CD4(+) bias in T cells cloned from a CML patient with active graft versus leukemia effect.

BACKGROUND: The ability to generate a GvL response by infusion of donor leukocytes (DL) in patients following relapse after BMT is now well documented and has been demonstrated to be particularly effective in patients with CML. METHODS: We generated T-cell lines from a patient who was undergoing an active GvL response following withdrawal of immunosuppression for cytogenetic relapse of CML. Cryopreserved pre-transplant leukemic cells were used as stimulators, to generate T-cell lines and oligoclonal lines from the lymphocytes. In total 38 sub-lines were generated from different bulk cultures. The lines were tested for their proliferative and cytotoxic capability to patient pre-transplant leukemic cells, PHA-transformed lymphoblasts, allogeneic CML cells, and autologous and allogeneic B-LCL. RESULTS: Four of the cloned lines tested recognized the patient's pre-transplant leukemic cells. Specifically, two were both cytotoxic and proliferative in response to patient leukemic cells and two were cytotoxic only. Six clonal lines recognized PHA blasts only and were proliferative; one was specific for PHA blasts and CML cells. The sub-lines were phenotyped for cell-surface markers and all were CD4(+) CD8(-) CD 16/56(-). The proliferative response of the leukemia-specific clonal lines could be blocked with anti-MHC Class II MAbs. DISCUSSION: These data suggest that CD4(+) cells play a crucial role in mediating the GvL effect in CML patients. Our observations can be used to delineate strategies for enhancing and investigating the GvL effect in CML.

A one-tube polymerase chain reaction protocol demonstrates CC chemokine overexpression in Graves' disease glands.

An adaptation of mixed oligonucleotide primed amplification of complementary DNA to detect the profile of CC chemokines in biological samples is presented. By introducing normalization, two correction coefficients, performing a single amplification reaction, and five parallel hybridizations, intrasample and intersample comparisons can be reliably made. This protocol of single tube PCR CC chemokine profiling was applied to tissue samples from an autoimmune thyroid condition, Graves' disease, and from a nonautoimmune condition, multinodular goiter. Results demonstrate overexpression of CC chemokines in Graves' disease, statistically significant for macrophage inflammatory protein-1alpha and -1beta, which correlated with the aberrant human leukocyte antigen class II expression by thyrocytes, as assessed by flow cytometry. Overexpression of CC chemokines probably plays a major role in determining the characteristics of the lymphocytes migrating to the thyroid gland and influences the course of the disease. The study of chemokine profile should be more informative than the study of isolated chemokines and cytokines, and as it can be applied to fine needle aspiration biopsies, it may be useful to clinical research.

HLA-DM and invariant chain are expressed by thyroid follicular cells, enabling the expression of compact DR molecules.

Thyroid follicular cells (TFC) in Graves' disease (GD) hyperexpress HLA class I and express ectopic HLA class II molecules, probably as a consequence of cytokines produced by infiltrating T cells. This finding led us to postulate that TFC could act as antigen-presenting cells, and in this way be responsible for the induction and/or maintenance of the in situ autoimmune T cell response. Invariant chain (li) and HLA-DM molecules are implicated in the antigen processing and presentation by HLA class II molecules. We have investigated the expression of these molecules by TFC from GD glands. The results demonstrate that class II+ TFC from GD patients also express li and HLA-DM, and this expression is increased after IFN-gamma stimulation. The level of HLA-DM expression by TFC was low but sufficient to catalyze peptide loading into the HLA class II molecules and form stable HLA class II-peptide complexes expressed at the surface of TFC. These results have implications for the understanding of the possible role of HLA class II+ TFC in thyroid autoimmune disease.

Single-cell analysis of intrathyroidal lymphocytes shows differential cytokine expression in Hashimoto's and Graves' disease.

Most human organ-specific autoimmune diseases such as Hashimoto's thyroiditis (HT) are considered to be Th1 mediated, and a quantitative dominance of Th1 cells in thyroid infiltrates from both Graves' disease (GD) and HT affected glands has been reported. However, Th2 dominance would be expected in GD, where thyroid hyperfunction induced by stimulating antibodies predominates over tissue destruction. We have analyzed the interleukin-4 (IL-4), interferon-gamma (IFN-gamma) production by T cells at the single-cell level, both in infiltrating lymphocytes isolated from digested GD and HT thyroid glands and in derived T cell lines, by direct intracellular cytokine detection. Results showed a heterogeneous pattern of cytokine production in bulk GD infiltrates and derived T cell lines, and a similar pattern was observed in the much larger HT infiltrates. Both type 1 and type 2 cytokines were simultaneously produced by the infiltrating populations, and T cells with both patterns as well as intermediate patterns similar to Th0 cells could be detected ex vivo. However, the larger T lymphocytes, presumably activated and responsible for the autoimmune damage, predominantly produced IL-4 in GD and IFN-gamma in HT. The specificity of the Th2 responses in GD was suggested by the enrichment in IL-4 production after antigen-specific expansion of two oligoclonal T cell lines. These data show that both type 1 and type 2 cytokines are produced in the thyroid glands affected by autoimmunity and that the difference between diseases may be the effect of a functionally dominant population at a given time. This in vivo chronically activated antigen-specific population, producing type 1 or type 2 cytokines locally, may be responsible for the effect finally leading to one of the disease states.

