Introduction
Congenital hypothyroidism (CH) is a prevalent neonatal endocrine disease resulting from insufficient thyroid hormone. If not diagnosed and treated promptly, it can lead to irreversible neurological deficits, and it affects approximately 1 in 2000–3000 live births worldwide [1]. Primary CH is mainly caused by abnormal thyroid gland development (thyroid dysgenesis, TD) and defects in thyroid hormone biosynthesis (dyshormonogenesis, DH). TD accounts for 80–85% of CH cases with various manifestations such as ectopic thyroid (50–60%), absent thyroid (20–30%), and thyroid dysplasia (5%) [2].
Currently, genetic research on TD is mainly focused on the monogenic form [3]. Well-established causes of TD include monogenic mutations in NKX2.1, PAX8, TSHR, and FOXE1. Furthermore, the extensive application of next-generation sequencing (NGS) in TD pathogenesis research has led to the discovery of additional causative genes, including NKX2-5, GLIS3, CDCA8, NTN1, TBX1, NNT, DUOX2, JAG1, and TUBB1. Most TD-related gene defects are also associated with other congenital malformations [2–6]. In fact, TD patients have shown a four-fold increase of extra-thyroidal congenital malformations [7], and the risk of TD is also increased in the context of syndromes. However, genetic causes are identified in less than 5% of TD cases, underscoring the critical need for further exploration of causative gene mutations [6].
TUBB1 (tubulin, beta 1 class VI) is located at 20q13.32; it encodes for β-tubulin, consisting of 451 amino acids, which interact with α-tubulin to form microtubules. As the cytoskeletal components of all eukaryotic cells, microtubules are involved in cell morphology, motility, differentiation, intracellular transport, signal transduction, and mitosis [8]. It has been reported that a TUBB1 defect might lead to thrombocytopaenia and platelet anisocytosis because TUBB1 plays a major role in proplatelet formation and platelet shape maintenance. In 2018, Stoupa et al. [9] first identified that TUBB1 mutation was the genetic cause of TD in a family, with further functional studies confirming that TUBB1 mutations impact thyroid development and function by disrupting microtubule assembly. In addition, the thyroid of Tubb1-/- mice exhibits early abnormal proliferation, delayed thyroid migration, and impaired thyroid hormone release. Nonetheless, further investigation is needed to validate TUBB1 as a potential candidate gene of TD in diverse populations with CH.
In a previous study we screened out the TUBB1 c.952C>T in 289 TD patients, all 4 patients who carried c.952C>T showed thyroid agenesis [10]. In combined bioinformatics analysis, MAF, multiple sequences alignment analysis, and American College of Medical Genetics and Genomics (ACMG) guidelines, c.952C>T mutation was assessed to be pathogenic. Thus, we aimed to further elucidate the pathogenic mechanisms of TUBB1 variants at the cellular level for the first time by evaluating the pathogenic effect of TUBB1 c.952C>T on the thyroid follicular epithelial cell line.
Material and methods
Patients
A total of 380 children with CH were enrolled in this study, comprising 91 cases of DH and 289 cases of TD (120 agenesis, 94 ectopy, and 75 hypogenesis). The ratio of male to female was 1:1.71. The average age of all subjects was 3.1 ± 1.9 years. Patients were selected from Shandong Province through a neonatal screening program conducted between 2007 and 2016. The criteria for inclusion were as follows: (1) neonates with thyroid-stimulating hormone (TSH) >9 µIU/mL[1] being recalled for further measurement within 2 months of birth, leading to a diagnosis of CH based on elevated TSH (> 4.2 μIU/ml) and low FT4 (< 12 pmol/L).; (2) patients were classified to DH and TD on the basis of thyroid ultrasound and Tc-99 m scans. The criteria for exclusion were as follows: patient with other congenital diseases (blood system, immune system, malignant tumour, and mental diseases). This study was approved by the Ethics Committee of the Affiliated Hospital of Qingdao University.
