Several types of animal models of Hashimoto’s thyroiditis have been established, as has spontaneous autoimmune thyroiditis in the NOD mouse. H-2h4 mice are a simple and reliable model for HT induction. This article describes this approach and evaluates the pathological process for a better understanding of the SAT murine model.
In recent years, Hashimoto’s thyroiditis (HT) has become the most common autoimmune thyroid disease. It is characterized by lymphocyte infiltration and the detection of specific serum autoantibodies. Although the potential mechanism is still not clear, the risk of Hashimoto’s thyroiditis is related to genetic and environmental factors. At present, there are several types of models of autoimmune thyroiditis, including experimental autoimmune thyroiditis (EAT) and spontaneous autoimmune thyroiditis (SAT).
EAT in mice is a common model for HT, which is immunized with lipopolysaccharide (LPS) combined with thyroglobulin (Tg) or supplemented with complete Freund’s adjuvant (CFA). The EAT mouse model is widely established in many types of mice. However, the disease progression is more likely associated with the Tg antibody response, which may vary in different experiments.
SAT is also widely used in the study of HT in the NOD.H-2h4 mouse. The NOD.H2h4 mouse is a new strain obtained from the cross of the nonobese diabetic (NOD) mouse with the B10.A(4R), which is significantly induced for HT with or without feeding iodine. During the induction, the NOD.H-2h4 mouse has a high level of TgAb accompanied by lymphocyte infiltration in the thyroid follicular tissue. However, for this type of mouse model, there are few studies to comprehensively evaluate the pathological process during the induction of iodine.
A SAT mouse model for HT research is established in this study, and the pathologic changing process is evaluated after a long period of iodine induction. Through this model, researchers can better understand the pathological development of HT and screen new treatment methods for HT.
Hashimoto's thyroiditis (HT), also known as chronic lymphocytic thyroiditis or autoimmune thyroiditis, was first reported in 19121. HT is characterized by lymphocyte infiltration and damage to thyroid follicular tissue. Laboratory tests are mainly manifested as increasing thyroid-specific antibodies, including anti-thyroglobulin antibody (TgAb) and anti-thyroid peroxidase antibody (TPOAb)2. The incidence of HT is in the range of 0.4%-1.5%, accounting for 20%-25% of all thyroid diseases, and this value has increased in recent years3. In addition, a large number of studies have reported that HT is associated with the oncogenesis and recurrence of papillary thyroid carcinoma (PTC)4,5; the potential mechanisms are still controversial. Autoimmune thyroiditis is also an important factor in female infertility6. Therefore, the pathogenesis of HT needs to be clear, for which a stable and simple animal model is essential.
To study the etiology of HT, two main kinds of murine models have been employed, including experimental autoimmune thyroiditis (EAT) and spontaneous autoimmune thyroiditis (SAT) in the present studies7,8. Susceptible mice were immunized with specific thyroid antigens (including the crude thyroid, purified thyroglobulin [TG], thyroid peroxidase [TPO], recombinant TPO ectodomain, and selected TPO peptides) to establish the EAT murine model. In addition, the adjuvants, including lipopolysaccharide (LPS), complete Freund's adjuvant (CFA), and other unusual adjuvants, are also used during the immunization to break down immune tolerance9,10,11,12,13,14,15,16,17.
The SAT model is an important model to study the spontaneous development of autoimmune thyroiditis, which is based on NOD.H-2h4 mice. The NOD.H-2h4 mouse is a new strain obtained from the cross of NOD and B10.A(4R) mice, followed by multiple backcrosses to NOD, with the autoimmune thyroiditis susceptibility gene IAk18,19. NOD.H-2h4 mice do not develop diabetes, but have a high incidence of autoimmune thyroiditis and Sjogren's syndrome (SS)19. Studies have found that intracellular adhesion molecule-1 (ICAM-1) is highly expressed in the thyroid tissue of NOD.H-2h4 mice at 3-4 weeks of age. Moreover, with the increase in iodine intake, the immunogenicity of the thyroglobulin molecule is enhanced, which further upregulates the expression of ICAM-1, which plays an important role in the process of monocyte infiltration21. This model simulates the autoimmune process while verifying the relationship between iodine dose and disease severity. The established method is stable, with a high probability of success. The SAT model has been applied to induce autoimmune thyroiditis for many years and continues to be an effective method to study the pathogenesis of autoimmune thyroiditis. However, the current construction method of the EAT model is more complicated and expensive; different laboratories use different immunization methods and injection sites. Furthermore, mice with different genetic backgrounds have different rates of induction, which need further study to reveal the potent mechanism.