Self-reactive cytotoxic gamma delta T lymphocytes in Graves' disease specifically recognize thyroid epithelial cells.

In this paper we report the isolation of a self-reactive cytotoxic gamma delta T cell line, 158RE.2, that originates from the T lymphocyte population infiltrating the thyroid gland of a patient with Graves' disease. Functional data using this cell line demonstrate that gamma delta T cells expanded in the thyroid tissue specifically recognize a ligand expressed by thyroid epithelial cells and cell lines of endocrine epithelial origin. The TCR expressed by these gamma delta T cells--V gamma I/V delta 5--is unusual in peripheral blood lymphocytes, and its specificity is clearly different from that observed in a high percentage of gamma delta T cells from PBL, which express the common TCR V gamma 9/V delta 2. The V gamma I/V delta 5 receptor is involved in the recognition of the ligand expressed by the thyroid cells, but not in the NK-like activity also displayed by 158RE.2. These cells express CD8 alpha alpha dimers, which participate in the thyroid ligand recognition but not in the NK-like activity. The epithelial cell recognition is not restricted by classical MHC class I or class II molecules, although the CD8 alpha alpha participation in the recognition suggests the involvement of nonclassical MHC molecules. These are the first data to be presented on self-reacting gamma delta T cells in human epithelium.

Hyperinducibility of HLA class II expression of thyroid follicular cells from Graves' disease. A primary defect?

Thyroid follicular cells (thyrocytes) from Graves' disease (GD) patients' thyroid glands express HLA class II molecules "ectopically." This phenomenon has been attributed to induction by locally produced cytokines and may be relevant to disease pathogenesis. We have compared IFN-gamma-mediated induction of HLA class II in thyrocytes from glands affected with GD and a nonautoimmune disease (MNG), to investigate a possible differential regulation of HLA expression between these two pathologies. HLA induction has been measured in primary thyrocyte cultures and control autologous macrophages stimulated or not stimulated with IFN-gamma. Comparison of flow cytometric data using an improved algorithm demonstrated that expression of HLA class II molecules is more readily induced in thyrocytes from GD than from MNG thyroid glands. This higher inducibility was parallel to a faster and stronger induction of HLA class II message in GD thyrocytes but did not correlate with the levels of HLA class II or class I originally expressed by thyrocytes in the tissue or with the degree of lymphocytic infiltration of the gland. There was no association with a particular HLA class II allele or with the presence of IFN-gamma and IL-2 in the tissue, as assessed by reverse transcription-PCR. No differences in the induction of class II were found in macrophages from each group of patients. These results suggest that an intrinsic feature of thyrocytes from GD patients is an up-regulation of HLA class II expression and that this is a characteristic that may facilitate the triggering of autoimmunity to "hyperinducible" thyroid glands.

Pancreas in recent onset insulin-dependent diabetes mellitus. Changes in HLA, adhesion molecules and autoantigens, restricted T cell receptor V beta usage, and cytokine profile.

Insulin-dependent diabetes mellitus (IDDM), in which only the pancreatic beta cells are destroyed by the autoimmune response, is the paradigm of organ-specific autoimmunity. As a result of a combination of factors, the number of immunohistologic/cellular/molecular studies of pancreas in IDDM is very limited. We report here studies conducted in the pancreata of two IDDM patients: one newly diagnosed (case 1) and one long standing (case 2). In case 1, we demonstrated the presence of morphologically normal viable beta cells without evidence of viral infection. In both cases the expression of the autoantigens defined by islet cell Abs and by glutamic acid decarboxylase was markedly reduced in the islet cells whereas expression of hsp60, another putative autoantigen, was normal. Over-expression of HLA class I was detected in 58% of the islets in pancreatic sections and in cultured beta cells in case 1 and also in 30% of islets in case 2 but it was not restricted to any insular cell type. In case 1, there was "inappropriate" HLA class II expression in islets cells but it was a rare finding and not beta cell specific. The analysis of the correlation between class I overexpression, residual insulin, and insulitis suggests that the first event is the increase of HLA class I expression. Of adhesion molecules, ICAM-1, VLA, VCAM, and LFA-3 were normal and only ICAM-1 was moderately overexpressed in and around the islets of case 1 insulitis, as was detected by immunofluorescence which showed that 18% of the islets of case 1 had CD8+ lymphocytes as the predominant population. Reverse transcription-PCR demonstrated moderate V beta skewing and the profile of cytokines expected in CTLs: IL-2, IL-4, IL-10, and IFN-gamma negative, perforin positive. In addition, IFN-alpha, IFN-beta, and IL-6 transcripts were detected in the case 1 pancreas, consistent with the existence of a silent viral infection. Overall, the results indicated that, differently from spontaneous animal models of diabetes, in the pancreas of IDDM patients there are no elements of the inductive phase of the autoimmune response.