Development of an athyroid mouse model using 131I ablation after preparation with a low-iodine diet

We optimized the protocol for thyroid ablation in living mice using radioactive iodine (RAI) and a low-iodine diet (LID). To examine the effect of LID on thyroid ablation, mice were randomly divided into 4 groups: Vehicle, 131I 2.775 MBq, 131I 5.55 MBq, and LID + 131I 2.775 MBq. The LID group was fed a LID for up to 7 days and then mice in the 131I 2.775, 131I 5.55, and LID + 131I 2.775 MBq groups were intravenously administrated with 131I, respectively. Scintigraphy imaging with 99mTc pertechnetate was performed once in 2 weeks for 4 weeks. After establishment of athyroid mice, control or athyroid mice were injected with human anaplastic thyroid cancer cells co-expressing sodium iodine symporter and enhanced firefly luciferase (ARO/NF) to evaluate RAI uptake. Scintigraphy imaging with 99mTc pertechnetate was performed with ARO/NF tumor-bearing mice. Scintigraphy imaging showed decreased thyroid uptake in the LID + 131I 2.775 MBq group compared to other groups. Scintigraphy images showed that tumor uptake was statically higher in athyroid mice than in control mice. These data suggest that these optimized conditions for thyroid ablation could be helpful to establish an in vivo mouse model.

I thyroid ablation has not yet been optimized [21][22][23] . The reported 131 I administration doses for thyroid ablation in mice is variable 21,[24][25][26] . The body iodine pool is one of the important factors associated with successful radioiodine ablation, and low-iodine diets (LIDs) have been widely used to decrease the body iodine pool. The effect of LIDs on radioiodine ablation of patients with thyroid cancer has been investigated, but few studies have evaluated the effect of LIDs in athyroid models using radioiodine. The current study established an athyroid mouse model using 131 I administration and also optimized the 131 I thyroid ablation protocol in this mouse model.

Results
Influence of diet on thyroidal 131 I uptake. Before the experiment, the mice were divided into 4 treatment groups; (1) the vehicle group, (2) 131 I 2.775 MBq, (3) 131 I 5.55 MBq, (4) and LID + 131 I 2.775 MBq. To evaluate the effect of LID effect in these mice, we monitored thyroid 131 I uptake by gamma camera imaging. The LID + 131 I 2.775 MBq group had considerably higher 131 I accumulation than groups without LID, as shown in Fig. 1. There were no significant differences in body weight among the groups ( Supplementary Fig. 2). 99m Tc pertechnetate scintigraphy imaging. To evaluate the effects of thyroid ablation according to 131 I treatment, we monitored gamma camera images of the thyroid gland using 99m Tc pertechnetate. There were no significant differences for 2 weeks after injection of 131 I to ablate the thyroid. However, starting at 3 weeks, 99m Tc pertechnetate accumulation in the thyroid gland gradually decreased. As shown in Fig. 2 Fig. 4a).

125
I bio-distribution study. Bio-distribution of 125 I was evaluated in order to validate thyroid uptake of radioactivity after establishment of the thyroid ablation model. As shown in Fig. 4b, thyroid uptake in the LID + 131 I 2.775 MBq group was significantly lower compared with uptake in the vehicle group (vehicle, LID + 131 I 2.775 MBq; 34.7 ± 7.8 vs. 5.11 ± 2.79% ID/g, respectively). Interestingly, uptake in other organs (lungs, heart, liver, stomach, etc.) was approximately 3-10-fold higher in the athyroid mice compared to the vehicle group. In vivo 99m Tc pertechnetate imaging of NIS expressing tumor. In vivo scintigraphy imaging with 99m Tc pertechnetate was performed to determine whether uptake of NIS-expressing tumors was affected by thyroid ablation. To normalize tumor uptake of 99m Tc pertechnetate, the bioluminescent signal of the tumor was    Immunohistochemical staining for thyroglobulin and CD68 proteins. On day 28, thyroid tissues were excised for immunohistochemical analysis and stained with thyroglobulin-or CD68-specific antibodies. In the LID + 131 I 2.775 MBq group, no staining was observed in the thyroid gland on thyroglobulin immunostained slides (x100) and darkly stained CD68-positive macrophages (inlet, x400) were found more frequently in the ablated thyroid gland (arrow) compared with the normal thyroid gland of control group (x100). uptake in the tumor region compared with tumors in the control group ( 125 I tumor uptake of control and athyroid groups: 10.6 ± 1.18 vs. 19.8 ± 2.67%ID/g, respectively, Fig. 6b). The athyroid mice group had lower accumulation of 125 I in the thyroid region compared to the control group ( 125 I thyroid uptake in the control and athyroid groups: 18.6 ± 3.81 vs. 1.76 ± 0.38, %ID/g, respectively).

