MCB
Home Help [Feedback] [For Subscribers] [Archive] [Search] [Contents]
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Nehyba, J.
Right arrow Articles by Bose, H. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Nehyba, J.
Right arrow Articles by Bose, H. R., Jr.

 Previous Article

Molecular and Cellular Biology, June 2002, p. 3942-3957, Vol. 22, No. 11
0270-7306/02/$04.00+0     DOI: 10.1128/MCB.22.11.3942-3957.2002
Copyright © 2002, American Society for Microbiology. All Rights Reserved.

A Novel Interferon Regulatory Factor (IRF), IRF-10, Has a Unique Role in Immune Defense and Is Induced by the v-Rel Oncoprotein

Jirí Nehyba,1 Radmila Hrdlicková,1 Joan Burnside,2 and Henry R. Bose, Jr.1*

Section of Molecular Genetics and Microbiology and Institute for Cellular and Molecular Biology, University of Texas at Austin, Austin, Texas 78712-1095,1 Delaware Biotechnology Institute, Newark, Delaware 197112

Received 23 October 2001/ Returned for modification 11 December 2001/ Accepted 19 February 2002


   ABSTRACT
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
The cloning and functional characterization of a novel interferon regulatory factor (IRF), IRF-10, are described. IRF-10 is most closely related to IRF-4 but differs in both its constitutive and inducible expression. The expression of IRF-10 is inducible by interferons (IFNs) and by concanavalin A. In contrast to that of other IRFs, the inducible expression of IRF-10 is characterized by delayed kinetics and requires protein synthesis, suggesting a unique role in the later stages of an antiviral defense. Accordingly, IRF-10 is involved in the upregulation of two primary IFN-{gamma} target genes (major histocompatibility complex [MHC] class I and guanylate-binding protein) and interferes with the induction of the type I IFN target gene for 2',5'-oligo(A) synthetase. IRF-10 binds the interferon-stimulated response element site of the MHC class I promoter. In contrast to that of IRF-1, which has some of the same functional characteristics, the expression of IRF-10 is not cytotoxic for fibroblasts or B cells. The expression of IRF-10 is induced by the oncogene v-rel, the proto-oncogene c-rel, and IRF-4 in a tissue-specific manner. Moreover, v-Rel and IRF-4 synergistically cooperate in the induction of IRF-10 in fibroblasts. The level of IRF-10 induction in lymphoid cell lines by Rel proteins correlates with Rel transformation potential. These results suggest that IRF-10 plays a role in the late stages of an immune defense by regulating the expression some of the IFN-{gamma} target genes in the absence of a cytotoxic effect. Furthermore, IRF-10 expression is regulated, at least in part, by members of the Rel/NF-{kappa}B and IRF families.


   INTRODUCTION
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
The interferon (IFN) regulatory factor (IRF) family is a group of transcription factors that have extensive homology in the DNA-binding domain (DBD) (36, 62). In mammals, nine family members have been identified: IRF-1 to -7, IRF-8 (IFN consensus sequence-binding protein), and IRF-9 (IFN-stimulated gene factor 3{gamma}/p48). In avian species, homologues of IRF-1, IRF-2, IRF-4, and IRF-8 have been described (39, 48, 63). An additional chicken IRF, cIRF-3, has been cloned that has the highest level of homology to IRF-7 but is not an IRF-7 orthologue (29). IRF family members have also been described in fish and amphibians (37, 106). Moreover, viral IRF-encoding genes are present in the genomes of several rhadinoviruses, including human herpesvirus 8, the virus associated with Kaposi's sarcoma (10, 61, 70).

The major function of the IRF family of transcription factors is the regulation of immune responses, especially those directed against viral infection. The immune response consists of innate defenses and an adaptive defense (22). The innate response is the first line of defense against a virus infection and involves the production of IFNs and other cytokines, the activation of complement, and natural killer cells. These events, in turn, stimulate the adaptive immune response, which enables recognition of antigens with a high degree of structural specificity. IRF family members play a crucial role in both the innate and adaptive responses. In response to viral infection, IRF-3, -5, and -7 rapidly translocate to the nucleus and activate the transcription of type I IFNs and other cytokines (58, 108). Type I IFNs are produced by a variety of cells and bind cell surface receptors activating the induction of genes with direct antiviral activity. The induction of genes such as those for Mx, 2',5'-oligo(A) synthetase (OAS), and guanylate-binding protein (GBP) may contribute to the establishment of the antiviral state in infected and neighboring cells (47). In addition, several IRFs are expressed directly in response to viral infection or type I IFN, resulting in amplification of the overall antiviral response and activation of the adaptive defense. The adaptive response is induced by antigens presented by major histocompatibility complex (MHC) class I or II molecules and leads to the clonal expansion of specific lymphoid populations. The expression of both MHC class antigens is also regulated, in part, by IRF transcription factors (13, 38, 104). Antigen-stimulated T cells and natural killer cells produce IFN-{gamma}, which functions to coordinate the innate and adaptive immune responses (7). IFN-{gamma} is an important inducer of IRF genes, MHC class genes, the gene for inducible nitric oxide synthetase, and other genes. Several IRFs are activated directly by antigen stimulation or by proliferation stimuli, which mimics activation by an antigen, such as concanavalin A (ConA) (62, 76). The targets of IRF transcription factors during the antiviral response are genes directly involved not only in virus elimination but also in differentiation, proliferation, and apoptosis of cells of the immune system. Disruption of this regulation leads to tumorigenesis, as documented by the association of cancer with failure to express IRF-1 or IRF-8, with mutations of IRF-2, and with the up-regulation of IRF-4 (43, 87, 102, 103).

The DBD of all IRFs, like that of c-myb, contains repeated tryptophan residues (62, 76). The DBD recognizes a GAAA sequence that is part of the motifs termed the IFN-stimulated response element (ISRE) and the IRF element (20). All IRFs, except IRF-1 and IRF-2, have an IRF association domain (IAD) that is responsible for interaction with other family members or transcription factors such as PU.1, E47, and STAT (78, 92, 93, 98). Another association domain (IAD2) that is present in IRF-1 and IRF-2 is important for their interaction with IRF-8 (68, 86). A nuclear localization signal has been identified in IRF-1, and similar sequences are present in other family members (39, 86). A bipartite nuclear retention signal located within the N terminus of the DBD has been identified in IRF-4, IRF-8, and IRF-9 (57). IRFs possess a transactivation domain in the middle of the protein (62).

