Overexpression of A1, an NF-kappaB-inducible anti-apoptotic bcl gene, inhibits endothelial cell activation.

A1 is an anti-apoptotic bcl gene that is expressed in endothelial cells (EC) in response to pro-inflammatory stimuli. We show that in addition to protecting EC from apoptosis, A1 inhibits EC activation and its associated expression of pro-inflammatory proteins by inhibiting the transcription factor nuclear factor (NF)-kappaB. This new anti-inflammatory function gives a broader dimension to the protective role of A1 in EC. We also show that activation of NF-kappaB is essential for the expression of A1. Taken together, our data suggest that A1 downregulates not only the pro-apoptotic and pro-inflammatory response, but also its own expression, thus restoring a quiescent phenotype to EC.

Bcl-2 and Bcl-XL serve an anti-inflammatory function in endothelial cells through inhibition of NF-kappaB.

To maintain the integrity of the vascular barrier, endothelial cells (EC) are resistant to cell death. The molecular basis of this resistance may be explained by the function of antiapoptotic genes such as bcl family members. Overexpression of Bcl-2 or Bcl-XL protects EC from tumor necrosis factor (TNF)-mediated apoptosis. In addition, Bcl-2 or Bcl-XL inhibits activation of NF-kappaB and thus upregulation of proinflammatory genes. Bcl-2-mediated inhibition of NF-kappaB in EC occurs upstream of IkappaBalpha degradation without affecting p65-mediated transactivation. Overexpression of bcl genes in EC does not affect other transcription factors. Using deletion mutants of Bcl-2, the NF-kappaB inhibitory function of Bcl-2 was mapped to bcl homology domains BH2 and BH4, whereas all BH domains were required for the antiapoptotic function. These data suggest that Bcl-2 and Bcl-XL belong to a cytoprotective response that counteracts proapoptotic and proinflammatory insults and restores the physiological anti-inflammatory phenotype to the EC. By inhibiting NF-kappaB without sensitizing the cells (as with IkappaBalpha) to TNF-mediated apoptosis, Bcl-2 and Bcl-XL are prime candidates for genetic engineering of EC in pathological conditions where EC loss and unfettered activation are undesirable.

High level expression of the Xlr nuclear protein in immature thymocytes and colocalization with the matrix-associated region-binding SATB1 protein.

The X-linked lymphocyte-regulated (Xlr) protein is a 30,000 Mr nuclear protein bearing homology with meiosis-specific proteins and expressed in late stage B lymphoid cell lines. In the present study we investigated its expression in the T lymphoid lineage. In adults, a high level of expression was detected in CD4-CD8- thymocytes. Most remarkably, the peak of Xlr expression occurred early during thymus cell ontogeny, precisely on days 14-15 of gestation, and was associated with the first wave of pre-T cell differentiation. Its onset preceded the rearrangement of TCR genes, as Xlr expression was conserved in thymus cells from RAG1(0/0) mice. The lower expression of Xlr on day 13 of fetal development, the bright Thy1+ phenotype of Xlr-positive cells, their large size, and their absence from subcapsular areas suggest that Xlr expression must be turned on within the thymus and not in prethymic precursors. From day 16 of gestation, Xlr expression decreased markedly. At birth and later, Xlr(high) cells were mostly large cells scattered throughout the cortical area. As shown by confocal microscopy, expression of Xlr closely overlapped that of SATB1, which binds special AT-rich DNA sequences associated with the nuclear matrix and plays an important regulatory role for many genes. The remarkably regulated expression of Xlr in the lymphoid cell lineage and of its homologue Xmr in the germ cell lineage suggests that they might play an important role in chromatin metabolism at critical stages of differentiation during which the genome undergoes irreversible rearrangements.

A20 inhibits NF-kappaB activation in endothelial cells without sensitizing to tumor necrosis factor-mediated apoptosis.

Expression of the NF-kappaB-dependent gene A20 in endothelial cells (EC) inhibits tumor necrosis factor (TNF)-mediated apoptosis in the presence of cycloheximide and acts upstream of IkappaBalpha degradation to block activation of NF-kappaB. Although inhibition of NF-kappaB by IkappaBalpha renders cells susceptible to TNF-induced apoptosis, we show that when A20 and IkappaBalpha are coexpressed, the effect of A20 predominates in that EC are rescued from TNF-mediated apoptosis. These findings place A20 in the category of "protective" genes that are induced in response to inflammatory stimuli to protect EC from unfettered activation and from undergoing apoptosis even when NF-kappaB is blocked. From a therapeutic perspective, genetic engineering of EC to express an NF-kappaB inhibitor such as A20 offers the mean of achieving an anti-inflammatory effect without sensitizing the cells to TNF-mediated apoptosis.

Accommodation of vascularized xenografts: expression of "protective genes" by donor endothelial cells in a host Th2 cytokine environment.

Organ xenografts under certain circumstances survive in the presence of anti-graft antibodies and complement, a situation referred to as "accommodation." We find that the endothelial cells (ECs) in hamster hearts that accommodate themselves in rats express genes, such as A20 and bcl-2, that in vitro protect ECs from apoptosis and prevent upregulation in those cells of proinflammatory genes such as cytokines, procoagulant and adhesion molecules. Hearts that are rejected do not express these genes. In addition, vessels of rejected hearts show florid transplant arteriosclerosis whereas those of accommodated hearts do not. Accommodated xenografts have an ongoing T helper cell type 2 (Th2) cytokine immune response, whereas the rejected grafts have a Th1 response. We propose a model for factors that contribute to the survival of xenografts and the avoidance of transplant arteriosclerosis.

Homotypic aggregation of CD103 (alpha E beta 7)+ lymphocytes by an anti-CD103 antibody, HML-4.

One monoclonal antibody, HML-4, directed against the alpha E beta 7 integrin (CD103), an integrin preferentially expressed on human intestinal intraepithelial lymphocytes (IEL), induced the homotypic aggregation of IEL and of a CD103+ MOLT16 cell line. Aggregation was an active adhesion event dependent on an intact cytoskeleton, on tyrosine phosphorylation but not on activation of protein kinase C. It was blocked by four other anti-CD103 antibodies but by none of the antibodies blocking known adhesion lymphocyte pathways. It was associated with a redistribution of the CD103 integrin in the areas of cell-cell contacts. These results indicated that HML-4-induced homotypic adhesion was mediated via CD103 and resulted from the binding of the integrin to an as yet undefined ligand expressed by CD103+ cells. This ligand was distinct from the epithelial ligand of CD103: in contrast with homotypic adhesion, heterotypic adhesion of CD103+ MOLT16 cells on two epithelial intestinal cell lines (DLD1 and HT29) was dependent on the presence of divalent cations, was not enhanced by HML-4, was inhibited by HML-1 but not by the three other antibodies with an inhibitory effect on homotypic adhesion. Finally, the study of conjugates between CD103+ and CD103- sublines derived from the MOLT16 cell line suggested that HML-4-induced homotypic aggregation resulted from homophilic CD103-CD103 interactions.