Sanger sequencing
DNA samples were extracted from the peripheral blood leukocytes of patients using a QIAamp Blood DNA Mini Kit (QIAGEN Company, America). All exons of TUBB1 along with their exon-intro boundaries were amplified by PCR [10]. The BigDye® Terminator Cycle Sequencing Kit and the automated sequencer ABI 3730XL were used for sequencing reaction of the PCR products. The sequencing data were analysed using Chromas software V.2.5.
Plasmids construction, cell culture, and transfection
The vector TUBB1-pcDNA3.1 and the control vector pcDNA3.1 were purchased from Synbio Technologies (Suzhou, China). The mutant expression vector TUBB1 c.952C>T-pcDNA3.1 was created using Fast Mutagenesis System (Transgen, Beijing). The primers used for PCR-based site-directed mutagenesis were as follows: Forward primer (5’-3’): ACAGTGGCCTGCATTTTCTGGGGCAAGAT; Reverse primer (5’-3’): AGAAAATGCAGGCCACTGTGAGGTAGCGG.
For the in vitro studies of mutations, Nthy-ori 3.1, an immortal human thyroid cell line, was purchased from Rongbai Biological (Shanghai, China) and cultured in RPMI 1640 medium (BI, Biological Industries) containing L-glutamine supplemented with 10% foetal bovine serum (FBS), 100 IU/mL of penicillin, and 100 µg/mL of streptomycin. The transfection experiments were conducted according to the lipofectamine 3000 Reagent Protocol (Thermo Scientific, USA).
RT–PCR analysis and western blotting
At 24h post-transfection, total RNA was extracted from Nthy-ori 3.1 cells using TRIzol agent (Invitrogen) following the manufacturer’s instructions. A total of 1000 ng of total RNA was mixed with HiScript II qRT SuperMix II (Vazyme) for the synthesis of complementary DNA. Each RT–PCR involved a 20 µL reaction system including 1 µL of cDNA, 0.5 µL of each forward and reverse primer, 10 µL of SYBR PCR master mix (Vazyme), and ddH2O. The primers used for qRCR were as follows: Forward primer (5’-3’): ATATGTGCCCCGAGCAGTCTT; Reverse primer (5’-3’): TCCGTGTAGTGGCCTTTGG. The reactions were incubated in 96-well plates on a real-time PCR system at 95°C for 2 min and 35 cycles of 95°C for 5 s and 60°C for 10 s. GAPDH served as the internal reference using the comparative threshold cycle method (2-△△CT) to quantify the expression level of the target gene.
At 48 h post-transfection, the cells were collected and lysed using the mixture of RIPA and PMSF with the ratio of 100:1 (Beyotime). The protein concentrations were determined using a BCA kit (Thermo Scientific), and 40 µg of protein was separated on 12% SDS–PAGE gels. Western blotting was performed according to the routine procedure using mouse monoclonal anti-beta 1 tubulin primary antibody (Sigma) (1:800) and goat anti-mouse secondary antibody combined with horseradish peroxidase (1:5000). The results were obtained using FluorChemQ (ProteinSimple) and the chemiluminescent HRP substrate (Immobilon Western; Millipore).
Cell proliferation assay
At 36 h post-transfection with pcDNA3.1, TUBB1- pcDNA3.1, and TUBB1 c.952C>T-pcDNA3.1, respectively, Nthy cells were successively digested, counted, and plated in 96-well plates (3 × 105 cells per cell), and then serum starvation treatment lasted for 12 h. After the cells were cultured for another 0 h, 24 h, 48 h, or 72 h, a solution of 10 µL of cck-8 (Vazyme) and 90 µL of DMEM was added to each plate, followed by incubation for 2 hours, The OD was measured at 450 nm using an enzyme reader (Thermo Fisher Scientific MULTISKAN GO).
Wounding healing assay
The 5 × 105 Nthy-ori3.1 cells were seeded into 6-well plates and transfected with pEGFP-N1, TUBB1-pEGFP-N1, and TUBB1R318W -pEGFP-N1, respectively, when the confluence of cells reached 70–90%. At 48 h after transfection, and when cells reached 100% confluence, a scratch was made in the centre of each well using a 200 μL pipet tip. After washing 3 times with PBS to remove the floating cells, the foetal bovine serum (FBS) in the culture medium was reduced to 2%. The cells were photographed at 0 h and 10 h using the cellSens Standard system. The migration rate was calculated according to the above data.