The transcription factors of the Rel/NF-{kappa}B family share a conserved N-terminal Rel homology region that contains the DNA-binding and dimerization domains conferring DNA binding to {kappa}B sites (25, 26, 31). Most of the genes regulated by Rel/NF-{kappa}B are involved in immune, inflammatory, or stress responses (11, 31, 49). The avian and mammalian Rel/NF-{kappa}B family members are c-Rel, RelA, NF-{kappa}B1 (p50), NF-{kappa}B2 (p52), and RelB. Altered regulation of Rel/NF-{kappa}B activities is associated with oncogenesis (23, 28, 81). v-Rel, the naturally occurring, acutely oncogenic member of the Rel/NF-{kappa}B family, was formed as a result of a nonhomologous recombination event between the c-rel proto-oncogene and an avian retrovirus (27). v-Rel rapidly induces an invariably fatal lymphoma in avian species (8). v-rel transforms cells by inappropriately activating or repressing genes coding for cytokines, transcription factors, chaperons, receptors, inhibitors of apoptosis, and adhesion molecules, which are normally regulated by c-Rel and other Rel/NF-{kappa}B family members (40).

In this report, we describe the characterization of a novel IRF (IRF-10). IRF-10 is expressed principally in cells of hematopoietic origin, but its level of expression in bursal cells is very low. The inducible expression of IRF-10 by type I and II IFNs in fibroblasts and by ConA in T cells has delayed kinetics and is dependent on protein synthesis. The cells exogenously expressing IRF-10 have a prolonged induction of GBP expression after treatment with IFNs. IRF-10 induces the expression of MHC class I in fibroblasts. IRF-10 also binds to the ISRE site of the MHC class I promoter. The expression of IRF-10 is induced in bursal cell lines by v-Rel and c-Rel and in fibroblasts by IRF-4. v-Rel and IRF-4 synergistically cooperate in the induction of IRF-10 expression.


   MATERIALS AND METHODS
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Cloning of chicken IRF-10 cDNA. A computer homology search for genes similar to IRF-4 identified a chicken expressed sequence tag (EST) clone, pat.pk0019.d2 (GenBank accession number A1980252), that encodes the C-terminal region of a novel IRF protein that we designated IRF-10. Clone pat.pk0019.d2 was isolated from a ConA-activated chicken splenic T-cell cDNA library as a part of the University of Delaware's chicken EST project (97). The 5' sequence encoding the N terminus of IRF-10 was cloned by 5' rapid amplification of cDNA ends (RACE). The first-strand cDNA was obtained by using Powerscript reverse transcriptase (Clontech Laboratories, Inc., Palo Alto, Calif.) from 5 µl of total RNA isolated from splenic lymphocytes activated for 48 h by ConA (50 µg/ml). SMART-RACE (Clontech) was performed with 2.5 µl of the first-strand substrate with a specific primer (5'-CCTCGTGGTATTTGCCCTTGTAGACGG-3'), and the PCR products were cloned into pGEM-T Easy (Promega, Madison, Wis.). The longest 5' RACE clone, p8N2, contained a translation start site and the 5' untranslated region of the IRF-10 cDNA. To verify that the partial cDNA sequences from p8N2 and pat.pk0019.d2 were from the same gene, the complete open reading frame (ORF) of IRF-10 was amplified with an Advantage-GC cDNA PCR kit (Clontech) and primers that flanked the ORF on both sides (5'-AGCCGGGATGGCGGAGCCGGGGTCTCCCAT-3', 5'-CCAGGCTTCGCGTCCCCTCAGCTGCC-3') and cloned into pGEM-T Easy. The plasmids designated pGE.IRF10-1 to -4 containing cloned PCR fragments were sequenced. Because the pat.pk0019.d2 cDNA clone has an incomplete 3' untranslated region, the sequence of the 3' terminus of the IRF-10 mRNA was determined from another University of Delaware EST clone (pnl-b.pk0007.e2), which contains a sequence overlapping the sequence of pat.pk0019.d2 in 482 nucleotides (nt).

Plasmids. pTZ-IRF10 was constructed by cloning an EcoRI-AccI fragment of pGE.IRF10-4 containing the 5' end of the IRF-10 ORF and the AccI-NotI fragment of pat.pk0019.d2 containing the 3' region of the IRF-10 ORF into pTZ-A10 via a NotI-MluI oligonucleotide adaptor (5'-GGCCGCGGTCGACGA-3', 5'-CGCGTCGTCGACCGC-3'). pTZ-A10 was derived by cloning an XhoI-EcoRI-AccI-MluI oligonucleotide adaptor (5'-AATTGCTCGAGGCGAATTCGAGTCTACGAGCGGCCGCACGCGT-3', 5'-AGCTACGCGTGCGGCCGCTCGTAGACTCGAATTCGCCTCGAGC-3') between the EcoRI and HindIII sites of pTZ-18R. pTZ-IRF4 was described previously (39). pTZ-IRF8 was created by cloning a 1,370-bp EcoRI-BamHI fragment of a chicken IRF-8 (IFN consensus sequence-binding protein) cDNA clone into pTZ-18R (48).

The REV-T-based retroviral vectors pREV-TW, pREV-C, pREV-C{Delta}Xba, pREV-IRF-4, and pREV-IRF-4{Delta}E6, which expressed v-rel, c-rel, c-rel{Delta}Xba, IRF-4, and IRF-4{Delta}E6, were described previously (39, 42, 75). pREV-IRF-10, expressing IRF-10, was derived by cloning the XhoI-MluI fragment from pTZ-IRF10 into the XhoI and BssHII sites of retroviral vector pREV-0. pCSV11S3 contains an infectious genome of chicken syncytial virus (CSV) (21). pDS3+pREP is a replication-competent retroviral vector derived from a genomic clone of the Schmidt-Ruppin strain of Rous sarcoma virus (subgroup A) by deletion of v-src (14, 77). pDSv-Rel, derived from pDS3, was used for the expression of v-Rel in fibroblasts (54). pcDNAchIFN1 and pcDNAchIFN-{gamma} were used for the production of IFN1 and IFN-{gamma} in COS-1 cells as previously described (89, 94).

