Vorinostat and Mithramycin A in combination therapy as an interesting strategy for the treatment of Sézary T lymphoma: a transcriptomic approach.

SAHA (vorinostat) is a histone deacetylase inhibitor approved by the USA Food and Drug Administration (FDA) for treating advanced refractory cutaneous T cell lymphomas. As SAHA alters the expression of many genes under control of the Sp1 transcription factor, we examined the effect of its association with the FDA-approved anticancer antibiotic Mithramycin A (MTR, plicamycin), a competitive inhibitor of Sp1 binding to DNA. Sézary syndrome (SS) cells, expanded ex vivo from peripheral blood mononuclear cells of 4 patients, were tested for their sensitivity to the drugs regarding cytotoxicity and differential responsive gene expression. Multivariate statistical methods were used to identify genes whose expression is altered by SAHA, MTR, and the synergist effect of the two drugs. MTR, like SAHA, induced the apoptosis of SS cells, while the two drugs in combination showed clear synergy or potentiation. Expression data stressed a likely important role of additive or synergistic epigenetic modifications in the combined effect of the two drugs, while direct inhibition of Sp1-dependent transcription seemed to have only limited impact. Ontological analysis of modified gene expression suggested that the two drugs, either independently or synergistically, counteracted many intertwined pro-survival pathways deregulated in SS cells, resistance of these tumors to intrinsic and extrinsic apoptosis, abnormal adhesion migration, and invasive properties, as well as immunosuppressive behavior. Our findings provide preliminary clues on the individual and combined effects of SAHA and MTR in SS cells and highlight a potential therapeutic interest of this novel pair of drugs for treatment of SS patients.

Specific deficiency of p56lck expression in T lymphocytes from type 1 diabetic patients.

Peripheral T lymphocyte activation in response to TCR/CD3 stimulation is reduced in type 1 diabetic patients. To explore the basis of this deficiency, a comprehensive analysis of the signal transduction pathway downstream of the TCR/CD3 complex was performed for a cohort of patients (n = 38). The main result of the study shows that T cell hyporesponsiveness is positively correlated with a reduced amount of p56(lck) in resting T lymphocytes. Upon CD3-mediated activation, this defect leads to a hypophosphorylation of the CD3zeta-chain and few other polypeptides without affecting the recruitment of ZAP70. Other downstream effectors of the TCR/CD3 transduction machinery, such as phosphatidylinositol 3-kinase p85alpha, p59(fyn), linker for activation of T cells (LAT), and phospholipase C-gamma1, are not affected. In some patients, the severity of this phenotypic deficit could be linked to low levels of p56(lck) mRNA and resulted in the failure to efficiently induce the expression of the CD69 early activation marker. We propose that a primary deficiency in human type 1 diabetes is a defect in TCR/CD3-mediated T cell activation due to the abnormal expression of the p56(lck) tyrosine kinase.

IL-2 and long-term T cell activation induce physical and functional interaction between STAT5 and ETS transcription factors in human T cells.

Activation of Stat5 by many cytokines implies that it cannot alone insure the specificity of the regulation of its target genes. We have evidenced a physical and functional interaction between members of two unrelated transcription factor families, Ets-1, Ets-2 and Stat5, which could contribute to the proliferative response to interleukin 2. Competition with GAS- and EBS-specific oligonucleotides and immunoassays with a set of anti-Stat and anti-Ets families revealed that the IL-2-induced Stat5-Ets complex recognizes several GAS motifs identified as target sites for activated Stat5 dimers. Coimmunoprecipitation experiments evidenced that a Stat5/Ets-1/2 complex is formed in vivo in absence of DNA. GST-pull down experiments demonstrated that the C-terminal domain of Ets-1 is sufficient for this interaction in vitro. Cotransfection experiments in Kit225 T cells resulted in cooperative transcriptional activity between both transcription factors in response to a combination of IL-2, PMA and ionomycin. A Stat5-Ets protein complex was the major inducible DNA-binding complex bound to the human IL-2rE GASd/EBSd motif in long-term proliferating normal human T cells activated by CD2 and CD28. These results suggest that the inducible Stat5-Ets protein interaction plays a role in the regulation of gene expression in response to IL-2 in human T lymphocytes.

Physical interaction of the bHLH LYL1 protein and NF-kappaB1 p105.

