Understanding cellular autoimmune processes


Title: Regulatory T cells in human autoimmune thyroid disease.

Authors: Marazuela M, Garcia-Lopez MA, Figueroa-Vega N, de la Fuente H, Alvarado-Sanchez B, Monsivais-Urenda A, Sanchez-Madrid F, & Gonzalez-Amaro R.

Reference: Journal of Clinical Endocrinology and Metabolism 91: 3639-3646, 2006



T regulatory cells have been shown to play a key role in peripheral tolerance in animal models of several autoimmune diseases, including autoimmune thyroid disease (AITD). To date there have been few studies on such cells in human autoimmune disease, especially thyroid disease.


To carry out a functional and phenotypic analysis of CD4 + T regulatory cells in human AITD.


The authors studied thyroid and peripheral blood lymphocytes (PBL) from 12 patients with Graves- disease (GD) and 8 with Hashimoto-s thyroiditis (HT). A further 8 with GD and 4 with HT were analysed using PBL alone and results were compared with PBL from 13 controls. Cells were analysed by flow cytometry and immunohistochemistry for expression of CD4, CD25 and more specific markers of T regulatory cells, namely Foxp3 and glucocorticoid-induced TNF receptor (GITR). The function of T regulatory cells was assessed in cell proliferation assays. Since transforming growth factor-β (TGF-β) is synthesised by a second population of T regulatory cells which is CD69 + in animal models, the production of this cytokine was measured by RNase protection and quantitative PCR assays.


PBL from AITD patients had an increased percentage of CD4 + T cells expressing CD69, GITR and Foxp3 compared to controls, and there was also an increase in cells brightly expressing CD25. Thyroid-derived lymphocytes had an increased proportion of these cells compared to PBL from the same patient. This change was confirmed by the observation of a prominent infiltration of the thyroid by CD25 + , CD69 + and GITR + lymphocytes, with a modest increase in Foxp3 + cells. There was an increase in TGF-β mRNA in thyroid tissue and thyroid lymphocytes synthesized TGF-β on stimulation with a monoclonal antibody against CD69. However the suppressive function of T regulatory cells in the PBL of AITD patients was lower than that of healthy controls.


Although there are abundant cells with the phenotypic properties of T regulatory cells in the PBL and thyroid populations in AITD, they are patently unable to halt the autoimmune process, and there is also an unexplained defect in PBL T regulatory cell function that may be related to this inability.


The most significant development in the understanding of cellular autoimmune processes recently has been the resurrection and rehabilitation of the T suppressor cell (or T regulatory cell). Despite experiments carried out in the 1970s which showed a clear role for such cells in maintaining tolerance to autoantigens, the difficulty in delineating markers and the vagaries of bioassays for function led many in the 1980s to question the existence of such cells. This is well-illustrated in the field of AITD where elegant experimental studies in a rat model of HT in the 1970s showed that thyroiditis could be prevented by transfer of T cells from healthy animals, but attempts to demonstrate thyroid antigen-specific T suppressor cells in human AITD could not be reproduced and in any case used non-physiological systems.

Careful definition of the phenotype of the cells which prevented autoimmunity indicated that they expressed CD4 and high levels of CD25; later, the more specific markers of GITR and Foxp3 were defined. The importance of these cells in man is best illustrated in the IPEX syndrome in which there is a genetic mutation in Foxp3 leading to ineffective T regulatory cell production. This is an inevitably lethal disease if untreated and the babies die with lymphoproliferation and autoimmune disease, including AITD. Recently it has become clear that, as in so much else of immunology, the situation is far more complicated than initially uncovered. One population of regulatory T cells can be defined as being generated in the thymus and they express CD25, Foxp3 and GITR as well as CTLA-4. There is a second group of T regulatory cells (sometimes called Th3-like) that develop in the periphery rather than the thymus. They too may express these surface markers, with the exception of Foxp3, and can be characterised by the synthesis of TGF- β and IL-10. Further cell populations can also mediate regulatory function, including those expressing CD69, but how these relate to the Th3-like cells is currently not clear.

The article by Marazuela et al. shows that all three types of T regulatory cell are increased in the thyroid in AITD with no apparent distinction between GD and HT. However there was a defect in the suppressor function of these cells when tested in proliferation assays in vitro, a clear reminder that phenotype does not equal function. In some ways this paradox is not surprising; firstly, we know there must be some defect in regulatory T cells for disease to occur in the first place, and secondly, such findings have been reported in rheumatoid arthritis despite evidence of normal or increased numbers of T regulatory cells. The explanation is less easy. Perhaps auto-reactive T cells are simply very difficult to regulate, and despite the presence of large numbers of T regulatory cells they are just unable to cope. The authors found a high expression of GITR ligand in the thyroid and this could have favoured proliferation of T regulatory cells while at the same time inhibiting function. Given the increase in TGF- β (and IL-10) in the thyroid, it may be that Th3-like regulatory cells are more resistant than other populations to such inactivation, but overall the observations lead us to the implication that down-regulation of ongoing autoimmunity by T regulatory cells (or their cytokines) is unlikely to be therapeutically effective until we have a greater understanding of the mechanisms involved.

It should also be recalled that in all the patients in this study, the AITD was longstanding and in the case of GD, the patients had relapsed after antithyroid drugs – they were therefore a rather selected population that had been exposed to drugs with immunomodulatory effects. It would be fascinating to know more about the earliest phase of disease and how this evolves; studies on T regulatory cells in postpartum thyroiditis would be useful in this regard. Also it is critical to keep in mind the relatively non-specific nature of the kind of studies that can currently be performed. In looking at phenotyped populations or functional assays of whole lymphocyte population proliferation, we cannot assess how thyroid-antigen specific T regulatory cells are expressing marker molecules or responding to autoantigen. In turn, this probably explains why clearer cut answers on the role of T regulatory cells have come from animal models.

( Summary and commentary prepared by Anthony Weetman )

Present summary and commentary are related to Chapter N- 7 of TDM

Full paper obtainable at http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=pubmed&cmd=Retrieve&dopt=AbstractPlus&list_uids=16804051&query_hl=3&itool=pubmed_docsum