The pATH-I10#8 bacterial expression plasmid encodes a TrpE/IRF-10 fusion protein. This plasmid was created by cloning the PvuII-HindIII fragment of the IRF-10 PCR fragment between the SmaI and HindIII sites of pATH1 (52). This fragment was PCR amplified from cDNA prepared from ConA-stimulated spleen cells with IRF-10-specific primer IRF10AB-1 (5'-AGTCCCGGGCAGCTGGACATCTCCGAGCCTTAC-3'), containing an SmaI site, and primer IRF10AB-2 (5'-CTGAAGCTTCTAGGGGTACTCCTCACCGAAGCACAGGT-3'), containing a STOP codon and a HindIII site. The sequence fidelity of the IRF-10 fragment cloned into pATH-1 was verified by sequencing.

Chickens, cell lines, viruses, and tissue culture. Embryonated eggs obtained from pathogen-free White Leghorn chickens were obtained from Hy-Vac, Adel, Iowa (SPF-SC strain), and Charles River SPAFAS, North Franklin, Conn. (SPAFAS strain). Chicken embryonic fibroblasts (CEF) were prepared from 10- or 11-day-old embryos of the SPF-SC strain. Cells were cultured with Dulbecco's modified Eagle's medium (DMEM) supplemented with 5% newborn calf serum (Atlanta Biologicals, Norcross, Ga.), 5% chicken serum (Gibco BRL Life Technologies, Grand Island, N.Y.), 100 U of penicillin, and 50 µg of streptomycin per ml. Secondary cultures of CEF were used for transfection of retroviral plasmid DNA by a calcium phosphate method as described previously (39). Retroviruses were harvested between 5 and 7 days after transfection, and infectious titers of Rel-expressing viruses were determined by an immunochemical titration assay with a monoclonal antibody (HY87) (42, 95). Vesicular stomatitis virus (VSV) (Indiana strain) was a gift from E. Ulug.

The simian COS-1 cell line was used for production of recombinant IFN1 and IFN-{gamma} (67, 89). The supernatant fluids from the COS-1 cell line contained 104 IU of antiviral activity/ml, as determined by a VSV cytopathic effect inhibition assay (89). All of the other cell lines used were of chicken origin. DT40 is a B-cell line established from an avian leukosis virus (ALV)-infected chicken bearing a bursa-derived lymphoid tumor (5). DT40 cells are transformed bursal stem cells that are continuously undergoing immunoglobulin gene diversification (51). The DT95 cell line, which is also derived from a chicken with an ALV-induced lymphoid leukosis, exhibits a more mature phenotype and secretes immunoglobulin M (IgM) (5). MSB-1 and RP-1 are T-cell lines established from a Marek's disease virus-induced lymphoma (1, 73). AEV-1 is an avian erythroblastosis virus-transformed erythroid cell line (80). BM-2 is a macrophage-like cell line derived by transformation of yolk sac cells by avian myeloblastosis virus (72). 123/12, 160/2, and 123/6T are B-cell, T-cell, and macrophage-like v-rel-transformed cell lines (42). The cell lines TWB2, TWB4, and TWB5 were obtained by transformation of bursal lymphocytes in vitro by REV-TW (CSV).

Preparation of white blood cells, lymphocytes, and bone marrow cells. Peripheral white blood cells and lymphocyte-enriched fractions from the bursa, thymus, and spleen used to prepare RNA for Northern analyses were isolated with Histopaque (Sigma Chemical Co., St. Louis, Mo.). Bone marrow, including the stroma, was isolated from the femur and tibiotarsus bones and used for RNA preparation.

LipofectAMIN-mediated delivery of pI · pC. Cells were seeded 48 h before transfection in 150-mm-diameter plates in DMEM without antibiotics. The cells were washed with Opti-MEM I medium (Gibco BRL Life Technologies) and overlaid with 10 µg of pI · pC (Sigma Chemical Co.) per ml and 10 µl of LipofectAMIN 2000 (Gibco BRL Life Technologies) per ml in 8 ml of Opti-MEM I per plate. After 3 h, 20 ml of DMEM with serum lacking antibiotics was added to the cultures.

In vitro transcription-translation. Proteins were synthesized by using pTZ-18R-based plasmids pTZ-IRF10, pTZ-IRF4, and pTZ-IRF8, which are described above. Transcription templates were prepared by linearization of the plasmids with MluI, BssHII, or BamHI, respectively. The proteins were produced in a TNT T7 Quick Coupled Transcription/Translation system (Promega), which employs a rabbit reticulocyte lysate. A 1-µg sample of linearized transcription templates was used per 50-µl reaction mixture. Proteins were cotranslationally labeled with [35S]methionine (Perkin-Elmer, Inc., Boston, Mass.), separated on sodium dodecyl sulfate (SDS)-8% acrylamide gels, fixed for 1 h in 45% methanol-10% acetic acid, dried, and subjected to autoradiography.

Electrophoretic mobility shift assay. Nuclear extracts were prepared from 2 x 107 DT40 cells infected with REV-IRF-10 (CSV) or with CSV by a procedure described previously (88). The concentration of total protein was determined with Bio-Rad Protein Assay Kit II (Bio-Rad Laboratories, Inc., Richmond, Calif.). Nuclear extracts were adjusted to the same protein concentration with extraction buffer C (88). A double-stranded DNA oligonucleotide probe was prepared by annealing of a primer (5'-AGCTCAAGC-3') to an oligonucleotide template, followed by primer extension with the Klenow fragment of DNA polymerase I (Promega) in the presence of [{alpha}-32P]dCTP. The template used (5'-TCCGCCTTTCGCTTTCGCTTCACAACGCTTGAGCT-3') contained an ISRE-binding site (underlined) of the chicken MHC class I {alpha}-chain gene promoter (55).

For DNA-binding assays, aliquots of nuclear extract containing 8 µg of total protein were combined with 50 fmol of the probe in a total volume of 20 µl. The binding buffer used contained 10 mM Tris-HCl [pH 7.8], 30 mM KCl, 0.1 mM EDTA, 5% glycerol, 0.5 mg of acetylated bovine serum albumin per ml, 0.05% Nonidet P-40, 0.3 mM dithiothreitol, 0.5 mM phenylmethylsulfonyl fluoride, and 0.05 µg of poly(dC-dI)-poly(dC-dI) (Sigma Chemical Company) per µl. The nuclear extracts were first preincubated with poly(dC-dI)-poly(dC-dI) in ice-cold buffer for 10 min, and after the addition of a radioactively labeled oligonucleotide, the reaction mixture was incubated for 30 min on ice. For supershift experiments, 2 µl of rabbit antiserum was added and the incubation was continued for an additional 15 min at 25°C. The complexes were separated from the free DNA probe by electrophoresis through a 4% polyacrylamide gel in 0.5x TBE (0.09 M Tris-borate, 2 mM EDTA [pH 8.0]). The gels were run at room temperature at 125 V, fixed for 5 min in 45% methanol-10% acetic acid, dried, and then analyzed by autoradiography.