The LYL1 gene was first identified upon the molecular characterization of the t(7;9)(q35;p13) translocation associated with some human T-cell acute leukemias (T-ALLs). In adult tissues, LYL1 expression is restricted to hematopoietic cells with the notable exclusion of the T cell lineage. LYL1 encodes a basic helix-loop-helix (bHLH) protein highly related to TAL-1, whose activation is also associated with a high proportion of human T-ALLs. A yeast two-hybrid system was used to identify proteins that specifically interact with LYL1 and might mediate its activities. We found that p105, the precursor of NF-kappaB1 p50, was the major LYL1-interacting protein in this system. The association between LYL1 and p105 was confirmed both in vitro and in vivo in mammalian cells. Biochemical studies indicated that the interaction was mediated by the bHLH motif of LYL1 and the ankyrin-like motifs of p105. Ectopic expression of LYL1 in a human T cell line caused a significant decrease in NF-kappaB-dependent transcription, associated with a reduced level of NF-kappaB1 proteins.

Secretion of extracellular factor(s) induced by X-irradiation activates the HIV type 1 long terminal repeat through its kappaB motif.

X-irradiation has been used in the treatment of several human diseases, including AIDS-related-malignancies. X-irradiation might induce the transcription and the replication of human immunodeficiency virus type 1 (HIV-1) and enhance nuclear factor kappa B (NF-kappaB). In the present article we show that the activation of the HIV-1 long terminal repeat (LTR) by direct X-irradiation can be mimicked by coculture of transfected cells with X-irradiated nontransfected (HIV-1-negative) cells. In the human colonic carcinoma cell line HT29, the activation seems to depend on an extracellular factor(s) released by a cell line treated with X-rays. The HIV-1 LTR cis-acting element conferring X-indirect responsiveness was identified as the kappaB tandem motif. The two main nuclear HIV-1 kappaB-binding complexes activated by X-direct and -indirect irradiation were the NF-kappaB p50/p65 and c-Rel/p65 heterodimers. Nuclear NF-kappaB activation was dependent on protein neosynthesis. It was partially inhibited by 100 microM pyrrolidine dithiocarbamate, a potent antioxidant drug, but was not correlated with a significant decrease in cellular IkappaBalpha. Furthermore, X-irradiation induces the expression of several cytokine genes generally associated with stress response and antibodies against interleukin 6 and TNF-alpha partially inhibited the X-indirect activation of the HIV-1 LTR. The use of protein kinase C (PKC)-specific inhibitor and of forskolin, an adenylate cyclase activator, suggests that a PKC-dependent pathway and the cAMP intracellular concentration could play a role in the X-indirect enhancement of HIV-1 LTR transcription in the HT29 cell line. In addition, supernatants of an X-irradiated HT29 cell culture activated the HIV-1 stimulation in infected peripheral blood monocytes.

Insulin stimulates haptotactic migration of human epidermal keratinocytes through activation of NF-kappa B transcription factor.

Insulin-mediated cell motility as well as the role of transcription factors in insulin-activated intracellular signal events have not been extensively studied. In this report we have examined whether insulin could mediate haptotactic migration of cultured human epidermal keratinocytes through activation of transcription factor NF-kappa B. Insulin caused a dose-dependent stimulation of keratinocyte migration that maximally reached 2-fold at 2 x 10(-7) M hormone. This phenomenon was independent of the nature of the extracellular matrix component (collagen I or laminin5/nicein) on which the cells migrated, indicating that a specific integrin-ligand complex is not required. A 10(-7) M insulin treatment of keratinocytes resulted in activation of a major kappa B DNA binding complex within 15 to 30 minutes, which was identified as the p65/p50 NF-kappa B heterodimer by electrophoretic mobility shift assays. The activation induced nuclear translocation of cytosolic pools of NF-kappa B factor. Pyrrolidine dithiocarbamate and N-acetyl-leucinyl-leucinyl-norleucinal H (two compounds that differentially inhibit I kappa B alpha degradation and, thus, NF-kappa B activation) reversed the insulin-stimulated keratinocyte haptotactic migration without affecting insulin receptor activation. These compounds inhibited the insulin-induced nuclear translocation of NF-kappa B as detected by confocal laser scanning microscopy. Taken together our experiments demonstrate that insulin stimulates haptotactic migration of human epidermal keratinocytes through activation of NF-kappa B transcription factor. They emphasize the ability of insulin to stimulate keratinocyte movement and provide a first clue to the mechanism of insulin-induced haptotactic signaling.

Temporal and subunit-specific modulations of the Rel/NF-kappaB transcription factors through CD28 costimulation.

