Involvement of an inducible factor in interferon-gamma-mediated accumulation of HLA gene transcripts

Different regulatory pathways employed in cytokine-enhanced expression of secretory component and epithelial HLA class I genes

The transmembrane secretory component (SC, or pIg receptor) plays a crucial role in mucosal immunity by translocating dimeric IgA and pentameric IgM through exocrine epithelia. This receptor is up-regulated by cytokines in parallel with increased epithelial HLA expression. By use of the human epithelial cell line HT-29m3, we show that IFN-gamma, TNF-alpha and IL-4 activate transcription of the SC gene. This activation was slow, suggesting mediation via newly synthesized protein factors. IFN-gamma and TNF-alpha, but not IL-4, also up-regulated expression of HLA class I genes. However, this gene induction was rapid and did not depend on new protein synthesis. Nuclear run-on experiments showed that the transcription rate of HLA class I genes nearly peaked after only 30 min of IFN-gamma or TNF-alpha stimulation, whereas the SC transcription rate did not peak until after 20-36 h of IFN-gamma, TNF-alpha or IL-4 stimulation. Gel electrophoresis mobility shift assays demonstrated binding of nuclear proteins from cytokine-stimulated HT-29 cells to consensus elements in the promoter of the SC gene, involving the binding site for the nuclear factor-kappaB p50 subunit after TNF-alpha stimulation, and IFN-stimulated response element after IFN-gamma stimulation (and weakly after TNF-alpha. Our observations in vitro likely parallel events in vivo by which activated mucosal T cells and macrophages enhance pIg receptor-mediated external transport of secretory IgA and IgM and up-regulate epithelial HLA expression.

Transforming growth factor-beta inhibits interferon-gamma-induced HLA-DR expression by cultured human fibroblasts

This study shows the induction of HLA-DR (DR) in fibroblasts by IFN-gamma and investigates the molecular mechanisms involved in the further DR down-regulation by TGF-beta 1. Kinetics of DR induction on human dermal fibroblasts by IFN-gamma showed that 1 hr of exposure was required to induce detectable levels of DR, and maximal DR expression was achieved only after 2 days of exposure to IFN-gamma. TGF-beta 1 inhibited DR induction by IFN-gamma, although complete inhibition never could be achieved, even with high concentrations of TGF-beta 1 and low concentrations of IFN-gamma. Inhibition was not accounted for by reduction in cell numbers, as TGF-beta 1 stimulated growth of the fibroblasts. Inhibition of DR induction was seen only if TGF-beta 1 was added during the first 24 hr of IFN-gamma treatment. TGF-beta 1 inhibited equally well if the cells were pretreated for as little as 1 hr and then washed before addition of IFN-gamma. TGF-beta 1 did not cause an overall suppression of protein synthesis. Northern blot analysis revealed that TGF-beta 1 greatly reduced the steady-state level of DR beta mRNA induced by IFN-gamma at 24 hr, and then DRP transcripts became undetectable at later stages. It is concluded that early intracellular signals must build up to stimulate maximum DR synthesis, which, later on, are inactivated or degraded by the action of TGF-beta 1. We suggest that these mechanisms regulating DR gene transcription involve the action of genes coding for specific IFN-gamma-inducible transcriptional factors that are turned on and off in an expeditious manner.

Involvement of two regulatory elements in interferon-gamma-regulated expression of human indoleamine 2,3-dioxygenase gene

Interferon (IFN)-gamma-induced expression of indoleamine 2,3-dioxygenase (IDO) gene is implicated in the antimicrobial and antiproliferative effects of IFN-gamma in cell cultures. Earlier studies identified a 96 base pair (bp) regulatory region upstream of the IDO gene that conferred IFN-gamma response to the chloroamphenicol acetyltransferase (CAT) gene linked to herpesvirus thymidine kinase promoter. The IFN-gamma-responsive region was further narrowed to a 67 bp fragment by 3' deletion. This 67 bp fragment contains several sequence elements of potential interest, including a 14 bp sequence homologous to the ISRE sequence found in IFN-alpha-inducible genes and two palindromic sequences (PE I and PE II) homologous to the GAS sequence identified in IFN-gamma-inducible genes. Site-directed mutagenesis studies showed that IFN-gamma-induced expression of IDO-CAT constructs involved cooperation between two elements: the ISRE homolog and the PE II (but not PE I). Either element alone with its flanking sequence was inadequate in conferring an IFN-gamma response to CAT reporter gene. Two IFN-gamma-regulated protein factors interacting with these two elements were identified. The factor binding to the ISRE region was induced with a slower kinetics, required new protein synthesis, and reacted with antibodies to IRF-1. The factor interacting with the PE II region appeared rapidly after treatment with IFN-gamma independently of new protein synthesis, and its binding to DNA probe was blocked by antibodies to p91 factor, reported to bind to GAS element.(ABSTRACT TRUNCATED AT 250 WORDS)

Effects of bile acids and cholestasis on major histocompatibility complex class I in human and rat hepatocytes

BACKGROUND/AIMS

Major histocompatibility complex (MHC) class I molecules, which are normally poorly expressed on the surface of hepatocytes, are overexpressed during cholestasis. The mechanisms responsible for this overexpression were examined.