The specificity of DNA binding was verified by the addition of competitor oligonucleotides added to the binding reaction mixture. The double-stranded competitor oligonucleotide containing the MHC class I ISRE site was obtained by annealing of the oligonucleotide template with this site (described above) with complementary oligonucleotide 5'-AGCTCAAGCGTTGTGAAGCGAAA GCGAAAGGCGGA-3'. The double-stranded oligonucleotide containing a mutated ISRE site was obtained by annealing of oligonucleotide 5'-TCCGCCTTTCAGCTTTCAGCTTCACAACGCTTGAGCT-3' with complementary oligonucleotide 5'-AGCTCAAGCGTTGTGAAGCTGAAAGCTGAAAGGCGGA-3'.

Antisera. The AI10-8 polyclonal antiserum was raised in a rabbit immunized with a TrpE/IRF-10 fusion protein purified from C600 bacteria transformed by the pATH-I10#8 expression plasmid. The TrpE/IRF-10 fusion protein contains amino acids 101 to 356 of chicken IRF-10. The goat anti-rabbit IgG biotinylated antibody was purchased from Kirkegaard & Perry Laboratories (Gaithersburg, Md.).

Western analysis. Western analysis was performed as described previously (42). Briefly, harvested cells were washed, resuspended, and boiled in SDS sample buffer and proteins were separated on an SDS-polyacrylamide gel with a Mini-PROTEAN II apparatus (Bio-Rad Laboratories). Proteins were transferred to a PolyScreen polyvinylidene difluoride membrane (Perkin-Elmer Life Science) and sequentially reacted with rabbit polyclonal antiserum AI10-8, goat anti-rabbit IgG biotinylated antibody, and streptavidin-linked alkaline phosphatase (Roche Molecular Biochemicals, Indianapolis, Ind.). Proteins were visualized by an enzymatic reaction with 5-bromo-4-chloro-3-indolyl phosphate and 4-nitroblue tetrazolium chloride as substrates (Roche Molecular Biochemicals). Calibrated prestained molecular weight markers from Bio-Rad Laboratories were used. Nuclear and cytoplasmic fractions of CEF were prepared in hypotonic buffer (50 mM Tris-HCl [pH 8], 1.1 mM MgCl2, 0.5% Triton X-100) as previously described (71).

Northern analysis and probes. Total RNA was isolated by RNAwiz (Ambion, Austin, Tex.). RNA was separated by electrophoresis in a 1% agarose gel in 20 mM MOPS [3-(N-morpholino)propanesulfonic acid buffer] and transferred to a Hybond N+ membrane (Amersham Pharmacia Biotechnology, Piscataway, N.J.). Parallel samples with ethidium bromide were analyzed under identical conditions, and the gel was photographed. Filters were hybridized with DNA fragments of the chicken genes labeled with [{alpha}-32P]dCTP by nick translation as described previously (42). The following list of probes includes the name of the gene product, followed, in parentheses, by the GenBank accession number, the positions of the first and last nucleotides of the fragment based on the GenBank sequence, and the reference describing the isolation of the gene used: cIRF-3 (U20338; 59 to 767; 29), c-Rel (X52193; 16 to 1865; 12), GBP (X92112.2; 1499 to 1876; 91), IRF-1 (L39766.1; 1 to 1935; 48), IRF-2 (X95478.1; 1 to 837; 63), IRF-4 (AF320331; 473 to 1357; 39), IRF-8 (L39767.1; 1 to 1640; 48), IRF-10 (AF380350; 186 to 1334; this report), Mx (Z23168; 1 to 2545; 6), and MHC class II ß chain (M26306.1; 673 to 936; 105). A probe for OAS was derived from University of Delaware EST clone ptr1c.pk002.b9 (AW240080.1). The fragment used corresponds to nt 831 to 1272 of the GenBank OAS sequence (accession number AB002585). The probe for MHC class I {alpha} chain was prepared by PCR amplification of cDNA from a v-Rel-transformed cell line. The sequence of the region amplified corresponded to nt 360 to 896 of the chicken MHC class I {alpha} chain (B2 allele) mRNA (GenBank accession number AF013492.1). There was only a single nucleotide difference in the sequence of the amplified fragment from the GenBank sequence.

Sequence analysis and phylogenetic tree construction. The nucleotide sequence of the gene for chicken IRF-10 was compared with the GenBank, EMBL, DDBJ, and PDB databases by using the BLAST 2.0 search engine, which is available at http://www.ncbi.nlm.nih.gov/BLAST (2, 3). The same search engine was used to compare the sequence encoded by the ORF for chicken IRF-4 with the protein sequence databases. All IRF sequences encoding the full-length proteins were retrieved on the basis of their similarity to chicken IRF-10. Additionally, a sequence derived from an EST clone (gb|AU090774.1) that encodes Japanese flounder IRF-8 and is likely incomplete at the C terminus was also included in a further analysis. A protein sequence alignment was then constructed by the ClustalX program (46). The alignment was visualized with the MacBoxshade 2.15 analysis tool (http://www.imtech.res.in/pub/oth/macboxshade/mac/macboxshade215.hqx). A phylogenetic tree was constructed by the neighbor-joining method as implemented in ClustalX (83). The tree was plotted by the tree-drawing program Unrooted (http://pbil.univ-lyon1.fr/software/unrooted.html). One thousand bootstrap replicates were generated by ClustalX for the bootstrap tests.

Nucleotide sequence of IRF-10. The sequence of chicken IRF-10 is a composite of the sequences of three cDNA clones: p8N2, pat.pk0019.d2, and pnl-b.pk0007.e2. Nucleotide 118, which was an A in the 5' RACE clone (p8N2), was changed to a G because other, independently cloned sequences (five clones) contained a G at this position and we concluded that the A at this position was a PCR artifact. The pat.pk0019.d2 and pnl-b.pk007.e2 (GenBank accession numbers AI980252 and AW198398) are University of Delaware EST clones. These two clones overlap at 482 nt and differ by a single nucleotide (the C at position 1117 of composite sequence is a T in pnl-b.pk0007.e2). This difference does not change the amino acid sequence.