Statistical analysis
Statistical analysis was performed using a paired Student’s t test. p < 0.05 represents statistical significance. Graphs were prepared using Graphpad Prism 8.0.
Results
Genetic screening of TUBB1 variations in 380 CH patients
The present study enrolled a total of 380 CH patients, consisting of 289 cases of TD and 91 cases of DH; the ratio of male to female was 1:1.71. Four heterozygous variants were found in TUBB1: c.112G>A, p.Gly38Arg; c.582g>T, p.Glu194Asp; c.952C>T, p.Arg318Trp; and c.1045G>A, p.Val349Ile (Supplementary Figure 1). A summary of these variants is shown in Table 1. Multiple sequence alignment was conducted among 11 species (Supplementary Figure 2). In combined bioinformatics analysis, MAF, multiple sequences alignment analysis, and ACMG guideline, c.952C>T, p.Arg318Trp was assessed as a pathogenic mutation. There were 4 patients carrying the TUBB1 c.952C>T mutation, who all exhibited thyroid agenesis. Notably, pathogenic TUBB1 variants were not detected in patients with DH. To eliminate the possibility of additional pathogenic mutations in TD-related genes that may be present in these 4 patients, an extended genetic investigation was carried out utilising whole exome sequencing to target specific genes, including NKX2.1, PAX8, TSHR, FOXE1, NKX2-5, GLIS3, CDCA8, NTN1, TBX1, NNT, DUOX2, and JAG1. Employing consistent variant detection and filtering procedures, no pathogenic mutations were identified within the aforementioned gene set.
|
Nucleotide change |
Amino acid change |
avsnp150 |
Minor allele frequencies |
Prediction analysis |
Number of carriers |
Thyroid phenotype |
ACMG classification |
||||
|
GnomAD |
1000G |
SIFT |
Polyphen2 |
Mutation taster |
CADD |
||||||
|
c.112G>A |
p.Gly38Arg |
rs144337011 |
0.00414 |
0.00225 |
0.0/D |
0.709/P |
1/D |
26.7 |
6 |
Variable |
LB |
|
c.582G>T |
p.Glu194Asp |
rs186106638 |
0.00005 |
0.0002 |
0.001/D |
0.057/B |
0.0999/D |
10.29 |
2 |
Ectopy |
VUS |
|
c.952C>T |
p.Arg318Trp |
rs121918555 |
0.00003 |
0.0002 |
0.0/D |
1.0/D |
1/D |
34 |
4 |
Agenesis |
P |
|
c.1045G>A |
p.Val349Ile |
rs115253190 |
0.00775 |
0.0074 |
0.015/D |
0.014/B |
1/D |
12.97 |
1 |
Hypoplasia |
VUS |
The c.952C>T mutant has lower expression compared with wild type
To evaluate the effect of the c.952C>T mutation on mRNA translation and protein expression, we transiently transfected the plasmids (pcDNA3.1, TUBB1-pcDNA3.1, TUBB1c.952C>T-pcDNA3.1) into nthy-ori 3.1 cells and measured the levels of TUBB1 mRNA and protein in each experimental group. RT-PCR analysis showed that the TUBB1 mRNA in the c.952C>T group was significantly lower than that in the WT group (p < 0.05) (Fig. 1A). Consistent with the mRNA findings, western blot analysis displayed an obvious decrease in TUBB1 protein levels in the c.952C>T group (Fig. 1B). Image J was used to analyse the band intensities from the western blot, revealing statistically significant differences between the c.952C>T and WT groups (p < 0.01) (Fig. 1C).
