Interferon (IFN) beta acts downstream of IFN-gamma-induced class II transactivator messenger RNA accumulation to block major histocompatibility complex class II gene expression and requires the 48-kD DNA-binding protein, ISGF3-gamma.

Interferon (IFN) gamma, a cardinal proinflammatory cytokine, induces expression of the gene products of the class II locus of the major histocompatibility complex (MHC), whereas IFN-alpha or -beta suppresses MHC class II expression. The mechanism of IFN-beta-mediated MHC class II inhibition has been unclear. Recently, a novel factor termed class II transactivator (CIITA) has been identified as essential for IFN-gamma-induced MHC class II transcription. We studied the status of IFN-gamma-induced CIITA messenger RNA (mRNA) accumulation and CIITA-driven transactivation in IFN-beta-treated cells and used cell lines that had defined defects in the type I IFN response pathway to address the roles of IFN signaling components in the inhibition of MHC class II induction. IFN-beta treatment did not suppress IFN-gamma-induced accumulation of CIITA mRNA. After cells were stably transfected with CIITA, endogenous MHC class II genes were constitutively expressed, and MHC class II promoters, delivered by transfection, were actively transcribed in CIITA-expressing cells. Expression of these promoters was significantly impaired by pretreatment with IFN-beta. These results suggest that IFN-beta acts downstream of CIITA mRNA accumulation, and acts in part by reducing the functional competence of CIITA for transactivating MHC class II promoters. IFN stimulated gene factor 3 (ISGF3) gamma was essential for IFN-beta to mediate inhibition of MHC class II induction, regardless of whether MHC class II transcription was stimulated by IFN-gamma or directly by CIITA expression. Results of these experiments suggest that inhibition of MHC class II in IFN-beta-treated cells requires expression of gene(s) directed by the ISGF3-IFN-stimulated response element pathway, and that these gene product(s) may act by blocking CIITA-driven transcription of MHC class II promoters.

Association of STATs with relatives and friends.

Members of the STAT family of transcription factors are present in species as diverse as mammals, insects and slime molds. Discovered as mediators of interferon-induced signals, the STATs were later shown to drive many different ligand-induced responses through receptor-induced tyrosine phosphorylation and dimerization. STAT1 also functions as a transcription factor, essential for the efficient constitutive expression of certain genes, without needing tyrosine phosphorylation, and phosphorylated STAT1 dimers mediate suppression - rather than activation - of some genes. STATs are present in the cytoplasm of untreated cells in multiprotein complexes, which might aid in their nuclear translocation and differential binding to DNA, thus contributing to the specificity of STAT action. This review explores the diverse protein-protein interactions that underlie the multiple functions of the STATs.

Jak-STAT pathways and transcriptional activation in response to IFNs and other extracellular signaling proteins.

Through the study of transcriptional activation in response to interferon alpha (IFN-alpha) and interferon gamma (IFN-gamma), a previously unrecognized direct signal transduction pathway to the nucleus has been uncovered: IFN-receptor interaction at the cell surface leads to the activation of kinases of the Jak family that then phosphorylate substrate proteins called STATs (signal transducers and activators of transcription). The phosphorylated STAT proteins move to the nucleus, bind specific DNA elements, and direct transcription. Recognition of the molecules involved in the IFN-alpha and IFN-gamma pathway has led to discoveries that a number of STAT family members exist and that other polypeptide ligands also use the Jak-STAT molecules in signal transduction.

A single phosphotyrosine residue of Stat91 required for gene activation by interferon-gamma.

Interferon-gamma (IFN-gamma) stimulates transcription of specific genes by inducing tyrosine phosphorylation of a 91-kilodalton cytoplasmic protein (termed STAT for signal transducer and activator of transcription). Stat91 was phosphorylated on a single site (Tyr701), and phosphorylation of this site was required for nuclear translocation, DNA binding, and gene activation. Stat84, a differentially spliced product of the same gene that lacks the 38 carboxyl-terminal amino acids of Stat91, did not activate transcription, although it was phosphorylated and translocated to the nucleus and bound DNA. Thus, Stat91 mediates activation of transcription in response to IFN-gamma.

Contribution of STAT SH2 groups to specific interferon signaling by the Jak-STAT pathway.