Nucleotide sequence accession no. The sequence of chicken IRF-10 was submitted to the GenBank database under accession number AF380350.


   RESULTS
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Cloning and sequence characterization of a novel IRF (IRF-10). The expression of IRF-4 and several other members of the IRF family has been shown to be upregulated in v-Rel-expressing cells, and an increased level of IRF-4 plays a role in the v-Rel-mediated transformation process (39). A computer homology search of EST databases identified a chicken EST clone (pat.pk0019.d2) isolated from activated T cells that encodes an N-terminally truncated protein similar to IRF-4 (97). The sequence of this clone was, however, significantly different from those of IRF-4 and other known IRF family members, indicating that this gene encodes a novel IRF. The new gene product was designated IRF-10 in accordance with the nomenclature employed for the mammalian IRF-encoding genes (84). Since Northern blot analysis demonstrated that IRF-10 expression is induced in lymphoid cells by v-Rel, the function of this gene product may be relevant to the v-Rel transformation process; therefore, the gene was further characterized.

The region encoding the missing N terminus of IRF-10 was obtained by 5' RACE, and the continuity of its sequence with the sequence of the EST clone was verified by PCR amplification of the entire ORF. The sequence of the 3'-terminal untranslated region was obtained from another EST clone (pnl-b.pk0007.e2) from liver cDNA (J. Burnside, unpublished data). The composite sequence of the avian IRF-10 cDNA is 1,560 nt long (Fig. 1A). The sequence contains a polyadenylation signal (AATAAA) at nt 1541 to 1546 located upstream of the poly(A) tract that follows immediately after nt 1560. The longest ORF encodes a protein of 416 amino acids. The start codon of this ORF is located at nt 41 to 43 and lies in favorable sequence context for the initiation of translation with a purine in the -3 position and a G in the +4 position (53).



View larger version (97K):
[in this window]
[in a new window]
  		FIG. 1. Characterization of the IRF-10 sequence. (A) Nucleotide and predicted amino acid sequences of chicken IRF-10 cDNA. The polyadenylation signal is doubly underlined. The last T nucleotide at position 1560 is followed in the pnl-b.pk0007.e2 clone by a tract of multiple A nucleotides. The 1,560-nt IRF-10 sequence plus the usual 200-nt poly(A) tract corresponds approximately to the 1.8- to 1.9-kb size of the IRF-10 mRNA estimated by Northern blot analysis (data not shown). (B) Sequence alignment of the predicted amino acid sequences of chicken IRF-10, IRF-4, and IRF-8. Light letters on a dark background identify identical amino acids. The lines under the sequences show the locations of the five functional regions of IRF family members. NT, N terminus; MR, middle region that includes the transactivation domain; CT, C terminus.

 
The nucleotide and predicted amino acid sequences of chicken IRF-10 were compared with all of the sequences in the combined nucleotide and protein sequence databases. One additional chicken sequence—EST clone pgf1n.pk008.d16 from fat tissue cDNA (L. Cogburn et al., unpublished data)—was found that encodes a protein identical to IRF-10. All of the additional sequences found represented other IRF family members. Compared to these IRF family members, the IRF-10 protein sequence showed the greatest homology to chicken, human, and mouse IRF-4 and IRF-8, being 43 to 44% identical to IRF-4 and 36 to 37% identical to IRF-8 proteins (Fig. 1B; data not shown). These homology scores are far lower than the scores between the chicken and mammalian orthologues of IRF-4 or IRF-8 (84 or 73% identity, respectively). IRF-10 had 30% or less sequence identity with the other IRF proteins. We also compared the homology between the individual functional domains of IRF-10 with those of its closest relatives, IRF-4 and IRF-8 (Table 1). The level of sequence identity in the less conserved domains (the IAD and especially the middle region and C terminus) demonstrated a closer relationship of IRF-10 to IRF-4 than to IRF-8. The sequences of the individual domains showed much less similarity between IRF-10 and IRF-4 or -8 than between avian and mammalian orthologues of IRF-4 or IRF-8, suggesting that IRF-10 has a long separate evolutionary history.


View this table:
[in this window]
[in a new window]
  		TABLE 1. Percentages of identical amino acids within predicted functional domains of IRF-10, IRF-4, and IRF-8

 
To better understand the position of IRF-10 in the evolutionary history of the IRF family, a phylogenetic tree was constructed from the sequences of the IRF proteins from mammals, birds, amphibians, and fish (Fig. 2; Table 2). The tree indicates that the IRF proteins can be classified into four subfamilies, IRF-1, IRF-3, IRF-4, and IRF-5. IRF-10 belongs to the IRF-4 subfamily together with IRF-4, IRF-8, and IRF-9. The avian IRFs (IRF-1, -2, -4, and -8) are more closely related to their mammalian orthologues than IRF-10 is to its closest relative, IRF-4. The tree also suggests the existence of a fish homologue of IRF-8 (Paralichthys olivaceus IRF-8) that is distinct from IRF-10. Additional fish members of the IRF-4 subfamily not included in Fig. 2 were tentatively identified by analysis of EST databases, i.e., an IRF-4 encoded by a clone from Japanese flounder (P. olivaceus) leukocytes (dbj|AU091159.1) and an IRF-9 encoded by a clone isolated from a zebra fish (Danio rerio) kidney (gb|BG737308.1). Early diversification also took place in the other three IRF groups, as the GenBank database contains fish EST sequences similar to those of avian and mammalian IRF-2 (gb|AF146733.1), cIRF-3 (gb|BE605317.1), IRF-5 (gb|AW174576.1 and gb|AW203040.1), and IRF-6 (gb|AW422721.1).



View larger version (31K):
[in this window]
[in a new window]
  		FIG. 2. Phylogenetic relationship of the IRF protein family. A neighbor-joining tree of the entire family was constructed based on alignment of protein sequences by the computer program ClustalX. Chicken (Gg), clawed toad (Xl), human (Hs), Japanese flounder (Po), mouse (Mm), quail (Cj), rat (Rn), sheep (Oa), and torafugu (Tr) IRF sequences were analyzed. The full-length sequences of the individual IRF family members and a sequence of P. olivaceus IRF-8 with an incomplete C terminus derived from the EST database entry were included (see Table 2). The four IRF subfamilies are indicated in the margins. The bootstrap confidence values shown at the nodes of the tree are based on 1,000 bootstrap replications.

