Cytotoxic effector molecule gene expression in acute renal allograft rejection: correlation with clinical outcome; histopathology and function of the allograft.

BACKGROUND: Cytotoxic effector molecule expression in human renal allograft biopsies has been closely associated with acute rejection. Here we studied whether intragraft expression of perforin, granzyme B, and Fas ligand correlates with long-term clinical outcome of acute rejection episodes. Furthermore, we examined the relation to histopathology and function of the allograft during rejection. METHODS: Twenty-two human renal biopsies were quantified for mRNA expression of perforin, granzyme B, Fas ligand, and Fas with reverse transcription-polymerase chain reaction. Expression levels were correlated with clinical outcome after 12 months, Banff rejection grades, and allograft function in the course of acute rejection. RESULTS: Only Fas ligand, but not perforin or granzyme B, showed significantly higher up-regulation in seven samples with therapy-resistant acute rejections versus eight samples with therapy-sensitive acute rejection. We found no relation between cytotoxic marker expression and Banff rejection grades or serum creatinine peak levels. CONCLUSIONS: Fas ligand may be useful as an early marker of therapy-resistant acute rejection. Cells that express Fas ligand but not classical soluble cytotoxic molecules might influence clinical outcome of acute rejection episodes.

Differential regulation of granulocyte-macrophage colony-stimulating factor mRNA and protein expression in human thyrocytes and thyroid-derived fibroblasts by interleukin-1 alpha and tumor necrosis factor-alpha.

In this study, we provide the first report on the production of granulocyte-macrophage colony-stimulating factor (GM-CSF) by human thyroid epithelial cells. Primary cultures of highly purified thyrocytes and thyroid-derived fibroblasts (n = 3) and three thyroid anaplastic and one largely papillary carcinoma cell lines were exposed to different potent GM-CSF stimulators, employing interleukin 1 alpha (Il-1 alpha) and tumour necrosis factor-alpha (TNF-alpha). Cytokine mRNA levels were monitored by semi-quantitative reverse transcriptase-PCR including an internal heterologous competitor fragment after 3, 6 and 18 h of culture. Culture supernatants were assayed for GM-CSF using a highly sensitive ELISA (detection limit < or = 0.5 pg/ml) after 24 h. Basal GM-CSF mRNA expression was higher in fibroblasts and SW 1736 cells compared with thyrocytes, C 634, 8505 C and HTh 74 cells. GM-CSF was spontaneously secreted by fibroblasts (means +/- S.E.M.; 43 +/- 15 pg/ml), SW 1736 (59 +/- 4 pg/ml), HTh 74 (34 +/- 4 pg/ml) and C 643 cells (12 +/- 1 pg/ml) but not by thyrocytes and 8505 C cells. Treatment with Il-1 alpha (10 U/ml) resulted in a marked increase of GM-CSF mRNA within 3 h and an increase or induction of protein expression in thyrocyte (2350 +/- 214 pg/ml), fibroblast (5242 +/- 1400 pg/ml), SW 1736 (20016 +/- 280 pg/ml) and C 643 cultures (1285 +/- 79 pg/ml). Stimulation with TNF-alpha (10 U/ml) yielded divergent results. No significant increase of GM-CSF mRNA or protein expression was found in thyrocytes although TNF-alpha receptor expression in these cells is well documented. Stimulation with TNF-alpha resulted in an increased GM-CSF production in fibroblasts (361 +/- 14 pg/ml), HTh 74 (148 +/- 51 pg/ml) and SW 1736 cultures (235 +/- 43 pg/ml). TSH (10 mU/ ml) did not stimulate GM-CSF secretion in thyrocytes and HTh 74 cells, both expressing the TSH receptor. Phorbol 12-myristate 13-acetate (10 ng/ml) enhanced GM-CSF mRNA and protein levels in all cell types investigated. Our data suggest that both thyrocytes and fibroblasts synthesize GM-CSF in response to Il-1 alpha, but only fibroblasts respond to TNF-alpha with a significant increase in GM-CSF. Anaplastic thyroid carcinomas are potential GM-CSF producers.