In response to specific ligands, various STAT proteins (signal transducers and activators of transcription) are phosphorylated on tyrosine by Jak protein kinases and translocated to the nucleus to direct gene transcription. Selection of a STAT at the interferon gamma receptor as well as specific STAT dimer formation depended on the presence of particular SH2 groups (phosphotyrosine-binding domains), whereas the amino acid sequence surrounding the phosphorylated tyrosine on the STAT could vary. Thus, SH2 groups in STAT proteins may play crucial roles in specificity at the receptor kinase complex and in subsequent dimerization, whereas the kinases are relatively nonspecific.

Defective TNF-alpha-induced apoptosis in STAT1-null cells due to low constitutive levels of caspases.

Signal transducers and activators of transcription (STATs) enhance transcription of specific genes in response to cytokines and growth factors. STAT1 is also required for efficient constitutive expression of the caspases Ice, Cpp32, and Ich-1 in human fibroblasts. As a consequence, STAT1-null cells are resistant to apoptosis by tumor necrosis factor alpha (TNF-alpha). Reintroduction of STAT1alpha restored both TNF-alpha-induced apoptosis and the expression of Ice, Cpp32, and Ich-1. Variant STAT1 proteins carrying point mutations that inactivate domains required for STAT dimer formation nevertheless restored protease expression and sensitivity to apoptosis, indicating that the functions of STAT1 required for these activities are different from those that mediate induced gene expression.

The role of p53 in regulating genomic stability when DNA and RNA synthesis are inhibited.

In addition to its induction by DNA damage, p53 is induced by drugs that starve cells for DNA and RNA precursors, or by inhibitors of DNA or RNA polymerase. In normal cells, the induction of p53 by dNTP starvation serves a protective role, mediating rapid, reversible cell-cycle arrest without DNA damage. In most cell lines, this first line of defense is missing, so that starvation for dNTPs causes DNA to break, thus increasing the probability of genomic instability, chromosome deletions and gene amplification. The mechanism of how p53 is induced remains unclear.

Cell type-specific signaling in response to interferon-gamma.

Type II interferon-gamma (IFN-gamma) is a pleiotropic cytokine that regulates many different cellular functions. The major signaling pathway activated by IFN-gamma involves sequential phosphorylation of the tyrosine residues of the Janus kinase (JAK) and signal transducer and activator of transcription (STAT) proteins, providing the primary mechanism through which gene expression is induced. However, recent work has revealed that the responses are complex, as shown by the activation of kinases in addition to JAKs, differential patterns of activation of STAT1, STAT3, and STAT5 in different cells, and activation of transcription factors other than STATs. This complexity is used to regulate biological functions differentially in a cell type-specific manner, by activating different specific signals and patterns of gene expression.

General method for cloning amplified DNA by differential screening with genomic probes.

Mutant Syrian hamster cell lines resistant to N-(phosphonacetyl)-L-aspartate, a potent and specific inhibitor of aspartate transcarbamylase, have amplified the gene coding for the multifunctional protein (CAD) that includes this activity. The average amount of DNA amplified is approximately 500 kilobases per gene copy, about 20 times the length of the CAD gene itself. A differential screening method which uses genomic DNAs as probes was developed to isolate recombinant phage containing fragments of amplified DNA. One probe was prepared by reassociating fragments of total genomic DNA from 165-28, a mutant cell line with 190 times the wild-type complement of CAD genes, until all of the sequences repeated about 200 times were annealed and then isolating the double-stranded DNA with hydroxyapatite. This DNA was highly enriched in sequences from the entire amplified region, whereas the same sequences were very rare in DNA prepared similarly from wild-type cells. After both DNAs were labeled by nick translation, highly repeated sequences were removed by hybridization to immobilized total genomic DNA from wild-type cells. A library of cloned DNA fragments from mutant 165-28 was screened with both probes, and nine independent fragments containing about 165 kilobases of amplified DNA, including the CAD gene, have been isolated so far. These cloned DNAs can be used to study the structure of the amplified region, to evaluate the nature of the amplification event, and to investigate gene expression from the amplified DNA. For example, one amplified fragment included a gene coding for a 3.8-kilobase, cytoplasmic, polyadenylated RNA which was overproduced greatly in cells resistant to N-(phosphonacetyl)-L-aspartate. The method for cloning amplified DNA is general and can be used to evaluate the possible involvement of gene amplification in phenomena such as drug resistance, transformation, or differentiation. DNA fragments corresponding to any region amplified about 10-fold or more can be cloned, even if no function for the region is known. The method for removing highly repetitive sequences from genomic DNA probes should also be of general use.

