 h. The DUOX1 protein expression was detected by Western blot.

Statistics analysis of variance (ANOVA) was used for data presented as mean ± standard deviation (SD) using the software of SPSS 17.0, followed by LSD (least significant difference) multiple comparison test. A P < 0.05 was considered to be statistically significant.

A heterozygous missense mutation (c.3920G>A, present in Genome Aggregation Database, Figure 1A) was identified in the DUOX1 gene in patient 1 (P1), causing an arginine-to-glutamine substitution at amino acid 1307 of DUOX1, which was located in a conserved sequence of the protein according to the multiple sequence alignment analysis (Figure 2A). This mutation was not found in 100 healthy control Chinese subjects. The relationship was not clear between DUOX1 mutation segregation and phenotype within the family because of the lack of relevant data. P1 was a female infant with birth weight of 2,500 g. She was born at full-term gestation by cesarean delivery and had no family history of thyroid disease. She was called back to hospital at 36 days (body weight 3,800 g, height 47 cm) after birth for further evaluation because of her high TSH concentration (>100 μIU/ml) on the filter paper neonatal screening test. Her thyroid gland was enlarged on the 99mTc thyroid scan and the perchlorate discharge test was positive. Levothyroxine (L-T4) replacement therapy was started immediately with the initial dose of 7.7 μg/kg. To maintain normal serum TSH and FT4 levels, replacement therapy dose was adjusted based on her clinical and hormonal evaluations during the subsequent follow-up. TSH, FT3, and FT4 levels remained normal following a temporary withdrawal of L-T4 therapy for 4 weeks when she was 2 years old. Her final diagnosis was therefore transient CH. The patient is currently 10.5 years old with normal physique and intelligence.

Patient 2 (P2) carried a DUOXA1 heterozygous missense mutation p.R56W (a C>T transversion at nucleotide 166 in exon 6, present in Genome Aggregation Database, Figure 1B) that changes the highly conserved arginine at amino acid 56 of DUOXA1 protein (Figure 2B) to tryptophan and resulted in severe congenital hypothyroidism, which was not identified in 100 healthy control Chinese subjects. P2 was a male subject (birth weight 2,750 g) with a p.R56W mutation, born by normal delivery at full-term gestation. He was called back to hospital after neonatal screening at 35 days from birth for a high level of TSH (>100 μIU/ml) on the filter paper screening test; he was subsequently diagnosed with CH with a low serum FT4 level (4.7 pmol/L). A 99mTc thyroid scan indicated that P2 had eutopic thyroid gland with mild diffuse goiter. L-T4 replacement therapy was started immediately at an initial dose of 8.4 μg/kg per day. To maintain normal serum TSH, FT3, and FT4 levels, the dose was adjusted according to clinical and hormonal evaluations. The patient continued to have hypothyroidism requiring thyroid hormone therapy at 2.5 years; this patient was therefore diagnosed with permanent CH. The patient is currently 9.5 years old and his physical and mental developments are normal. She currently receives L-T4 with a daily dose of 5.9 μg/kg.

The relative expression of DUOX1 mRNA decreased with mutant constructs in comparison with both wild type constructs (P < 0.01), and the group with p.R1307Q and p.R56W combination (p.R1307Q + p.R56W) had the least expression of DUOX1 mRNA compared with other groups (P < 0.01). With respect to the relative expression of DUOXA1 mRNA, groups with p.R56W mutant (DUOX1 + p.R56W or p.R1307Q + p.R56W) were far lower than other groups except for the group with DUOX1 only (P < 0.01) (Figure 3).

Consistent with the results of mRNA expression, both p.R1307Q and p.R56W were associated with decreased expression of DUOX1 protein compared with the wild type protein on the western blot analysis (P < 0.01). In particular, the expression of DUOX1 protein in HeLa cells transfected with both p.R1307Q and p.R56W in combination decreased most significantly (P < 0.01) (Figure 4). The size of the bands of DUOX1 protein is presented in Supplementary Figure 2.

As shown in Figure 5, transfection with wild-type DUOX1 or DUOXA1 each alone significantly increased H₂O₂ production compared with empty vector transfection. Transfection with mutant p.R1307Q or p.R56W alone resulted in reduced H₂O₂ production compared with their corresponding wild-type counterparts, respectively (P < 0.01). The group transfected with both wild-type DUOX1 and DUOXA1 in combination (DUOX1 or DUOXA1) produced the most amount of H₂O₂ (P < 0.01) while, in contrast, the group with transfection of both mutants p.R1307Q and p.R56W in combination (p.R1307Q + p.R56W) impaired the H₂O₂-generating system to the maximum extent (P<0.01). These patterns were consistent with the Western blot analysis results above (Figure 4). Thus, these DUOX1 and DUOXA1 mutants were able to substantially inhibit the functional activities of the H₂O₂-generating system, demonstrating the functional consequences of the two mutations.

DUOX1 protein synthesis with transfection of all expression vectors for 24 h was blocked with the addition of CHX. Compared with the quantity of DUOX1 protein at the beginning of adding CHX (0 h), the degradation rate of mutant DUOX1 protein had no significant difference compared with wild type DUOX1 protein at 6, 12, and 24 h after transfection (P > 0.05; Figure 6).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.