Discussion
Temporary LIDs are recommended before radioactive iodine treatment or scanning in thyroidectomized thyroid cancer patients. LIDs are widely used to deplete body iodine pools, which increases the availability of radioactive iodine to NIS-expressing tissues [27][28][29][30] . The American Thyroid Association (ATA) recommends an LID defined by an intake of <50 μg/day for 1-2 weeks before 131 I administration, while the British Thyroid Association recommends an LID for 2 weeks before 131 I administration 28,31 . However, there is no consensus about the necessity of LIDs for ablation of the entire thyroid gland, such as 131 I treatment for Graves' disease, and there are limited reports regarding the relationship between LID and 131 I thyroid ablation outcomes in animal models.
In the current study, we developed an effective 131 I thyroid ablation by performing an in vivo experiment to investigate the effects of an LID on radioiodine uptake in the thyroid gland. Prior to in vivo experiments, the determination of the minimum 131 I dose for effective thyroid ablation is required. However, 131 I doses ranging from 1.04 to 37 MBq (28-1000 μCi) were used in previous thyroid ablation reports 21,[24][25][26] . The 131 I ablation dose can produce systemic and local adverse effects, including weight loss, salivary gland dysfunction, tracheal damage, and breathing difficulty 32,33 . Higher doses tends to show higher ablation success rates, but more frequent adverse effects 34 . Therefore, it is important to determine the optimal dose of 131 I for successful thyroid ablation.
In order to optimize the dose of RAI for effective thyroid gland ablation, mice were divided into 4 groups: vehicle, 131 I 2.775 MBq; 131 I 5.55 MBq; and an LID + 131 I 2.775 MBq. Scintigraphy imaging was performed with 99m Tc pertechnetate in all groups before 131 I ablation to check the baseline status of the thyroid gland. Accumulation of 99m Tc pertechnetate or radioiodine was observed in tissues that endogenously express NIS, including the thyroid, stomach, and salivary glands, as well as the urinary bladder, which eliminates the radionuclide from the body. During administration of the LID, there were no significant differences in body weights between the groups. There were also no significant differences in body weight between the RAI ( 131 I 2.775 and 5.55 MBq) and control groups. One week after administration of the LID, all mice were intravenously administrated 131 I 2.775 or 5.55 MBq and gamma camera imaging was performed. Thyroid uptake of 131 I on the gamma camera image was, at first, used as an indicator of successful intravenous injection of 131 I and to quantitatively assess thyroid 131 I accumulation. Accumulation of 131 I in the thyroid gland was visualized in all groups, but more intense thyroid uptake was observed in the LID + RAI group compared to uptake in the other groups, which suggests that depleting the body iodine pool via an LID resulted in enhanced 131 I accumulation in the thyroid gland, which increases 131 I ablation by delivering high radiation doses to the thyroid gland.
In accordance with thyroid uptake of 131 I, scintigraphy imaging with 99m Tc pertechnetate 4 weeks after 131 I ablation revealed significantly lower thyroid uptake of 99m Tc pertechnetate in the LID + RAI group compared with uptake in the non-LID groups and we further measured thyroid size and serum T4 levels to confirm the thyroid ablation effect, these results revealed decreased thyroid size and T4 level in LID + RAI group compared with other groups. In addition, we investigated the biodistribution of 125 I and histological changes in the thyroid region. The results of the bio-distribution assessment using 125 I showed that the LID + RAI group had lower radioactivity in the ablated thyroid gland and higher radioactivity in other organs, including the lungs, heart, liver, and stomach, compared to levels in the vehicle group. Immunohistochemical analysis using thyroglobulin staining confirmed more prominent destruction of thyroid follicles in the LID + RAI group. In contrast, we observed increased density of CD68-positive macrophages in the thyroid gland. We assumed that the dead follicular cells and cellular debris resulting from the radioiodine ablation are phagocytosed by macrophages. The degree of CD68 expression in damaged thyroid tissue might be related to the degree of damage in thyroid follicular cells 35 .
The results of 99m Tc pertechnetate scans for NIS-expressing tumors revealed significantly increased uptake in the tumors and other organs in athyroid mice compared to control mice. These scans revealed almost faint uptake in the thyroid bed of the athyroid mice, indicative of successful 131 I thyroid ablation; the thyroid ablation also significantly increased tracer uptake in the NIS-expressing tumor xenograft. The biodistribution of radioiodine in the athyroid mouse model is more similar to that observed in thyroidectomized thyroid cancer patients than euthyroid mouse models; therefore, the athyroid mouse model might be more appropriate for developing radioiodine therapy than other euthyroid mouse models.
Overall, we successfully optimized conditions for thyroid ablation with a week of LID and low dose of RAI. Our results could be useful for investigation of RAI refractory thyroid cancers in mouse models that mimic real clinical situations by removing the thyroid gland before 131 I therapy.