Activation of phosphatidylinositol 3-kinase in response to interleukin-1 leads to phosphorylation and activation of the NF-kappaB p65/RelA subunit.

The work of Reddy et al. (S. A. Reddy, J. A. Huang, and W. S. Liao, J. Biol. Chem. 272:29167-29173, 1997) reveals that phosphatidylinositol 3-kinase (PI3K) plays a role in transducing a signal from the occupied interleukin-1 (IL-1) receptor to nuclear factor kappaB (NF-kappaB), but the underlying mechanism remains to be determined. We have found that IL-1 stimulates interaction of the IL-1 receptor accessory protein with the p85 regulatory subunit of PI3K, leading to the activation of the p110 catalytic subunit. Specific PI3K inhibitors strongly inhibit both PI3K activation and NF-kappaB-dependent gene expression but have no effect on the IL-1-stimulated degradation of IkappaBalpha, the nuclear translocation of NF-kappaB, or the ability of NF-kappaB to bind to DNA. In contrast, PI3K inhibitors block the IL-1-stimulated phosphorylation of NF-kappaB itself, especially the p65/RelA subunit. Furthermore, by using a fusion protein containing the p65/RelA transactivation domain, we found that overexpression of the p110 catalytic subunit of PI3K induces p65/RelA-mediated transactivation and that the specific PI3K inhibitor LY294,002 represses this process. Additionally, the expression of a constitutively activated form of either p110 or the PI3K-activated protein kinase Akt also induces p65/RelA-mediated transactivation. Therefore, IL-1 stimulates the PI3K-dependent phosphorylation and transactivation of NF-kappaB, a process quite distinct from the liberation of NF-kappaB from its cytoplasmic inhibitor IkappaB.

Molecular cloning of a gene selectively induced by gamma interferon from human macrophage cell line U937.

A cDNA clone encoding a gamma interferon (IFN-gamma)-inducible mRNA in human cells of the macrophage lineage was isolated and characterized. The corresponding gene, gamma.1, was selectively induced by IFN-gamma, responding a hundredfold better to IFN-gamma than to IFN-alpha. The induction was rapid and transient, with maximal mRNA accumulation at about 3 h and decline to the basal level after 48 h. Transcriptional activation could be detected as early as 5 min after IFN-gamma stimulation and accounted entirely for the mRNA accumulation. The induction of gamma.1 by IFN-gamma was cell-type restricted, being seen only in macrophages and endothelial cells. In addition, phorbol ester-induced differentiation of promyelocytic HL-60 cells and promonocytic THP-1 cells rendered the gamma.1 gene inducible by IFN-gamma. The 1.0-kilobase gamma.1 cDNA sequence encoded a small predicted polypeptide of 38 amino acids and had a conserved sequence associated with rapidly turning over mRNAs. In vitro translation of the gamma.1 transcript yielded a 4,000-dalton polypeptide.

Preferential amplification of rearranged sequences near amplified adenylate deaminase genes.

In a previous study of three independent families of mutants selected for overproduction of adenylate deaminase (AMPD), we were not able to isolate a cDNA probe for the gene and so could not demonstrate its amplification directly. In addition to overproduction of AMPD, four proteins of unknown function, designated W, X, Y1, and Y2, accumulated, and by using the corresponding cDNA probes, we demonstrated amplification of all four genes. In independent mutant clones, sometimes all and sometimes only a subset of these genes were amplified. Assuming that all five genes are linked, the pattern of their coamplification suggested a genetic map in which AMPD lies between W and Y1. We show here that a two-step chromosome walk joins the W and Y1 genes, that the AMPD gene is the only expressed sequence between them, and that its amplification is indeed responsible for overproduction of the AMPD protein. In the course of this work, we cloned and studied two novel joints which mark rearrangements on either side of the AMPD gene. Each joint was generated independently in a single first-step mutant at single or low copy number. Remarkably, each joint was amplified preferentially in every second- and third-step mutant derived from the first-step line in which it was originally present, suggesting that the two independent rearrangements each generated amplification-prone structures.

Structure of the gene for CAD, the multifunctional protein that initiates UMP synthesis in Syrian hamster cells.