However, the development of thyroiditis in the SAT model is associated with sodium iodide, sexual dimorphism, and the rearing conditions. To reveal the appropriate procedure of autoimmune thyroiditis in the SAT model, this article described the method of induction of autoimmune thyroiditis in different conditions. In addition, it allows for the study of the pathogenesis and immunological progress of autoimmune thyroiditis in different stages of this disease.
The protocol described below was approved by the care and use guidelines established by the Institutional Animal Care and Use Committee of Sichuan University.
1. Preparation
2. Induction of thyroiditis
3. Measurements
The histological changes were strikingly different in female and males, the duration of iodine intake, and the solution of NaI. As shown in Figure 1, ~10% of NOD.H-2h4 mice developed SAT even without iodine induction at the age of 24 weeks, and all the mice eventually developed thyroiditis. When given regular water, there was no significant difference in the histological changes between males and females. The addition of NaI to the drinking water accelerated the development of thyroiditis. In the solutions of 0.005%, 0.05%, and 0.5%, SAT reached maximal severity in weeks 16, 8, and 8, respectively, after giving the NaI water. Once thyroiditis was induced, lymphocyte infiltration continued throughout the mouse's life. During the induction, female mice seemed to have a trend in developing more severe thyroiditis than males under the same condition, but there were no significant differences in the severity of lymphocytic infiltration, which might be limited by the number of samples.
When given regular water without additional iodine, the level of TgAb did not rise significantly at the age of 16 weeks, and no significant difference was found between males and females. TgAb levels in NOD.H-2h4 mice began to rise at 24 weeks, with no difference between sexes, and TgAb levels continued to rise until 72 weeks (Figure 2A). With the addition of NaI to drinking water, the levels of TgAb began to rise at weeks 16, 8, and 8 with solutions of 0.005%, 0.05%, and 0.5% NaI, respectively. However, the levels of TgAb showed no difference between males and females in any of the solutions (Figure 2B–D). Regardless of whether NaI water was given, the levels of TgAb reached their highest at the age of 72 weeks (or 64 weeks after NaI water was first given).
There was a longer delay in the detectability of TPOAb levels compared to TgAb. These antibodies were rarely detected in NOD.H-2h4 mice with regular water at the age of 24 weeks. When fed with regular water, females exhibited much higher TPOAb levels than males at 72 weeks (Figure 2E). The same trend of intersex difference was also emerging in NOD.H-2h4 mice during the duration of 16, 8, and 8 weeks when given NaI water of 0.005%, 0.05%, and 0.5%, respectively (Figure 2F–H) Remarkably, however, when the concentration of NaI was over 0.05% during the induction, the levels of TPOAb reached their highest at 64 weeks after NaI water was first given, and there was no difference between males and females (Figure 2G,H).
In addition, serum TSH levels in male NOD.H-2h4 mice were significantly higher than those of females at different feeding stages, regardless of whether NAI water was given or not. When given regular water, the TSH levels of NOD.H-2h4 mice began to increase at the age of 24 weeks, and a similar tendency was observed in the iodine diet groups with the concentrations of 0.005%, 0.05%, and 0.5% at 16, 8, and 8 weeks, respectively, after iodine was given separately. TSH levels continued to rise throughout the induction period, peaking at 64 weeks. Furthermore, there was a significant difference in TSH levels between males and females in the rest of the NOD.H-2h4 mice (Figure 3A–D). NOD.H-2h4 mice did not experience goiter or hypothyroidism after iodine ingestion. Regardless of whether iodine agents were administered, there was no sex-related difference in T4 levels in NOD.H-2h4 mice. However, with the duration of induction increasing, the levels of T4 had a tendency to decrease in both sexes in the iodine water groups (no significant difference) (Figure 3E–H).