Materials and Methods
Animals. Female Balb/c nude mice, 5.5 weeks old and weighing averages of 18.7 ± 0.46 g (mean ± SD) were purchased (Hamamatsu, Shizuoka, Japan). The mice were maintained under specific pathogen-free conditions in order to adapt to the experimental conditions for 1 week before starting the experiment. The animals were maintained at room temperature (20-25 °C) at 40-70% relative humidity.
Experimental protocol. All procedures were reviewed and approved by the Kyungpook National University Animal Care and Use Committee, and performed in accordance with the Guiding Principles for the Care and Use of Laboratory Animals. Mice were randomly divided into different treatment groups for thyroid ablation: (1) the vehicle group (n = 10), which received no 131 I and a LID, (2) 131 I treatment group (n = 10), which received 131 I 2.775 MBq by intravenous injection, (3) 131 I treatment group (n = 10), which received 131 I 5.55 MBq by intravenous injection, (4) 131 I treatment combined with LID group (n = 10), which received a LID for 7 days prior to radioiodine administration as well as 131 I 2.775 MBq by intravenous injection. The body weights of the mice were measured once a week before gamma imaging. Blood samples were taken to estimate T4 concentration at days 0, 14, and 28. Gamma camera images of the thyroid were obtained following intravenous injection of 99m Tc pertechnetate (18.5-22.2 MBq) at days −1, 14, and 28 after 131 I administration, respectively. At the end of the experiments, the animals were sacrificed and 125 I bio-distribution was performed. The tissues were analyzed by hematoxylin and eosin (H&E) staining and immunohistochemistry. A schematic of thyroid ablation procedure is shown in Supplementary Fig. 1. 99m Tc pertechnetate whole-body imaging. To evaluate the efficacy of the thyroid ablation, whole-body scans were performed with 99m Tc-pertechnetate. All mice were administrated with 99m Tc-pertechnetate (18.5-22.2 MBq) and static gamma camera images were obtained for 20 min using a 2 mm pinhole collimator (Infinia II, GE Healthcare, Milwaukee, WI, USA). Additionally, 131 I gamma camera images were obtained at that time of 131 I injection to confirm thyroid uptake of 131 I. The mice were maintained under isoflurane (Forane, ChoongWae Co., Seoul, Korea) anesthesia during injection and scanning. We set regions of interest (ROIs) in the thyroid region of the mice to obtain a total count of thyroid uptake. Background activity was measured using circular ROI of the same size on the head. Quantification of thyroid uptake was defined as the % of thyroid gland (TG)/background (BG), which was calculated as the total count of thyroid uptake divided by the total background count.
Sonography of the thyroid gland. Ultrasonography of the thyroid gland was performed to measure size of the gland non-invasively using Prospect 3.0 ultrasound imaging system (S-Sharp Corporation, New Taipei, Taiwan). Mice were anesthetized with 1-2% isoflurane in 100% O 2 and then placed in ventral side up position. B mode imaging with 40 MHz transducer was acquired on thyroid area of mice after replenishment of the ultrasound gel on the neck of mice. Serum T4 measurement. Serum T4 measurements were made using serum samples pooled from retro orbital blood puncture, as described previously 36 . On days −1, 14, and 28, serum T4 concentrations were estimated by radioimmunoassay using RIA-gnost-T4 Kits (Cisbio Bioassays, France) according to the manufacturer's instructions. Serum samples and standard solutions with radiolabelled tracer were incubated in antibody-coated test tube. After incubating for 2 h, the solution was removed. A gamma counter (Cobra II, Packard, Perkin Elmer, MA, USA) was used to measure the radioactivity adhering to the tube.
Histologic examination. The mice were sacrificed by cervical vertebrae dislocation on day 29 after 131 I administration. After the anterior neck was incised vertically, the thyroid lobes were excised and fixed with 4% formalin overnight. The specimens were embedded in paraffin, sectioned into 4-µm-thick sections and mounted on slides. The specimen section slides were deparaffinized and stained using H&E. Immunohistochemical staining for thyroglobulin (Abcam, Cambridge, MA, USA) at a 1:250 dilution and CD68 (Abcam) at a 1:200 dilution were performed. The H&E and immunohistochemical stained slides were analyzed by light microscopy.
Bio-distribution of 125 I. The mice were intravenously injected with 1.85 MBq of 125 I. Four hours later, blood samples were taken from all mice and the mice were sacrificed. Ex vivo radioactivity measurements were taken from the lungs, heart, liver, stomach, spleen, intestine, kidney, muscle, and thyroid using a gamma-counter (Cobra II). The data were expressed as the percentage injected dose per gram tissue (%ID/g). Statistical analysis. All data are expressed as means ± standard deviation (SD) and were statistically analyzed by t-test using GraphPad Prism 5, version 5.01 (GraphPad Software, Inc. USA). P values less than 0.05 were considered statistically significant.