 

View this table:
[in this window]
[in a new window]
  		TABLE 2. Protein sequences used for phylogenetic tree constructiona

 
Characterization of the IRF-10 protein. The protein product of the cloned IRF-10 cDNA was translated in vitro and compared with proteins, IRF-4 and IRF-8, encoded by the two most closely related genes (Fig. 3A). The IRF-10 cDNA encoded a protein with a molecular mass of 46 kDa, which closely agrees with the calculated theoretical molecular mass. IRF-10 was also analyzed in vivo by Western blotting. The coding region of IRF-10 was inserted into a retroviral vector and expressed in the B-cell line DT40 and fibroblast cultures (Fig. 3B and C). Exogenous IRF-10 was abundantly expressed in both cell types, while no endogenous protein was detected in control cells infected with CSV alone. Retrovirus-expressed IRF-10 had a size similar to that of the endogenous protein detected in the C4-1 B-cell line transformed by the S2A3v-rel oncogene (Fig. 3B). In contrast to IRF-4, which was previously shown to have two closely migrating species, IRF-10 was expressed as a single species (39). IRF-10 also differed from IRF-4 in its subcellular localization (Fig. 3C). While most of the IRF-4 protein exogenously expressed in CEF was found in the cytoplasm, approximately two-thirds of the IRF-10 protein was located in the nucleus (39). Like that of IRF-4, the expression of IRF-10 was maintained for several weeks in fibroblasts and in the DT40 cell line. This is in contrast to fibroblast cultures or DT40 cells exogenously expressing either IRF-1 or IRF-8, where the cells expressing these IRF family members were rapidly eliminated due to their antiproliferative or apoptotic effect on the cells (data not shown).



View larger version (36K):
[in this window]
[in a new window]
  		FIG. 3. Analysis of IRF-10 protein. (A) Protein product of IRF-10 ORF translated in vitro with rabbit reticulocyte lysate. The mobility of the protein is compared with those of chicken IRF-4 and IRF-8. Molecular mass markers in kilodaltons are shown on the right. (B) Comparison of retrovirus-expressed IRF-10 in the DT40 B-cell line with endogenous IRF-10 present in C4-1 B cells transformed by S2A3v-rel. Western blot analysis of whole-cell lysates. (C) Expression of IRF-10 in CEF cultures. Western blot analysis of whole-cell lysates (W) and lysates of cytoplasmic (C) and nuclear (N) fractions. For experiments shown in both panels B and C, cells were infected either with a retrovirus expressing IRF-10 or with the CSV helper only. The lysates from 4 x 105 cells were loaded in each lane of the gel. Detection was performed with IRF-10 antiserum AI10-8. The positions of IRF-10 bands are shown on the right. Just above the IRF-10 band is a prominent background band that was present in all of the cells analyzed. (D) DNA-binding complexes in the nuclei of DT40 cells infected with a retrovirus expressing IRF-10 in comparison with complexes from control CSV-infected cells. Oligonucleotide probe containing the ISRE site from the promoter of the chicken gene encoding the {alpha} chain of the MHC class I protein was used for the analysis. Binding reaction mixtures shown in lanes 1 to 4 contained preimmune rabbit antiserum (PRE) or IRF-10-specific antiserum AI10-8 ({alpha}IRF10). Supershifted high-molecular-weight complexes (S) and complexes partially depleted by IRF-10-specific antiserum (D) are indicated. To verify that IRF-10 binding is specific, competitor oligonucleotides (COMPET) with a wild-type (WT) or mutated (MUT) ISRE site were added in 50-fold excess to the reaction mixtures containing nuclear extract from DT40 cells expressing IRF-10 (lanes 5 and 6).

 
IRF-10 was further tested in vivo for the ability to bind to the ISRE present in the {alpha} chain of the MHC class I promoter. Electrophoretic mobility shift assays revealed differences in DNA-binding complexes in the nuclei of DT40 cells expressing IRF-10 in comparison with complexes detected in the nuclei of control cells (Fig. 3D, lanes 1 and 2). The presence of IRF-10 in the DNA-binding complexes from IRF-10-expressing cells was verified with IRF-10 specific antiserum (Fig. 3D, lane 4).

Tissue distribution of IRF-10. The expression of IRF-10 in various organs and cells was determined by Northern blot analysis (Fig. 4A). The highest level of expression was detected in cells of hematopoietic origin. A very low level of IRF-10 expression was detected, however, in bursal cells. IRF-10 was also expressed in lung tissue, but it was not detected in other organs (data not shown). In order to compare the expression pattern of IRF-10 with those of other family members, parallel blots were hybridized with IRF-4, IRF-8, cIRF-3, IRF-1, and IRF-2 probes. All of the members of the IRF family examined differed in tissue distribution. With the exception of IRF-8, their expression was limited to hematopoietic cells and lung, liver, and intestine tissues. In striking contrast to IRF-10, IRF-4 was expressed at the highest level in bursae and bursal lymphocytes. The highest level of IRF-8 expression was detected in the spleen and splenic lymphocytes, followed by other hematopoietic cells and organs. Surprisingly, IRF-8 expression was also detected in most other organs, including oviduct, testis, gizzard muscle, intestine, lung, liver, and brain tissues (Fig. 4B). The IRF-1 and cIRF-3 expression pattern was most similar to that of IRF-10; however, their overall expression level in hematopoietic organs was much lower than that of IRF-10 (IRF-1 and cIRF-3 blots were exposed 3.7 times longer). IRF-2 tissue distribution was unique among the IRFs analyzed, with the highest expression in bursae and splenic lymphocytes. In conclusion, IRF-10, like IRF-4 and IRF-8, is highly expressed in organs of hematopoietic origin but, in contrast to them, its expression level in bursal cells is low.