Stimulation of highly purified primary T lymphocytes through CD2 and CD28 adhesion molecules induces a long-term proliferation, dependent on persistent autocrine secretion of interleukin 2 (IL-2), high and prolonged expression of inducible CD25/IL-2 receptor alpha chain (IL-2Ralpha), and secretion of growth factors such as the granulocyte-macrophage colony-stimulating factor (GM-CSF). CD28 costimulation appears to activate cytokine gene expression through conserved kappaB-related CD28 response (CD28RE) or cytokine 1 (CK-1) elements in addition to canonical NF-kappaB-binding sites. In this report, we assess: 1) the evolution of the expression, over an 8-day time period, of the Rel/NF-kappaB family of proteins in costimulated versus TcR/CD3-stimulated primary T cells; 2) the impact of changes on the in vitro occupancy of GM-CSF kappaB and CK-1, as well as IL-2Ralpha kappaB sites; and 3) the differential regulation of newly synthesized p65 and c-Rel by IkappaB proteins. We show that CD2+CD28 stimulation specifically induces, at maximal T cell proliferation phase, sustained nuclear overexpression of NFKB2 p52 and c-Rel subunits which might rely on long-lasting processing of p100 precursor for p52 and increased neosynthesis of c-Rel. This up-regulation correlates with sustained occupancy of GM-CSF kappaB and CK-1 elements by both proteins. Conversely, these subunits do not appear to bind to the IL-2Ralpha kappaB site. Costimulation, but not TcR/CD3 stimulation, appears supported by sustained down-regulation of both IkappaBalpha and -beta regulators. Furthermore, contrary to p65, c-Rel appears to display little affinity for p105, p100 and IkappaBalpha regulators.

Biogenesis of MHC class I antigens: involvement of multiple chaperone molecules.

To analyze the early events occurring during the folding and assembly of major histocompatibility complex class I antigens, we used a panel of P815 mouse mastocytoma transfectants expressing wild-type or mutant human leukocyte antigen (HLA)-Cw3 proteins. We observed that newly synthesized unassembled HLA-Cw3 heavy chains (Cw3 alpha) specifically associate with three major long-lived proteins denoted p105, p88 and p78, according to their size. These proteins display different kinetics of interaction. The association of p105 is transient, while p78, which we identified as the immunoglobulin binding protein BiP, interacts permanently with Cw3 alpha chains. Furthermore, the binding of p88, a calnexin candidate, seems delayed compared to that of p105 and p78. As the great majority of newly synthesized Cw3 alpha proteins expressed in P815 cells can associate with cotransfected human beta 2-microglobulin (beta 2m), our observations suggest that multiple molecular chaperones cooperate in the folding of class I heavy chains. We were unable to coimmunoprecipitate detectable levels of these proteins with oligomerized Cw3 alpha chains. However, we could still detect p78/BiP in transient association with mutant HLA-Cw3 heterodimers which were delayed in the endoplasmic reticulum (ER) compared to their wild-type counterparts. In this case, the dissociation of BiP preceded the ER to Golgi transport of these proteins. These results suggest that BiP release is neither related to the process of class I oligomerization nor to the ER retention of class I assembly intermediates.

Cytotoxic T lymphocyte recognition of secreted HLA class I molecules.

The cytolytic responses of DBA/2 mice against syngeneic transfected P815 mastocytoma cells expressing either membrane-associated (HLA-Cw3) or -secreted hybrid (HLA-Cw3 x H-2 Q10b) molecules were compared. In spite of the absence of serologically detectable hybrid molecules on their plasma membrane, cells secreting these molecules elicited a CTL response similar to that of cells expressing the membrane associated HLA-Cw3 molecules, in terms of both MHC-restriction and peptide specificity. Together with the observation that syngeneic mice were capable of rejecting the injected secreting cells, these results imply that secreted HLA class I molecules can function as minor histocompatibility Ag and suggest that processing of both the membrane-bound and the -secreted forms of a protein may follow common or overlapping pathways.

Hybrid genes between HLA-A2 and HLA-A3 constructed by in vivo recombination allow mapping of HLA-A2 and HLA-A3 polymorphic antigenic determinants.

