Thyroid regeneration and carcinogenesis

Division Of Basic Sciences - Nci

Investigators

Linked publications, trials & patents

Abstract

(1) Thyroid regeneration. Even though thyroid is considered as a dormant organ, it retains a capacity to regenerate under certain conditions. We used partial thyroidectomy as an acute thyroid injury model to study thyroid regeneration. After partial thyroidectomy, we found a cluster of thin and elongated cells in extrafollicular area near the tracheal cartilage and muscle, some of which were positive for thyroid differentiation marker, NKX2-1. This suggested that these cells may be a precursor to thyroid follicular cells. In fact, a newly formed follicle-like structure with NKX2-1 positive cells were found connected at the end or in the middle of the cluster. This NKX2-1-positive cells cluster were also observed after partial thyroidectomy of TSH receptor (Tshr)-null mice, suggesting that TSH signaling is not required for at least this type of thyroid regeneration. When we transplanted bone marrows obtained from GFP transgenic mice, and the recipient mice subjected to partial thyroidectomy, GFP-positive cells were found in the elongated cluster of cells in the recipient mice. This suggested that this NKX2-1-positive cluster of cells could be at least partly derived from bone marrow cells. In order to further understand the relationship between the thin elongated cells and immune cells, spatial gene expression profiling was carried out using laser microdissection of a cluster of thin elongated cells, the surrounding tissues, and the normal thyroid tissues, followed by RNA sequencing. The RNAseq analysis demonstrated that many immune-related genes are highly expressed in the NKX2-1-positive cell cluster. In situ hybridization showed that many immune-related genes such as Igha, Igkc, H2-Aa, H2-Ab1, H2-Eb1, Cd74, and Cxcl12 were highly expressed in the NKX2-1-positive cell cluster, followed by the area next to the NKX2-1-positive cell cluster and then the normal thyroid follicular area, whereas the expression of NKX2-1 was in the opposite direction; the normal follicular area > the area next to the NKX2-1-positive cell cluster > NKX2-1-positive cell cluster. These results are in agreement with our hypothesis that immune cells may be the origin of thyroid follicular cells at least in this mode of thyroid regeneration. Immunohistochemistry demonstrated that CD74 was the highest among the aformentioned immune related proteins, and ubiquitously expressed in the NKX2-1-positive cell cluster. In order to prove our hypothesis, single cell RNA sequencing was carried out. The results show that both Nkx2-1 and Cd74 appear to be located in the thyroid cluster on UMAP, suggesting that both genes may be co-expressed in a thyroid follicular cell or a precursor to a thyroid follicular cell. In situ hybridization results support this conclusion. Double immunohistochemisty with NKX2-1 and CD74 is currently ongoing to further validate this result. If indeed CD74-expressing cells become thyroid follicular cells, this is a novel paradigm shifting new finding. The results will provide a novel way to understand the physiology and diseases of the thyroid, and the way to restore thyroid functions in humans after loss of thyroid tissues due to various thyroid diseases including cancer. (2) Thyroid carcinogenesis. Multi-step carcinogenesis process is one of the ways carcinogenesis occurs in many tissues, including the thyroid. In order to understand whether NKX2-1 plays any role during thyroid carcinogenesis, we used three different mouse thyroid carcinogenesis models; radiation and low iodine diet-induced, thyroid follicular cells specific KrasG12D-induced, and thyroid follicular cells specific HrasG12V-induced. (2-1) Radiation and low iodine diet-induced carcinogenesis. Thyroid-specific Nkx2-1 conditional knockout (cKO) mice were used for this study that developed thyroid adenoma under exposure to iodine deficiency diet, and adenoma and carcinoma in higher frequency under a combination of radiation and iodine deficiency as compared to control diet. Radiation only did not result in any adenoma or carcinoma. Iodine deficiency caused oxidative stress, which subsequently resulted in DNA damage, leading to transformation of thyroid follicular cells. The combination of radiation and iodine deficiency caused higher production of reactive oxygen species (ROS) in the thyroid of cKO mice that have naturally elevated basal levels of oxidative stress. This was accompanied by a feedback induction of antioxidants, superoxide dismutase 3 (SOD3) and glutathione peroxidase 2 (GPX2). Elevated expression of Slc5a5 (encoding sodium/iodide symporter, also known as NIS) and Tpo (encoding thyroid peroxidase) was also observed in cKO mouse thyroids under iodine deficiency, suggesting that the disturbance of thyroid hormone synthesis may lead to ROS production in the thyroid. In addition, Acox2 (encoding Acyl-coA Oxidase 2) was identified as an up-regulated oxidase gene in cKO thyroids, which produces hydrogen peroxide through beta-oxidization of fatty acid and could also have contributed to higher ROS production in cKO thyroid. We propose a novel role for NKX2-1 as a regulator of ROS production homeostasis in the thyroid. Its disturbance would dispose follicular cells more vulnerable to the ROS-producing carcinogens. (2-2) KrasG12D-induced carcinogenesis. Mice used in this study were Nkx2-1(fl/fl);Scgb3a2(fl/fl);KrasG12D;TPO-Cre and those having each one of Nkx2-1 or Scgb3a2 gene deleted, or both genes deleted under the expression of KrasG12D in thyroid follicular cells-specific fashion. In this carcinogenesis model, the effect of Scgb3a2 gene deletion was also examined as described in Project 2. Histopathological examination of thyroids was carried out on 1 year old of mice and those that died earlier than 1 year due to thyroid lesions. The results confirmed that without KrasG12D expression, there was almost no adenoma/carcinoma developed in all groups of mice, confirming oncogenic activity of KrasG12D. Nkx2-1 gene deletion resulted in increased incidence of carcinoma, while Scgb3a2 gene deletion resulted in increased incidence of adenoma, suggesting that NKX2-1 serves as tumor suppressor while SCGB3A2 as tumor promoter. In order to understand how NKX2-1 and/or SCGB3A2 interact with the KRAS signaling pathway to interfere or promote thyroid carcinogenesis, single cell RNA sequencing was carried out and the analysis is ongoing. (2-3) HrasG12V-induced carcinogenesis In order to examine the effect of loss of Nkx2-1 gene expression in HrasG12V-induced thyroid carcinogenesis, Nkx2-1(fl/fl);HrasG12V;TPO-Cre mice were used, which expressed HrasG12V in thyroid follicular cells specific fashion. In these mice, adenoma incidence was higher when they were fed diet of LowI (low iodine)+PTU (propyl thiouracil), compared with control diet or LowI+PTU-fed wild-type mice, however, carcinoma was almost never developed. This suggested that the HrasG12V does not exhibit oncogenic role in thyroid carcinogenesis, and the loss of Nkx2-1 only promoted earlier events such as hyperplasia and adenoma, but did not help augmenting the oncogenic role of HrasG12V.