ZEBRAFISH AS MODEL SYSTEM TO STUDY THE ROLE OF THE GLIS3 TRANSCRIPTION FACTOR IN THE PATHOGENESIS OF CONGENITAL HYPOTHYROIDISM

Congenital Hypothyroidism (CH) is the most common congenital endocrine disease that can be broadly classified as failure of the gland to develop normally (dysgenesis) or inadequate thyroid hormone (TH) production from eutopic thyroid gland (dyshormonogenesis). Biochemically, CH patients are characterized by low circulating levels of T4 and T3 associated with an increased production of TSH due to the feedback mechanisms controlling the hypothalamic-pituitary-thyroid (HPT) axis. CH is considered a disease with a strong genetic component, but with a largely missing explanation for its heritability. At present, genetic variants in candidate genes explain only the 10% of CH cases.
The extraordinary progresses in high-throughput screening technologies lead to the identification of several genes involved in the thyroid development and TH synthesis. Recently, the transcription factor GLI-Similar protein 3 (GLIS3) has emerged as a new candidate gene for CH, but its role in thyroid development and function remains largely unexplored. GLIS3 is a member of the Kruppel-like zinc-finger transcription factors that can acts as activator or repressor of gene expression. Homozygous and compound heterozygous GLIS3 variants have been associated with NDH syndrome, characterized by neonatal diabetes (T1D and T2D), CH and polycystic kidney. Additionally, several missense heterozygous GLIS3 variants have been also identified in cohorts of patients with isolated CH. Interestingly, heterozygous GLIS3 variants are associated with other mutations in genes involved in thyroid functioning, supporting the hypothesis of the oligogenic CH origin. In both NDH and isolated CH patients, the thyroid disease is extremely heterogeneous. In fact, most of the affected cases present variable thyroid dysgenesis (athyreosis, thyroid hypoplasia and ectopy), whereas dyshormonogenesis with eutopic in situ gland is also reported. Evidences from Glis3 knockout mice indicate a relevant role of Glis3 in TH biosynthesis and postnatal follicle proliferation, with a mechanism of action downstream of the TSH/TSHR signalling and thyroid cell proliferation. However, no significant thyroid developmental defects were observed in this particular mice model. Since GLIS3 mutations are variably associated with thyroid dysgenesis, the aim of this study is to gain insight on GLIS3 activity during the early steps of thyroid development, using zebrafish as a model system. In zebrafish (zf), we observed the expression of glis3 transcript from the early developmental stages onwards, with a particularly evident signal at 1 dpf (day post fertilization) in the pharyngeal endoderm, the embryonic tissue that will give rise to endocrine organs, like thyroid and pancreas, and in the pronephric ducts. In apparent contrast with mouse data, glis3 is absent in the differentiated thyrocytes of zebrafish embryos at 2-3 dpf. Transient knockdown zebrafish embryos (called glis3_MOs), obtained by the microinjection of specific glis3 morpholinos, revealed a reduced expression of the early thyroid markers nkx2.4, and pax2a, at 1dpf. The defective specification of the thyroid primordium at this early developmental stage was not associated with reduced proliferation or increased apoptosis of thyroid precursors, thus indicating that this phenotype was likely due to alterations in the commitment of endodermal cells toward the thyroid fate. Such defect resulted in a reduction in size of the differentiated thyroid precursors with a diminished expression of both tg and slc5a5, and later on in the number of functional thyroid follicles. At 5dpf, the stage in which the thyroid gland is functional and responds to the HPT-axis, decreased levels of T4 associated with tshba elevation were also seen in glis3_MOs. The specificity of the thyroid defects was confirmed by the rescue phenotype after co-injection of the wild-type zf-glis3 transcript (WT mRNA) and the glis3 morpholino. In contrast with morpholino knock-down, the overexpression of the glis3 mRNA leads to an increase of the number of the thyroid precursors at 1dpf, thus leading to the differentiation of a larger thyroid tissue and an elevation of T4 levels at 5 dpf. Taken together, our results demonstrate that glis3 acts at the endodermal level controlling the commitment of endocrine precursors, potentially representing a required factor for the early specification of thyroid primordium. Given the current knowledge, GLIS3 is reported to interact with a pivotal element within the Sonic hedgehog (Shh) pathway. Previous experiments in the mouse model indicated a potential role for such pathway in thyroid development, but the underlying mechanisms and the stage of Shh action in this context are still elusive. Our experiments revealed that the expression of shha (the zf homologous of human Shh) was significantly reduced in the pharyngeal endoderm of glis3_MOs. Consistently, the injection of a morpholino against shha abolished the expression of glis3 in the endodermal layer, as well as the treatment with Cyclopamine (a Shh-antagonist) caused a reduced or absent expression of glis3, thus confirming the co-involvement of shha and glis3 during the thyroid cell specification. Interestingly enough, the overexpression of glis3 mRNA failed to rescue the thyroid defects in the Cyclopamine-treated embryos, suggesting that glis3 would act as downstream effector of the Shh pathway. In conclusion, this is the first evidence of the possible interaction between Shh and Glis3 during the early specification of thyroid primordium. Since the mechanisms involved in such a delicate event are largely unknown, our data provide important insights as glis3 appears as an endoderm factor required for the commitment of endoderm cells toward the thyroid fate. These data shed new light into the molecular mechanisms potentially involved in CH pathogenesis.