View larger version (51K):
[in this window]
[in a new window]
  		FIG. 4. Expression pattern of chicken IRF-10, IRF-4, IRF-8, cIRF-3, IRF-1, and IRF-2 in various tissues from a 1-month-old chicken. (A) Total RNA (10 µg) isolated from bursa, bursal lymphocytes (BURSAL L.), thymus, thymic lymphocytes (THYMIC L.), spleen, splenic lymphocytes (SPLENIC L.), bone marrow, white blood cells (WBC), intestine, liver, and lung was subjected to Northern analysis. (B) Total RNA (10 µg) isolated from the various organs was hybridized with a probe against IRF-8. The blots hybridized with IRF-10, IRF-4, IRF-8, and IRF-2 were exposed for 15 h, and the blots hybridized with IRF-1 and cIRF-3 were exposed for 56 h. The specific activity of all of the probes was approximately 2 x 105 cpm/ng. The probes are described in Materials and Methods. The intensity of the rRNA stained with ethidium bromide is shown at the bottom (rRNA).

 
IRF-10 expression is inducible by IFNs and ConA. The expression of IRFs may be induced by some or all of the following stimuli: virus infection, double-stranded RNA (dsRNA), IFNs, or proliferative signals (62). Therefore, the kinetics of induction of IRF-10 expression by dsRNA, type I and II IFNs, and ConA was analyzed and compared with that of IRF-4, IRF-8, cIRF-3, IRF-1, and IRF-2 (Fig. 5; see also Fig. 6 and 7). CEF cultures were treated with dsRNA (10 µg of pI · pC per ml) with LipofectAMIN 2000, and samples for Northern blot analysis were taken at 0.5, 1, 2, 4, and 7 h after treatment (Fig. 5). Parallel cultures were treated simultaneously with dsRNA and cycloheximide, cycloheximide alone, or LipofectAMIN 2000. IRF-10 transcription was not elevated 4 and 7 h after dsRNA treatment above the background of LipofectAMIN 2000 alone (Fig. 5, lanes 5, 6, and 13 to 15). IRF-4 mRNA was not detected in fibroblasts in this experiment. The expression of IRF-8 and cIRF-3 was slightly induced 2 and 7 h after treatment and was not inhibited by cycloheximide. The expression of IRF-1 was strongly induced beginning at 2 h after treatment and was not inhibited by cycloheximide. The expression of IRF-2 did not significantly change during the experiment. The expression of c-rel increased 4 to 7 h after treatment of cells with dsRNA, and this increase was independent of cycloheximide. In conclusion, in contrast to IRF-8, IRF-1, and cIRF-3, IRF-10 did not respond to dsRNA.



View larger version (113K):
[in this window]
[in a new window]
  		FIG. 5. Induction of IRF-10, IRF-4, IRF-8, cIRF-3, IRF-1, IRF-2, and c-rel expression by dsRNA (dsR). Fibroblasts cultivated for 2 weeks after explantation from chicken embryos were treated with pI · pC (10 µg/ml) (lanes 1 to 6) by using LipofectAMIN 2000 (LF) as described in Materials and Methods. Parallel cultures were simultaneously treated with cycloheximide (CHX; 10 µg/ml) (lanes 7 to 9). The control cells were treated with cycloheximide (lanes 10 to 12) from Sigma or LipofectAMIN 2000 (lanes 13 to 15) alone. RNA was isolated from these cells at the time points indicated. Total RNA (10 µg per lane) was subjected to Northern analysis and hybridized with probes described in Materials and Methods. The intensity of the rRNA staining with ethidium bromide is shown at the bottom (rRNA).

 


View larger version (105K):
[in this window]
[in a new window]
  		FIG. 6. Induction of IRF-10, IRF-4, IRF-8, cIRF-3, IRF-1, IRF-2, and c-rel expression by IFN1 and IFN-{gamma}. Fibroblasts cultivated for 2 weeks after explantation from chicken embryos were treated with 200-fold-diluted supernatant fluids from COS-1 cells expressing IFN1 or IFN-{gamma} (IFN1, lanes 1 to 7; IFN-{gamma}, lanes 13 to 19). Parallel cultures were simultaneously treated with cycloheximide at 10 µg/ml (IFN1 + CHX, lanes 8 to 12; IFN-{gamma} + CHX, lanes 20 to 22). RNA was isolated at the time points indicated. Total RNA (10 µg per lane) was subjected to Northern analysis and hybridized with the probes described in Materials and Methods. The intensity of the rRNA stained with ethidium bromide is shown at the bottom (rRNA).

 


View larger version (71K):
[in this window]
[in a new window]
  		FIG. 7. Induction of IRF-10, IRF-4, IRF-8, cIRF-3, IRF-1, IRF-2, and c-rel expression by ConA. The spleens from 2-month-old chickens were used as a source of lymphocytes. The splenic cells were purified with Histopaque. Lymphocytes (80 x 106/ml) were treated with ConA (Sigma Chemical Co.) at 25 µg/ml (lanes 1 to 7). A parallel sample was simultaneously treated with cycloheximide at 10 µg/ml (ConA + CHX, lane 8). RNA was isolated at the time points indicated. Total RNA (10 µg per lane) was subjected to Northern analysis and hybridized with the probes described in Materials and Methods. The intensity of rRNA staining with ethidium bromide is shown at the bottom (rRNA).

 
CEF cultures were treated with IFN1 or IFN-{gamma}, and samples were taken for Northern blot analysis at the various time points indicated (Fig. 6). Parallel cultures were treated simultaneously with cycloheximide and either IFN1 or IFN-{gamma}. The increases in IRF expression induced by IFN1 were transient and peaked at 2 to 4 h after treatment (Fig. 6, lanes 4 and 5). The expression of IRF-10 was induced relatively late, at 4 h, and was dependent on protein synthesis (Fig. 6, lanes 5 and 11). IRF-4 expression was not detected after IFN1 treatment. IRF-8 was rapidly induced by IFN1 treatment after 1 h, with a peak after 2 h (Fig. 6, lanes 3 and 4). Induction of cIRF-3 expression was more rapid than that of IRF-10, but its peak at 4 h coincided with the peak of IRF-10 expression (Fig. 6, lane 5). The induction of IRF-1 expression was similar to that of IRF-8 but could be detected as early as 30 min following exposure to IFN1 (Fig. 6, lane 2). The level of IRF-2 induction was very low (Fig. 6, lanes 3 and 4). CEF cultures treated with recombinant IFN-{gamma} also increased the expression of IRF-10. The level of IRF-10 mRNA increased 2 h after treatment and reached a plateau at 4 h (Fig. 6, lanes 15 and 16). IRF-10 expression was not induced with IFN-{gamma} in the presence of cycloheximide (Fig. 6, lanes 15, 16, 20, and 21). As expected, IRF-4 expression was not induced by IFN-{gamma}. However, the expression of IRF-8 and IRF-1 was strongly and rapidly induced by IFN-{gamma} starting as early as 30 min after treatment (Fig. 6, lane 13). Expression of cIRF-3 began 1 h after treatment, and its level remained relatively low (Fig. 6, lane 14). The induction of IRF-2 expression was very low in level and delayed (Fig. 6, lane 16). The expression of c-rel did not change after treatment with either of the two IFNs. With the exception of IRF-10, the induction of all of the other IRFs with IFNs was independent of protein synthesis.