Peripheral T cell activation in long-term renal transplant patients: concordant upregulation of adhesion molecules and cytokine gene transcription.

In renal transplant, patients, the number of T cells expressing high levels of LFA-1 (LFA-1-bright) and of T cells expressing CD57 increases in response to viral infection, even if the latter is asymptomatic. Their role in long-term renal transplant patients with cytomegalovirus (CMV) antigenemia and concomitant transplant dysfunction was investigated. For this purpose, this study used triple-color flow cytometry, fluorescence-activated cell sorting of peripheral blood T cells (CD3+/LFA-1-dim or -bright and CD8+/CD57+ or CD57- subsets), and subsequent semiquantitative reverse transcription-polymerase chain reaction. Cytokine mRNA levels for interleukin (IL)-1 beta, IL-2, IL-4, IL-8, IL-10, tumor necrosis factor alpha, and interferon-gamma, as well as Granzyme A and IL-2R p55 and p75 transcripts were determined and compared in peripheral blood mononuclear cells and in separated T cell subsets. Although in patients with CMV infection and/or rejection, cytokine transcripts were readily detected and the levels in the CD3+/LFA-1-bright subsets were, by orders of magnitudes, higher than in the LFA-1-dim subset, hardly any cytokine message was found in patients without CMV infection or rejection episodes or in control subjects. The expression of Granzyme A, which is involved in cytotoxic T lymphocyte-mediated cytotoxicity, was not upregulated in LFA-1-bright T cells, which is in discordance with cytokine levels. Differences between CD57+ and CD57- T cells were limited to the IL-2R p55 mRNA, of which the former expressed significantly less than the latter. It is concluded that upon virus-induced activation of peripheral blood T cells, an effector type that is marked by high inflammatory but small cytotoxic potential is produced. The results of this study propose that these cells represent a correlate of persistent immune activation and are liable to produce graft dysfunction, although they are unable to clear the organism from virus infection because of their lack of cytotoxic potential.

The enigma of CD57+CD28- T cell expansion--anergy or activation?

Expansion of a CD57+CD8+ T lymphocyte subset has been reported in HIV and human cytomegalovirus (HCMV) infection. Almost all of these T cells lack CD28 expression. While CD28- cells are often associated with anergy, some authors believe their expansion in HIV infection precipitates immunodeficiency. We studied 15 randomly chosen patients with immune activation and observed that CD57+CD28- T cell expansion may occur in various conditions and to the same degree as in HIV infection without resulting in immunodeficiency. Triple colour flow cytometry also revealed that the CD57 and CD28 antigens are coexpressed in only 3% of CD8+ T cells, irrespective of the underlying condition, so that almost all CD57+CD8+ cells are always CD28-. Analysis of Fas (CD95) expression with respect to CD28 expression on CD4+ and CD8+ T cells from 10 additional patients indicated no increased commitment to apoptosis in CD28- T cells. Semiquantitative polymerase chain reaction (PCR) comparing CD28+ and CD28-CD8+ T cells with respect to cytokine gene expression (tumour necrosis factor-alpha (TNF-alpha), interferon-gamma (IFN-gamma), IL-1beta) in five renal transplant patients with expansion of the CD57+ subset detected no cytokine gene expression deficit in CD28- T cells. A direct association of increased proportions of CD57+CD28-CD8+ T cells with immunodeficiency/anergy is disputed.

Delayed-type hypersensitivity-like mechanisms dominate late acute rejection episodes in renal allograft recipients.

