The mechanisms by which glucocorticoid receptor (GR) mediates glucocorticoid (GC)-induced apoptosis are unknown. We studied the role of mitochondrial GR in this process. Dexamethasone induces GR translocation to the mitochondria in GC-sensitive, but not in GC-resistant, T cell lines. In contrast, nuclear GR translocation occurs in all cell types. Thymic epithelial cells, which cause apoptosis of the PD1.6 T cell line in a GR-dependent manner, induce GR translocation to the mitochondria, but not to the nucleus, suggesting a role for mitochondrial GR in eliciting apoptosis. This hypothesis is corroborated by the finding that a GR variant exclusively expressed in the mitochondria elicits apoptosis of several cancer cell lines. A putative mitochondrial localization signal was defined to amino acids 558–580 of human GR, which lies within the NH2-terminal part of the ligand-binding domain. Altogether, our data show that mitochondrial and nuclear translocations of GR are differentially regulated, and that mitochondrial GR translocation correlates with susceptibility to GC-induced apoptosis.

No abstract

Glucocorticoids (GCs) are integral components in the treatment protocols of acute lymphoblastic leukemia, multiple myeloma, and non-Hodgkin lymphoma owing to their ability to induce apoptosis of these malignant cells. Resistance to GC therapy is associated with poor prognosis. Although they have been used in clinics for decades, the signal transduction pathways involved in GC-induced apoptosis have only partly been resolved. Accumulating evidence shows that this cell death process is mediated by a communication between nuclear GR affecting gene transcription of pro-apoptotic genes such as Bim, mitochondrial GR affecting the physiology of the mitochondria, and the protein kinase glycogen synthase kinase-3 (GSK3), which interacts with Bim following exposure to GCs. Prevention of Bim up-regulation, mitochondrial GR translocation, and/or GSK3 activation are common causes leading to GC therapy failure. Various protein kinases positively regulating the pro-survival Src-PI3K-Akt-mTOR and Raf-Ras-MEK-ERK signal cascades have been shown to be activated in malignant leukemic cells and antagonize GC-induced apoptosis by inhibiting GSK3 activation and Bim expression. Targeting these protein kinases has proven effective in sensitizing GR-positive malignant lymphoid cells to GC-induced apoptosis. Thus, intervening with the pro-survival kinase network in GC-resistant cells should be a good means of improving GC therapy of hematopoietic malignancies.

Recent data cast new light on the mechanisms by which glucocorticoids (GCs) elicit apoptosis of thymocytes and leukemia cells. Here we attempt to integrate recent studies by others and us, which provide a novel insight to this apoptotic process. In the last few years it was made clear that there is a tight cooperation between genomic and non-genomic effects exerted by GC receptors (GRs). GC invokes major alterations in the gene expression profile through GR-mediated transactivation and transrepression, which ultimately tip the balance between pro-survival and pro-apoptotic proteins. Although essential in shaping the cell's proteome, these genomic effects are insufficient to elicit apoptotic death and additional signals are required for activating the pro-apoptotic proteins. Several non-genomic effects have been described that occur immediately following exposure to GC, which are imperative for the induction of apoptosis. We have recently observed that GC induces instant GR translocation to the mitochondria in GC-sensitive, but not in GC-resistant, T lymphoid cells. This response contrasts the nuclear translocation of GR occurring in both cell types. We propose that the sustained elevation of GR in the mitochondria following GC exposure is crucial for triggering apoptosis.

T cell development in the thymus is controlled by thymic epithelial cells (TE). While it is accepted that TE interact with maturing T cells, the mechanisms by which they trigger 'death by neglect' of double-positive (DP) thymocytes are poorly understood. We and others have demonstrated a role for TE-derived glucocorticoids (GCs) in this process. We have studied TE-induced apoptosis using an in vitro system based on co-culturing a thymic epithelial cell line (TEC) with DP thymic lymphoma cells or thymocytes (DP thymic cells). Here, we demonstrate that nitric oxide (NO*) is also involved in this death process. The inducible nitric oxide synthase (iNOS) inhibitors N(G)-methyl-L-arginine and 1,4-PBIT attenuated TEC-induced apoptosis of DP thymic cells. Co-cultivation of TEC with DP thymic cells increased the expression of iNOS in TEC. A concomitant increase in NO* was detected by staining with DAF-FM diacetate. Moreover, the iNOS-regulating cytokines IL-1alpha, IL-1beta and IFNgamma were up-regulated upon interaction of TEC with DP thymic cells. Neutralizing IL-1R or IFNgamma reduced TEC-induced apoptosis of DP thymic cells. Cardinally, NO* synergizes with GCs in eliciting apoptosis of DP thymic cells. Our data indicate that a cross-talk between DP thymic cells and TEC is required for proper induction of iNOS-up-regulating cytokines with a subsequent increase in iNOS expression and NO* production in TEC. NO*, in turn, cooperates with GCs in promoting death by neglect. We suggest that NO* together with GCs fine-tune the T cell selection process.