Sexual dimorphism in the pathological process and TPOAb levels were found in the progeny of NOD.H-2h4 mice, but not TgAb. Similar to other studies25,26, TPOAb levels were significantly increased in model mice after receiving dietary iodide for 2-4 months, and there was a sex-related difference. In addition, TSH levels were higher in male than female NOD.H-2h4 mice with or without dietary iodide supplementation, which is similar to other mouse strains. This difference remained throughout the lives of the animals, which may not have beeb influenced by the levels of TPOAb or TgAb.When exposed to 0.05% NaI, most of the mice developed thyroiditis in 8 weeks, which had a similar effect with 0.5% NaI. While 0.005% NaI had a similar effect on NOD.H-2h4 mice fed with regular water, the concentration of 0.05% NaI (after 8 weeks of induction) might be the preferable condition to induce autoimmune thyroiditis. T4 levels in NOD.H-2h4 mice had no sex-related differences and were not affected by iodine agents, which is consistent with other studies27,28. This may be due to the relatively short onset time of the disease and the compensatory thyroid function of the mice, resulting in a phenomenon similar to subclinical hypothyroidism in the mice. We speculate that the TSH axis of the mice would eventually become decompensated and produce differences over time.
Figure 1: Pathological changes and lymphocyte infiltration inflammatory scores of thyroid glands in NOD.H-2h4 mice with/without the iodine diet. (A,C,E,G) Pathological changes of the thyroid gland under the iodine diet of 0%, 0.005%, 0.05%, and 0.5% by HE staining; 200x magnification, N = 10/group. (B,D,F,G) Thyroid inflammation was determined according to the lymphocyte infiltration area; grading of the degree of lymphocyte infiltration: 0: almost no lymphocyte infiltration; 1+: more than one-eighth of the gland is invaded; 2+: one-quarter of the gland is invaded; 3+: one-quarter to one-half of the gland is invaded; 4+: more than one-half of the gland is destroyed. N = 10/group. Significant differences: *p < 0.05; **p < 0.01. Abbreviations: F = female; M = male; NW = N weeks. Please click here to view a larger version of this figure.
Figure 2: Autoantibodies in NOD.H-2h4 mice on regular and iodide-containing water. (A–D) Autoantibodies to Tg in NOD.H-2h4 mice on regular water (at the age of 8, 16, 24, and 72 weeks separately) and iodide water-0.005%, 0.05%, and 0.5% (at 0, 8, 16, and 64 weeks after the iodine water was begun). (E–H) Autoantibodies to TPO in NOD.H-2h4 mice on regular water (at the age of 8, 16, 24, and 72 weeks separately) and iodide water-0.005%, 0.05%, and 0.5% (at 0, 8, 16, and 64 weeks after the iodine water was begun). Values for TgAb are reported as ELISA optical density (OD) 490 nm (mean ± standard error of the mean SEM) and values for TPOAb are reported as the geometric mean in flow cytometry. N = 8/group. Significant differences: *p < 0.05; **p < 0.01; ***p < 0.001. Abbreviations: Tg = thyroglobulin; TgAb = autoantibodies to Tg; TPO = thyroid peroxidase; TPOAb = autoantibodies to TPO; F = female; M = male; NW = N weeks. Please click here to view a larger version of this figure.
Figure 3: TSH and T4 levels in NOD.H-2h4 mice on regular and iodide-containing water. (A–D) TSH and (E–H) T4 levels on regular water (at the age of 8, 16, 24, and 72 weeks separately) and iodide water-0.005%, 0.05%, and 0.5% (at 0, 8, 16, and 64 weeks after the iodine water was begun). TSH values were measured by radioimmunoassay (mU/L, mean ± standard error of the mean SEM). N = 8/group. Significant differences: *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. Abbreviations: TSH = thyroid stimulating hormone; F = female; M = male; NW = N weeks. Please click here to view a larger version of this figure.