METHODS

The expression of class I molecules, assessed by flow cytofluorimetry, and the class I messenger RNA (mRNA) transcripts, assessed by Northern blot analysis, were measured on normal human hepatocytes in primary culture.

RESULTS

Chenodeoxycholic acid induced an overexpression of MHC class I molecules, whereas ursodeoxycholic acid did not. The level of class I mRNA closely reflected that of the membrane protein. Moreover, cholestasis, induced in the rat by ligation-section of the common bile duct, increased the MHC class I mRNA level. Actinomycin D inhibited bile acid-induced class I transcription of rat hepatocytes in primary culture, whereas cycloheximide did not. Finally, class I mRNA expression was induced in hepatocytes by phorbol myristate acetate and by forskolin. This hyperexpression, as well as that observed with chenodeoxycholic acid, was suppressed by an inhibitor of protein kinase C and protein kinase A.

CONCLUSIONS

Taken together, these results suggest that chenodeoxycholic acid, as interferon, activates protein kinase C and protein kinase A, resulting in the induction of MHC class I expression.

Stimulation of beta 2-microglobulin synthesis in lymphocytes after exposure to Cuprophan dialyzer membranes

We tested beta 2-microglobulin (beta 2m) synthesis by peripheral blood lymphocytes of non-uremic donors after contact with dialyzer membranes. Under serum-free conditions, lymphocytes were incubated with different dialyzer membranes for four hours. After subsequent culture in the absence of membranes to yield an overall culture time of 24 hours, RNA was extracted. Messenger RNA for beta 2m was quantitated by Northern blotting with specific probes. Increased beta 2m mRNA was seen after a minimum of one hour contact time with Cuprophan both in the presence and absence of cycloheximide. While no consistent stimulation was seen with polyacrylonitrile or polycarbonate-polyether membranes, respectively, the stimulation of beta 2m mRNA with Cuprophan was of the same order of magnitude as that obtained with interferon gamma. Stimulation of beta 2m mRNA by Cuprophan was stereospecifically diminished by 5 mM L-fucose. Apart from stimulated transcription of beta 2m we could also show increased surface expression of MHC class I molecules, using FACS technique. The results further prove complement-independent blood cell activation by Cuprophan membranes. It is uncertain whether the results are relevant for beta 2m amyloidosis.

Regulation of cellular gene expression by interferon-gamma: involvement of multiple pathways

Interferons (IFNs) classified as type I (IFN-alpha and -beta) and type II (IFN-gamma) interact with different receptors and regulate the expression of a number of genes in common, whereas the expression of certain other genes is regulated differentially by the type I or type II IFNs. Regulation of cellular gene expression by IFN-gamma was studied with the help of two cDNA clones (called C5-4 and C13) isolated in our laboratory and cDNA clones for human leukocyte antigen (HLA) class I (HLA-B) and class II (HLA-DR alpha, -DR beta) genes. The results indicate that IFN-gamma induced the expression of the cognate genes, but in different manners. IFN-gamma induced the transcription of all four genes as determined by nuclear run-on transcription analyses, but the induction of C5-4 genes transcription was inhibited by cycloheximide and anisomycin, indicating that some newly synthesized protein, presumably induced by IFN-gamma, was required for the transcriptional activation of the C5-4 gene. On the contrary, IFN-gamma-induced transcription of HLA class I, class II and C13 genes was unaffected by cycloheximide or anisomycin. However, these inhibitors completely blocked the accumulation of the HLA class II gene transcripts, but not HLA class I or C13 gene transcripts. Results suggest that some newly synthesized protein factor was required for IFN-gamma-induced accumulation of HLA class II gene transcripts and played a role at a step subsequent to the transcriptional activation by IFN-gamma. Evidence was obtained which suggests that the putative protein factor(s) required was induced by IFN-gamma. These studies indicate that IFN-gamma regulates the expression of cellular genes through multiple pathways.