Episodes of acute rejection (aRx) may occur in functional renal allografts even at a very late stage post-Tx. Histopathology in early and late aRx looks quite similar--however, there is a slower deterioration of graft function in late aRx, suggesting that pathogenetic immune mechanisms are different. In order to investigate this phenomenon we studied the gene expression pattern (IL-1beta, 2, 4, 8, 10, IFNgamma, TNFalpha, GrnA, IL-2R p55/p75) in PBMC and core biopsies from long-term renal allograft recipients with histologically proven late aRx and compared it with transplant and non-transplant controls using a semiquantitative RT-PCR technique. PBMC and graft-infiltrating cells of patients with late aRx showed an upregulation, especially of IFN-gamma, IL-4, IL-10, and TNFalpha transcripts. While IL-2 mRNA was only detected in PBMC of two patients with late aRx who were not on cyclosporine, upregulation of intragraft IL-2 mRNA allowed the best discrimination between aRx and the other groups (sensitivity: 83%, specificity: 93%). In contrast to several reports on early Rx we did not notice an elevation of granzyme A transcripts in comparison with the controls, suggesting that cell-mediated inflammatory processes (CD4+ T cell-mediated DTH) dominate the late aRx, while early aRx is characterized by the additional involvement of cytotoxic T cell response. This may explain the distinct clinical course. Additionally, in a pilot study we successfully treated late aRx in 10/12 patients with the anti-CD 4 mAb, 16H5. Our encouraging therapeutic results underline the pathogenetic role of CD4+ T cells and support our hypothesis on DTH-like mechanisms in late aRx.

Different cytokine mRNA profiles in Graves' disease, Hashimoto's thyroiditis, and nonautoimmune thyroid disorders determined by quantitative reverse transcriptase polymerase chain reaction (RT-PCR).

Intrathyroidal lymphocytes are a source of cytokines thought to stimulate or maintain the immune process within the thyroid in Graves' disease (GD) and Hashimoto's thyroiditis (HT). Quantitative assessment of the cytokine profile may provide important clues as to the Th1/Th2 balance prevailing in these diseases. We analyzed cytokine mRNA expression levels in thyroid tissue samples from 13 patients with GD, 2 with HT, 5 with nontoxic multinodular goiter (NTG), and 4 with thyroid autonomy (nodular = TAnod and perinodular = TAperi tissue) using multispecific competitor fragments with primer sequences for IL-1 beta, IL-2, IL-4, IL-6, IL-8, IL-10, IFN-gamma, CD25, and CD3 delta-chain mRNA. Patients with GD were subdivided into two groups according to their serum levels of antibodies to thyroperoxidase (anti-TPO; GDhigh > 4000 U/mL, GDlow < or = 200 U/mL). These levels correlated positively with the CD3 delta-chain mRNA levels (r = 0.83) and with the T cell infiltration (r = 0.71) as determined by immunohistochemistry. Patients with GDhigh demonstrated 2- to 4-fold higher IL-4 mRNA levels (as compared to all other investigated groups) and significantly higher IL-10 mRNA levels as compared to HT, GDlow, and TAnod patients. Patients with GDhigh also had significantly higher levels of IFN-gamma, IL-1 beta, IL-8, and CD25 mRNA as compared to GDlow. The highest IFN-gamma, IL-2, and CD25 mRNA levels were found in HT. The lowest mRNA levels of all the investigated groups were detected in TAnod. No significant differences in IL-6 and IL-8 mRNA levels were found between most of the patient groups. In summary, patients with GDhigh showed a shift to a more Th2-driven cytokine pattern. In contrast, the increase mRNA levels of Th1-related cytokines found in HT indicate predominantly T cell-mediated cytotoxic processes.

Late-acute renal allograft rejection and symptomless cytomegalovirus infection.

The role of cytomegalovirus infection in allograft injury is controversial. A subgroup of renal graft recipients who had histologically proven late-acute rejection did not respond to conventional anti-rejection therapy (80% graft loss within 1 year). These patients showed an expansion of memory-type CD8 peripheral-blood T cells that expressed interferon-gamma mRNA and an association with clinically symptomless cytomegalovirus infection (82% PCR positive, 42% antigenaemia). Antiviral therapy with ganciclovir resulted in stable improved graft function in 17 of 21 treated patients with cytomegalovirus-associated late-acute rejection. The results underline the clinical relevance of cytomegalovirus-related graft injury and offer a novel therapeutic approach.