The effect of c.952C>T mutation on thyroid cell proliferation and migration
To further examine the effect of the c.952C>T mutant on proliferation and migration of Nthy-ori 3.1 cells, cell counting kit 8 (CCK8) and wound healing assay were used. At 48 h, 72 h, 96 h, and 120 h post-transfection, the total number of cells was detected. The results revealed that the proliferation trend in the c.952C>T mutation group was notably suppressed compared to the WT group, with a statistically significant difference observed at 72 h and 96 h (p < 0.05) (Fig. 2A). The wound healing was pictured and measured at 0 h and 10 h after scratch. The relative migration distance of the NC, WT, and c.952C>T groups was 30.1%, 43.1%, and 35.7%, respectively. The c.952C>T mutation appears to have a restraining effect on the migration of thyroid cells, although it was not statistically significant (Fig. 2BC).
Discussion
Thyroid morphogenesis is a coordinated spatial and temporal process from bud to gland, mainly including 6 stages: formation of the thyroid placode (D20–22 in humans, E8.5–9.5 in mice); conversion of the placode to the thyroid bud (D24–28, E9.5–10); downward migration of the thyroid primordium to a pretracheal position (D30–40, E11.5–13); bifurcation of the primordium (D37, E12.5–13.5); formation of the left and right thyroid lobes (D45–50, E13.5); and folliculogenesis (D60, E15–16) [6, 11, 12]. Any factors affecting the development and migration of the thyroid during these critical stages can result in thyroid dysplasia [13].
The thyroid transcription factors NKX2.1, FOXE1, PAX8, and HHEX were found to be co-expressed during specific developmental stages in mice (E8–8.5) and humans (E20–22), which act both individually and in concert to regulate the expression of thyroid specific genes and determine the specification of thyroid progenitors from the ventral endoderm [14, 15]. In the mouse model, the absence of Nkx2-1 results in a normal thyroid bud that subsequently disappears around E10.5–11 [16], while the thyroid cell precursors in PAX8-/- mouse disappear at approximately E11–11.5 [17], indicating the crucial roles of Nkx2.1 and Pax8 in cell survival. Foxe1 plays a significant role in the survival and migration of thyroid buds, as Foxe1-/- mouse showed thyroid athyreosis or ectopy [6, 18]. Hhex-/- embryos exhibit the arrest of thyroid development at E9.5, demonstrating potential involvement in cell proliferation [15], although mutations in HHEX have not been identified in TD patients. Additionally, the activation of TSHR by TSH at E15 is essential for maintaining the size and structure of the thyroid gland. TSHR mutations were the most frequent genetic cause of TD, with a prevalence of 4.3% [3, 19]. However, less than 5% of all TD cases were found carrying causative mutations in the 4 aforementioned genes.
Léger et al. [20] reported a 7.9% incidence of first-degree relatives in children with TD. Previous studies have shown that the familial occurrence of TD varies between 2% and 12%, implying the involvement of genetic factors [21, 22]. The extensive use of the next generation sequencing (NGS) has proven advantageous in identifying the root cause of TD cases. TUBB1 was first identified as the novel candidate gene for TD by Stoupa et al. [9] in 2018. In Tubb1-/- mouse, there were notable increases in thyroid anlage surface area and proliferation ratio at E9.5, a slight delay in thyroid migration at E11.5, and fusion of the midline thyroid primordium with the ultimobranchial bodies at E13.5, leading to thyroid hypoplasia. Patients who carried TUBB1 mutations showed a range of thyroid phenotypes, from thyroid ectopic to asymmetry, hypoplasia, and hemithyroid [9]. Considering that the genetics of TD in animal models was not completely consistent with that in humans, the pathogenesis of TUBB1 in TD and the association between TUBB1 mutation and TD are still needed for further verification in human subjects.
In the present study, we screened out 4 heterozygous variants of TUBB1 among 380 CH patients, including c.112G>A, c.582g>T, c.952C>T, and c.1045G>A. After conducting an extensive series of bioinformatics analyses, the only variant identified as pathogenic was c.952C>T. Consequently, we proceeded to perform additional analyses and functional experiments on this pathogenic variant. Numerous studies have identified a shared genetic basis between TD and DH, with genetic factors such as DUOX2, TG, TSHR, JAG1, FOXE1, and GLIS3 showing overlap [23, 24]. We screened the TUBB1 pathogenic variant in TD and DH patients, and the results indicate that TUBB1 variants confer genetic susceptibility to TD but not DH. Utilising data from GnomAD as the control group, burden test analysis demonstrated a significant association between TUBB1 variants and TD (odds ratio [OR] = 5.323, p = 0.008). The mutation rate of TUBB1 in TD patients was 1.38% (4/289), slightly higher than the 1.1% reported in the previous research [9].