HLA-A2 and -A3 genes have been modified in their third exon (second domain) by using in vivo recombination. In this method Escherichia coli are transfected with a plasmid which contains two highly homologous sequences (e.g., the third exons of HLA-A2 and -A3) and has been linearized by cleavage between these two sequences. Circularization takes place in the bacteria by homologous recombination leading to hybrid A2-A3 sequences. The analysis by DNA sequencing of a number of such recombinants shows that they indeed occur by homologous recombination (no insertions or deletions) and that the probability of crossing over decreases as the distance from the free end of DNA in the homologous region increases. No double recombinants were observed. These hybrid exons were reinserted into either HLA-A2 or HLA-A3 genes, thus generating a panel of functional hybrid genes containing one or several HLA-A2 specific substitutions in an HLA-A3 background or vice versa. These genes were expressed by transfection into murine P815-high transfection efficiency recipient cells. Serologic analysis leads to the conclusion that expression of polymorphic antigenic determinants specific for HLA-A2 (detected with M58, A2A28M1, and CR11.351 mAb) is linked to the presence of threonine residue (amino acid (AA) 142) and/or histidine residue (AA 145) and valine residue (AA 152). The expression of specific HLA-A3 polymorphic determinants (recognized by GAP-A3 mAb) is correlated with the existence of a asparagine residue (AA 127) and a aspartic residue (AA 161). But aspartic residue 161 contributes with glutamic acid residue 152 in the formation of the A3 epitope recognized by the anti-A3 mAb X1.23.2.

Acquisition of HLA class I W6/32 defined antigenic determinant by heavy chains from different species following association with bovine beta 2-microglobulin.

Murine, rat, rabbit and guinea pig class I heavy chains, which do not react with W6/32 monoclonal antibody when they are expressed in association with autologous beta 2-microglobulin (beta 2-m), can acquire such a reactivity once they are expressed at the surface of cells cultured in conditions which allow their association with bovine beta 2-m. Sequence comparison of beta 2-ms suggests that glutamine at position 89 might be critical for the induction of the W6/32 defined antigenic determinant. However, in the murine species, certain class I heavy chains, in spite of their association with bovine beta 2-m, do not express this determinant. Using genetically engineered hybrid class I molecules and selected congenic strains of mice this negative property was shown to be related to the presence of a cysteine residue at position 121 which allows covalent association of beta 2-m to class I heavy chains (Bushkin, Y., J-S. Tung, A. Pinter, J. Michaelson, and E. A. Boyse. 1986. Unusual association of beta 2-microglobulin with certain class I heavy chains of the murine major histocompatibility complex. Proc. Natl. Acad. Sci. USA 83:432). Therefore, expression of the W6/32 defined antigenic determinant implicates both the beta 2-m and the second domain of the heavy chain, but its expression (or exposure) is prevented by the covalent fixation on cysteine 121 of the light chain.

Creation of an HLA-A2/HLA-Aw69 alloantigenic determinant on an HLA-A3 molecule by site-directed mutagenesis.

The HLA-A2 and HLA-Aw69 molecules share an antigenic determinant not expressed by HLA-Aw68 and HLA-A3. Comparison of the amino acid (aa) sequences of these molecules and previous studies of the antigenic determinant expressed by different HLA-A2 X HLA-A3 hybrid molecules had established that three aa at positions 95, 97, and 107 were possibly involved in the formation of this determinant. The HLA-A3 gene was therefore mutagenized to replace successively at these positions the HLA-A3-specific aa by the HLA-A2 residues. A single substitution at position 107 of a glycine by a tryptophan residue is sufficient for full expression by HLA-A3 molecules of the HLA-A2/Aw69 shared antigenic determinant without modification of the other serological reactivities characteristic of the HLA-A3 molecules. Previous studies of ethyl methanesulfonate mutants having shown the involvement of aa 161 in this determinant, we assume that the two aa residues 107 and 161 are close to each other.

Serological and structural analysis of HLA class I molecules: beta 2-microglobulin interacts with the two external domains of the HLA class I heavy chain.

The serological reactivities of HLA class I molecules were studied in relation to structural modifications of these molecules, including shuffling of external exons and exchange of human beta 2-microglobulin for beta 2-microglobulin from different species. Two major clusters (I and II) of monomorphic and polymorphic antigenic determinants could be delineated. beta 2-Microglobulin participates in the formation of the two clusters, indicating that the light chain interacts tightly with the two external domains of the HLA class I heavy chain. However, external molecules can modify these interactions and alter the antigenic structure of the overall molecule. Thus, fixation on HLA class I molecules of the Fab fragment of a monoclonal antibody directed at antigenic determinants associated with cluster II resulted in enhanced fixation of a monoclonal antibody (B10.6) related to cluster I. The structural and functional implications of these results are discussed.

Recognition of hybrid HLA molecules expressed on murine P815 cells using human alloreactive cytotoxic T lymphocytes.

The HLA-A2 and -A3 genes were used to construct intra-exon hybrids in which part of the third exon (corresponding to the second domain) was of one type and the rest of the other. Murine P815 cells expressing these hybrid constructs were assayed with human alloreactive cytotoxic T-lymphocyte lines specific for either HLA-A2 or HLA-A3. Specific recognition patterns were obtained which indicate that, in some cases, a small portion of the HLA-A2 sequence in an HLA-A3 background is sufficient for recognition by HLA-A2-specific cytotoxic T cells.