HT occurs due to an autoimmune system disorder caused by lymphocytes infiltrating the thyroid gland, further impairing thyroid function, while producing thyroid-specific antibodies. Serum TSH, TgAb, and TPOAb levels in HT patients are significantly elevated27. At present, two main kinds of murine models are widely used to study the etiology of autoimmune thyroiditis: EAT and SAT29. EAT mice are mostly immunized using proteins and adjuvants to create an abnormal immune environment in vivo. This approach has been used for many years and continues to be effective30,31. In addition, novel approaches to induce thyroiditis include injecting dendric cells (DCs) pulsed with Tg32, the expression of Tg or TPO in vivo by adenoviral vectors or plasmids31, transplanting the thyroid gland of allogeneic mice, and injecting LPS. However, the successful construction of the thyroiditis model is usually more restricted by the genetic background of the experimental murine model, syngeneic antigens, and cell biology technologies. Furthermore, specific adjuvants may induce a more complex immune environment, making it difficult to investigate the immunological process of autoimmune thyroiditis. Therefore, we believe that the SAT of NOD.H-2h4 mice can better illustrate the pathogenesis of Hashimoto’s thyroiditis compared with the EAT model. Specifically, immunization with potent adjuvants and thyroglobulin is not necessary for SAT33. To speed up the progress of SAT, iodine is added to the drinking water of NOD.H-2h4 mice. Although the resulting thyroiditis does not share the same mechanism by which thyroiditis occurs in humans, iodine is indeed an important influencing factor of thyroiditis. Furthermore, in this model, iodine agents can quickly induce thyroiditis in NOD.H-2h4 mice, and NOD.H-2h4 mice that do not receive iodine may eventually develop thyroiditis as well. As the experimental protocol in most laboratories is different, we performed a more comprehensive evaluation to the NOD.H-2h4 mice during the induction of thyroiditis.
In the process of model construction, some points need careful attention. Adding drinking water regularly and evaluating the general condition of the mice avoids accidental mouse deaths. It is recommended to keep no more than five mice per cage during the induction to provide enough NaI feeding to each mouse. However, the number of mice can also change following the specific institution’s rules and regulations for animal. As sexual dimorphism is found in our laboratory, single-sex mice are suggested for the experiment (for experiments with special requirements for the detection of thyroid-specific antibodies, female are recommended). It is suggested to select 8-week-old mice, as 10% of the mice develop spontaneous thyroiditis by the age of 16 weeks and all the mice develop thyroiditis eventually without NaI water. The thyroid lesions persist throughout the lifetimes of the mice and the next generation develop thyroiditis when born. Therefore, for the preservation of mice, it is recommended to use 8-week-old mice.It is important to note that TPOAb is more suitable as a detection indicator than TgAb when using NOD.H-2h4 mice to mimic thyroiditis. However, when measuring the levels of TPOAb, the duration of induction needs to be more delayed than TgAb.
In conclusion, this paper describes the establishment of a SAT model with NOD.H-2h4 mice and the exploration of the influence of different factors on this model. Although it does not perfectly simulate the pathogenesis and mechanism of autoimmune thyroid disease, the NOD.H-2h4 mouse is a stable animal model with great potential in the field of autoimmune thyroid disease. Given that this is an easy animal model to construct and is highly reproducible, it is hoped that this method will help improve applications of SAT murine models in different research institutions.
The authors have nothing to disclose.
Mouse monoclonal antibodies to human TPO (used as positive controls) were provided by Dr. P. Carayon and Dr. J. Ruf (Marseille, France). The authors thank all the participants in this study and the members of our research team. This work was in part supported by grants from the Postdoctoral Sustentation Fund of West China Hospital, Sichuan University, China (2020HXBH057) and the Sichuan Province Science and Technology Support Program (Project No. 2021YFS0166)
Butorphanol tartrate | Supelco | L-044 | |
Dexmedetomidine hydrochloride | Sigma-Aldrich | 145108-58-3 | |
Enzyme-linked immunosorbent assay (ELISA) well | Sigma-Aldrich | M9410-1CS | |
Ethanol | macklin | 64-17-5 | |
Freund’s Adjuvant, Complete | Sigma-Aldrich | F5881 | |
Freund’s Adjuvant, Incomplete | Sigma-Aldrich | F5506 | |
Goat anti-Mouse IgG | invitrogen | SA5-10275 | |
Midazolam solution | Supelco | M-908 | |
Mouse/rat thyroxine (T4) ELISA | Calbiotech | DKO045 | |
Paraformaldehyde | macklin | 30525-89-4 | |
Propidium iodide | Sigma-Aldrich | P4864 | |
Sodium Iodine | Sigma-Aldrich | 7681-82-5 | |
Thyroglobulin | Sigma-Aldrich | T1126 | |
Thyroglobulin ELISA Kit | Thermo Scientific | EHTGX5 | |
TSH ELISA | Calbiotech | DKO200 | |
Xylene | macklin | 1330-20-7 |