The TUBB1 protein contains 3 domains comprising 451 amino acids. The amino-terminal domain (residues 1–206), is responsible for regulating guanosine triphosphate (GTP) binding and hydrolysis; the intermediate domain (207–384) is involved in hydrolysis and interacting with monomers within and between protofilaments; and the carboxy-terminal domain (385-451) is essential for binding to microtubule-associated proteins [25]. Among these TUBB1 mutations found in TD patients, c.479C>T, c.318C>G, and c.35delG are all situated in the N- terminal, while the mutation (c.952C>T, p.R318W) is located in the intermediate domain. The arginine at position 318 is highly conserved, and the replacement with tryptophan alters the hydrophobic parameter from -4.5 to -0.9, which might reduce the stability of mutant proteins. In vitro functional analysis has demonstrated that the mutation c.952C>T affects the expression of TUBB1 gene at both the mRNA and protein level.
Microtubules participate in executing key cellular functions, such as intracellular transport of cargos, cell migration, division, proliferation, and apoptosis by polymerisation and depolymerisation cycle [26]. The mutation c.952C>T was initially identified in a patient with congenital macrothrombocytopaenia, and in vitro studies in Chinese hamster ovary (CHO) cells demonstrated that the mutation dominantly affects protein stability and microtubule assembly [27]. Our findings indicate that the c.952C>T mutation could significantly inhibit the proliferation of a thyroid cell line. Consistently, all 4 patients harbouring the c.952C>T mutation showed thyroid agenesis, underscoring its detrimental effect on cell survival during thyroid development. Notably, these 4 patients in our study cohort displayed normal platelets (Supplementary File — Tab. S1, Suppl. Fig. S3), suggesting variable penetrance of the mutation, while the c.952C>T mutation did not appear to affect the migration of thyroid cells. In fact, the migration of wild-type thyroid progenitors is intricately regulated by epithelial cadherin and characterised by collective migration [28], which relies on the matrix environment for the directional movement and cell migration [13]. Consequently, accurately assessing the effect of the c.952C>T mutation on migration ability through wound healing assay may be limited by the absence of a matrix environment, necessitating further experimental investigations.
In conclusion, we demonstrated that TUBB1 variants confer genetic susceptibility to TD, but not to DH. The pathogenic variant in TUBB1 was identified in 1.38% (4/289) of our Chinese TD patient cohort. Burden test analysis confirmed the correlation between TUBB1 variants and TD. Functional experimental findings revealed that the c.952C>T mutation decreased the expression of TUBB1 gene and inhibited the proliferation of thyroid cells. Nevertheless, further research and investigation are required to fully elucidate the pathogenic mechanism of TUBB1 variants in TD.
Data availability statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.
Ethics statement
This study was approved by the Ethics Committee of the Affiliated Hospital of Qingdao University (QYFY WZLL 28515). Written informed consent was obtained from all subjects.
Author contributions
Conceptualisation: S.L., M.L.; Data curation: Y.W., Fa.W.; Formal analysis: Miaomiao Li; Funding acquisition: M.L.; Fa.Wang, S.L.; Methodology: Fe.W.; Project administration: C.S.; Writing original draft: C.S.; Writing review and editing: M.L.
Funding
The present work was supported by grants from the Natural Science Fund Project of Shandong Province (ZR2024QH075) and the Qingdao Scientific Technology Demonstration Project of Benefiting the People (24-1-8-smjk-5-nsh) and the National Natural Science Foundation of China (82071683, 82270829).
Acknowledgments
We thank all the patients and team members for participating in the study.
Conflicts of interest
The authors declare that there are no conflicts of interest.
