Thyroid xenografts from patients with Graves' disease in severe combined immunodeficient mice and NIH-beige-nude-xid mice

Objective: To compare human thyroid xenografts from patients with Graves' disease in severe combined immunodeficient (SCID) mice and triple immunodeficient NIH-beige-nude-xid (NIH-3) mice to obtain an improved animal model for studying these xenografts. Design: Animal study. Participants and animals: Patients with Graves' disease; SCID and NIH-3 mice. Interventions: Thyroid tissue from six patients with Graves' disease was xenografted to SCID and NIH-3 mice; in addition, peripheral blood mononuclear cells (PBMC) from 12 patients with Graves' disease were grafted intraperitoneally to separate SCID and NIH-3 mice. Outcome measures: Levels of human immunoglobulin (IgG), thyroperoxidase antibodies (TPO-Ab), thyroglobulin (Tg-Ab), and expression of thyrocyte intercellular adhesion molecule-1 (ICAM-1) and histocompatibility leukocyte antigen (HLA-DR) in mice after xenografting. Results: IgG was detected in all mice grafted with Graves' thyroid tissue and some mice grafted with PBMC; levels of human IgG peaked 6 to 10 weeks after xenografting. Human IgG levels reached a mean of 500 mg/L (standard error of the mean [SEM] 150 mg/L) in the NIH-3 mice with thyroid xenografts. This was similar to results in SCID mice with thyroid xenografts, which had a mean level of human IgG of 640 mg/L (SEM 230 mg/L). PBMC xenografting resulted in a mean IgG level of 1200 mg/L (SEM 250 mg/L) in NIH-3 mice, which was similar to the mean level of 1000 mg/L (SEM 280 mg/L) in SCID mice. The rate of rise in human IgG in the sera of the NIH-3 mice with thyroid xenografts was similar to that in the SCID mice. TPO-Ab were also detected in some mice with Graves' thyroid grafts and in a few mice injected with PBMC, with levels peaking 4 to 6 weeks after xenografting. TPO-Ab levels reached a mean 109.3 U/mL (SEM 57.2 U/mL) in the NIH mice with thyroid xenografts, which were similar to the mean level of 91.7 U/mL (SEM 34.2 U/mL) in the SCID mice. There were no significant differences in the Tg-Ab levels in each type of mice (13.9 [SEM 12.1] U/mL v. 17.9 [SEM 7.9] U/mL). Eight weeks after xenografting into mice, the expression of xenograft thyrocyte ICAM-1 decreased significantly in both the SCID and NIH-3 mice (from 43.4%, SEM 4.9%, to 35.9%, SEM 4.6%, in the NIH-3 mice, p < 0.05, and from 43.4%, SEM 4.9%, to 32.5%, SEM 5.2%, in the the SCID mice, p < 0.05). However, the expression of thyrocyte HLA-DR did not change significantly in the NIH-3 mice (from 11.5%, SEM 3.3%, to 10.8%, SEM 3.3%), whereas it decreased significantly in the SCID mice (from 11.5%, SEM 3.3%, to 4.2%, SEM 2.0%, p < 0.02). Conclusions: Not only SCID mice but also NIH-3 mice may be useful as animal models for xenografted thyroid tissue, which will help us elucidate the pathogenesis of autoimmune thyroid disease. NIH-3 mice are superior to SCID mice in maintaining the expression of thyrocyte HLA-DR in Graves' thyroid xenografts at levels as high as those before xenografting; this maintenance of expression may be due to the lack of natural killer cells in NIH-3 mice.