Transfection of murine LMTK- cells with purified HLA class I genes. VII. Association of allele- and locus-specific serological reactivities with respectively the first and second domains of the HLA-B7 molecule.

The individual contributions of the first two external domains of the HLA-B7 heavy chain to the expression of allele-specific (B7) and locus-specific (B and C) antigenic determinants were investigated using hybrid class I genes. Hybrid genes were constructed in vitro by exon shuffling between the parent genes HLA-B7, HLA-Cw3, HLA-A3, and H-2Kd, and their expression was monitored following transfection into mouse L cells. The results show that most allele-specific antigenic determinants are associated with the first external domain of the B7 heavy chain, whereas all the locus-specific antigenic determinants tested map to the second external domain.

HLA class I genes: from structure to expression, serology and function.

HLA class I genes have been isolated from phage and cosmid libraries and assayed by transfection into murine L cells. The transfection step proved to be very important because of the large number of genes (and pseudogenes) in this family. All functional genes characterized so far in this way are "classical" class I genes, i.e. members of the HLA-A, -B or -C families. Three of these have been sequenced (HLA-A3, -Aw24; HLA-Cw3) in addition to the pHLA 12.4 pseudogene. Sequence comparisons indicate, in particular, extreme conservation of the 3' non-coding region between allelic HLA-A locus genes; the general organization of all these genes (8 exons) is very similar. Restriction mapping around the functional genes has been performed to investigate the degree of conservation (e.g. between HLA-A3 regions from 2 different individuals) and examine allelism at the DNA level (e.g. between HLA-A3 and HLA-Aw24 regions). Exon shuffling experiments followed by serological analysis of the expressed product indicate that, as expected, specificities are determined by the first two domains of the molecule. However, further constructs show that as soon as a single exon is exchanged most specific reactivities disappear. CTL analysis of murine cells expressing HLA molecules has run into many difficulties but still holds promise for the study of structure-function relationships in this system.

Delineation of three subsets of class I human T antigens (HTA) on Molt-4 cells: serologic and regulatory relationship to HLA class I antigens.

Three subsets of class I human T antigens (HTA) were serologically identified on the surface of the Molt-4 T lymphoma cell line. The HTA 1 subset is defined by NAI/34, D47, or 10H3.9 cross-reactive m.Ab. and by BL6 m.Ab. The HTA 2 and HTA 3 subsets are defined by M241 and 4A7.6 m.Ab., respectively. We obtained no evidence of any additional HTA subset. The different HTA antigens share only few epitopes with human leukocyte antigens (HLA-A, -B, and -C). Interestingly, these epitopes all belong to the same cluster defined on HLA class I molecules, but differ from one HTA subset to another. These results would therefore suggest that HTA and HLA class I antigens display a limited structural homology, but have a conserved epitopic area whose detailed structure differs for each HTA subset. Furthermore, the cell surface expression of each HTA class I molecule type is differently enhanced by natural interferon (IFN)-alpha or -gamma. This result additionally supports the serologic delineation of HTA subsets, and suggests that the corresponding genes in Molt-4 cells, are subjected to distinct regulations.

Putative conformation of mouse Fc gamma-receptor.

Three cell surface molecules (m.w. = 115,000, 90,000, and 70,000) binding to the Fc portion of complexed IgG have been isolated from the murine mastocytoma line P815. Various results suggested that the 90,000 and 70,000 dalton components are generated from the 115,00 dalton molecule by spontaneous proteolytic clevages and release of 23,000 dalton fragments. It was demonstrated that these cleavages occur during cell culture and not when freshly harvested mouse spleen cells are used as an Fc gamma R cell source. The survey of the Fc gamma R molecular forms obtained from P815 and spleen cells together with their reduction products led us to conclude that the mouse Fc gamma-receptor for complexed IgG is a single chain molecule (115,000 daltons) folded into five globular subunits (m.w. eta 23,000) linked by loose connecting regions accessible to proteolytic enzymes. Three of these subunits that compose the 70,000-dalton fragment are linked by di-sulfide bonds. Furthermore, a 140,000-dalton Fc gamma-binding molecule, not identified after cell surface labeling, could be detected after internal labeling. This component could be a cytoplasmic precursor of the Fc gamma R molecule. The structural model we present here might in addition shed some light on the discrepancy that appears through the various biochemical studies performed so far on Fc gamma-receptors.