Two adjacent fragments of genomic DNA spanning the gene for CAD, which encodes the first three enzymes of UMP biosynthesis, were cloned from a mutant Syrian hamster cell line containing multiple copies of this gene. The mutant was selected for resistance to N-(phosphonacetyl)-L-aspartate, a potent and specific inhibitor of aspartate transcarbamylase, the second enzyme in the pathway. The sizes and positions of about 37 intervening sequences within the 25-kilobase CAD gene were mapped by electron microscopy, and the locations of the 5' and 3' ends of the 7.9-kilobase CAD mRNA were established by electron microscopy and by other hybridization methods. The coding sequences are small (100 to 400 bases), as are most of the intervening sequences (50 to 300 bases). However, there are also several large intervening sequences of up to 5,000 bases each. Two small cytoplasmic polyadenylated RNAs are transcribed from a region just beyond the 5' end of the CAD gene, and their abundance reflects the degree of gene amplification.

Properties of dispersed, highly repeated DNA within and near the hamster CAD gene.

Dispersed, highly repeated DNA sequences were found within and near the Syrian hamster gene coding for the multifunctional protein CAD. Most of the repeated sequences were homologous to each other and had similar properties. They hybridized to many cytoplasmic polyadenylated RNAs and to 7S and 4.5S cytoplasmic non-polyadenylated RNAs. Cloned DNA fragments containing repeated sequences were transcribed in vitro by RNA polymerase III. The repeated sequences from Syrian hamsters share many properties with the Alu family of repetitive DNA from humans. The hamster sequences were homologous to total repetitive human DNA but only very weakly homologous to two cloned members of the human Alu family.

Use of a selectable marker regulated by alpha interferon to obtain mutations in the signaling pathway.

We have selected mutations in genes encoding components of the signaling pathway for alpha interferon (IFN-alpha) by using a specially constructed cell line. The upstream region of the IFN-regulated human gene 6-16 was fused to the Escherichia coli guanine phosphoribosyltransferase (gpt) gene and transfected into hypoxanthine-guanine phosphoribosyltransferase-negative human cells. These cells express gpt only in the presence of IFN-alpha. They grow in medium containing hypoxanthine, aminopterin, and thymidine plus IFN and are killed by 6-thioguanine plus IFN. Two different types of mutants were obtained after treating the cells with mutagens. A recessive mutant, selected in 6-thioguanine plus IFN, was completely resistant to IFN-alpha but responded normally to IFN-gamma and, unexpectedly, partially to IFN-beta. A constitutive mutant, selected in hypoxanthine-aminopterin-thymidine alone, was abnormal in expressing endogenous genes in the absence of IFN. Both types revert infrequently, allowing selection for complementation of the defects by transfection.

Functional subdomains of STAT2 required for preassociation with the alpha interferon receptor and for signaling.

Two members of the STAT signal transducer and activator of transcription family, STAT1 and STAT2, are rapidly phosphorylated on tyrosine in response to alpha interferon (IFN-alpha). Previous work showed that in the mutant human cell line U6A, which lacks STAT2 and is completely defective in IFN-alpha signaling, the phosphorylation of STAT1 is very weak, revealing that activation of STAT1 depends on STAT2. We now find that STAT2 binds to the cytoplasmic domain of the IFNAR2c (also known as IFNAR2-2) subunit of the IFN-alpha receptor in extracts of untreated cells. STAT1 also binds but only when STAT2 is present. The activities of chimeric STAT2-STAT1 proteins were assayed in U6A cells to define regions required for IFN-alpha signaling. Previous work showed that a point mutation in the Src homology 2 (SH2) domain prevents STAT2 from binding to phosphotyrosine 466 of the IFNAR1 subunit of the activated receptor. However, we now find that the entire SH2 domain of STAT2 can be replaced by that of STAT1 without loss of function, revealing that other regions of STAT2 are required for its specific interaction with the receptor. A chimeric protein, in which the N-terminal third of STAT2 has replaced the corresponding region of STAT1, did preassociate with the IFNAR2c subunit of the receptor, became phosphorylated when IFN-alpha was added, and supported the phosphorylation of endogenous STAT1. These results are consistent with a model in which STAT2 and STAT1 are prebound to the IFNAR2c subunit of the resting receptor. Upon activation, the IFNAR1 subunit is phosphorylated on Tyr-466, allowing the SH2 domain of STAT2 to bind to it; this is followed by the sequential phosphorylation of STAT2 and STAT1.

Evolution and stability of chromosomal DNA coamplified with the CAD gene.

We have compared clones of Syrian hamster cells selected for the first amplification of the CAD gene with clones selected for further amplification. The large domain amplified initially was not reamplified as an intact unit. Instead, subregions were reamplified preferentially, and parts of the initial array were often lost. These events reduced the average amount of coamplified DNA accompanying each copy of the selected gene. The degree of amplification was small in the first step (about three extra copies of CAD per cell), but second-step amplifications to a high copy number (up to 60 extra copies per cell) occurred frequently. After several separate steps of amplification, highly condensed arrays that brought many CAD genes close together were formed. In striking contrast to the stability of these highly amplified arrays, the low-copy chromosomal arrays formed early were quite unstable and were often lost completely within 1 or 2 months of growth without selection. The results suggest that different mechanisms may be involved in the first step of amplification and in the later evolution of an already amplified array.

Interactions between STAT and non-STAT proteins in the interferon-stimulated gene factor 3 transcription complex.

The first STAT-containing transcription factor to be studied, the alpha-interferon-induced ISGF3, is composed of a Stat1:2 heterodimer and a weak DNA-binding protein, p48, that is a member of a growing family of proteins similar to the so-called interferon regulatory factor (IRF-1). The p48 and Stat1:2 heterodimer do not associate stably in the absence of DNA, but we show that amino acids approximately 150 to 250 of Stat1 and a COOH-terminal portion of p48 exhibit physical interaction, implying contact that stabilizes ISGF3. Moreover, amino acid exchanges within the Stat1 contact region diminish or abolish the functional activity of Stat1. This protein interaction domain may be important in other STAT proteins to recruit partners to multiprotein transcription factors.

Single-copy and amplified CAD genes in Syrian hamster chromosomes localized by a highly sensitive method for in situ hybridization.

Syrian hamster cells resistant to N-(phosphonacetyl)-L-aspartate (PALA), a specific inhibitor of the aspartate transcarbamylase activity of the multifunctional protein CAD, overproduce this protein as a result of amplification of the CAD gene. We have used a sensitive in situ hybridization technique to localize CAD genomes in spreads of metaphase chromosomes from several independent PALA-resistant lines and from wild-type PALA-sensitive cells. The amplified genes were always found within chromosomes, usually in an expanded region of the short arm of chromosome B9. In wild-type cells, the CAD gene was also on the short arm of chromosome B9. In one mutant line, 90 to 100 CAD genes were found within an expanded B9 chromosome and 10 to 15 more were near the distal end of one arm of several different chromosomes. Another line contained most the genes in a telomeric chromosome or large chromosome fragment. The amplified genes were in chromosomal regions that were stained in a banded pattern by trypsin-Giemsa. A few double minute chromosomes were observed in a very small fraction of the total spreads examined. The it situ hybridizations were performed in the presence of 10% dextral sulfate 500, which increases the signal by as much as 100-fold. Using recombinant DNA plasmids nick-translated with [125I]dCTP to high specific radioactivity, 10 CAD genes in a single chromosomal region were revealed after 1 week of autoradiographic exposure, and the position of the unique gene could be seen after 1 month.

Mutant cells that do not respond to interleukin-1 (IL-1) reveal a novel role for IL-1 receptor-associated kinase.

Mutagenized human 293 cells containing an interleukin-1 (IL-1)-regulated herpes thymidine kinase gene, selected in IL-1 and gancyclovir, have yielded many independent clones that are unresponsive to IL-1. The four clones analyzed here carry recessive mutations and represent three complementation groups. Mutant A in complementation group I1 lacks IL-1 receptor-associated kinase (IRAK), while the mutants in the other two groups are defective in unknown components that function upstream of IRAK. Expression of exogenous IRAK in I1A cells (I1A-IRAK) restores their responsiveness to IL-1. Neither NFkappaB nor Jun kinase is activated in IL-1-treated I1A cells, but these responses are restored in I1A-IRAK cells, indicating that IRAK is required for both. To address the role of the kinase activity of IRAK in IL-1 signaling, its ATP binding site was mutated (K239A), completely abolishing kinase activity. In transfected I1A cells, IRAK-K239A was still phosphorylated upon IL-1 stimulation and, surprisingly, still complemented all the defects in the mutant cells. Therefore, IRAK must be phosphorylated by a different kinase, and phospho-IRAK must play a role in IL-1-mediated signaling that does not require its kinase activity.