Induction of IRF-10 and other IRF family members by ConA was analyzed in splenic cells by Northern blot analysis at different times after treatment (Fig. 7). Parallel cultures were treated with ConA and cycloheximide, and RNA was isolated 4 h after treatment. Exposure of splenic lymphocytes to ConA led to induction of IRF-10, IRF-4, IRF-1, and cIRF-3, which continued during the 21-h time frame evaluated. IRF-10 expression was induced at 4 h after exposure of splenic cells to ConA and was not observed in the presence of cycloheximide (Fig. 7, lanes 5 and 8). In contrast, induction of IRF-4 expression started earlier (2 h after treatment), immediately reached a plateau, and was cycloheximide independent (Fig. 7, lanes 4 and 8). Surprisingly, IRF-8 expression was unresponsive to ConA treatment. cIRF-3 expression was induced by ConA at 2 h (Fig. 7, lane 4). IRF-1 expression was strongly induced by ConA 2 h after treatment, while IRF-2 expression was not changed. Expression of c-rel transiently increased between 2 and 4 h after ConA treatment (Fig. 7, lanes 4 and 5). With the exception of that of IRF-10 and IRF-1, induction was independent of protein synthesis. Another distinct feature of IRF-10 ConA-mediated induction of expression was the continued increase at the later time points.

These experiments demonstrated that IRF-10, in contrast to IRF-4, is induced by both types of IFNs in primary fibroblasts. IRF-10, like IRF-4, is also induced by ConA in splenic cultures. The kinetics of induction of IRF-10 by IFNs and ConA is delayed relative to that of other IRFs and depends on protein synthesis.

IRF-10 modulates the induction of the IFN-induced genes for GBP and OAS. IRF family members play an important role in antiviral responses; therefore, the ability of IRF-10 to induce some IFN-responsive genes was evaluated. Fibroblast cultures expressing high levels of IRF-10 from retroviral vectors were established (Fig. 3C). The induction of several IFN-responsive genes, of which GBP, OAS, and Mx are the best characterized, is regulated, in part, by IRF family members (9, 64, 90). To characterize the IFN induction pattern of these genes in IRF-10-expressing cells, control and IRF-10-expressing cultures were treated with IFN1 and the kinetics of induction of these IFN target genes was evaluated by Northern blot analyses (Fig. 8). Overexpression of IRF-10 resulted in a slight increase in GBP expression even without IFN treatment, and IFN1-mediated induction of GBP in these cells was stronger than in control cells and was prolonged over a 24-h period. Surprisingly, IFN1-mediated induction of OAS was weaker and briefer in IRF-10-expressing cells than in control cells. Mx induction by IFN1 was unaffected in cells overexpressing IRF-10 in comparison to normal CEF (data not shown).



View larger version (30K):
[in this window]
[in a new window]
  		FIG. 8. IRF-10 modulates the expression of GBP- and OAS-encoding genes involved in an antiviral response. CEF expressing IRF-10 or control cells expressing the helper virus (CSV) were treated with 1,000-fold-diluted supernatant fluids from COS-1 cells expressing IFN1. RNA was isolated at the time points indicated. Total RNA (10 µg per lane) was subjected to Northern analysis with the indicated probes (described in Materials and Methods). The radioactive signal on Northern blots was quantified by a Phosphorimager 225SI (Molecular Dynamics, Sunnyvale, Calif.), and the intensities of the hybridizing bands are shown in relative phosphorimager units (RPU).

 
IRF-10 contributes to the regulation of expression of MHC class I. IRF-10 is unique among IRFs in its delayed induction and dependency on protein synthesis, suggesting that it may play a role in the late stages of an antiviral defense. The purpose of the delayed defense is to present viral antigens to stimulate a specific immune response, which can ultimately recognize and eliminate infected cells. Viral antigens are presented together with MHC class I and II molecules on antigen-presenting cells. The regulation of MHC class I is partially under the control of IFNs, and some IRF family members have been shown to mediate this antiviral action at the promoter level (13, 38, 104). Therefore, we investigated whether IRF-10 also participates in the regulation of MHC class I expression.

The expression of MHC class I and IRF-10 in fibroblasts after treatment with IFNs was analyzed (Fig. 9). IFN1 and IFN-{gamma} induced the expression of MHC class I transcription beginning at 5 h after treatment, and the steady-state level of RNA increased at the later time points (Fig. 9, lanes 3 to 6). At that time, endogenous IRF-10 already reached its highest levels (Fig. 9, lane 3). To determine if IRF-10 may mediate the induction of MHC class I by IFNs, the expression of MHC class I in IRF-10-overexpressing cells before and after treatment with IFN1 and IFN-{gamma} was analyzed (Fig. 9, lanes 7 to 12). When IRF-10 was expressed in fibroblasts exogenously, the levels of MHC class I were significantly elevated (Fig. 9, lane 7). Moreover, in these cells, the expression of MHC class I was further enhanced by treatment with IFN1, suggesting that IRF-10 contributes to MHC class I regulation in fibroblasts (Fig. 9, lanes 9 to 12). MHC class II expression was not detected in control or IRF-10-expressing fibroblasts before or after treatment with IFNs (data not shown). Taken together, these results suggest that IRF-10 contributes to the regulation of MHC class I in the response to IFNs.



View larger version (102K):
[in this window]
[in a new window]
  		FIG. 9. IRF-10 modulates the expression of MHC class I. CEF cultures expressing IRF-10 or control cells expressing helper virus (CSV) were treated with 1,000-fold-diluted supernatant fluids from COS-1 cells expressing IFN1 or IFN-{gamma}. RNA was isolated at the time points indicated. Total RNA (10 µg) was subjected to Northern analysis with MHC class I and IRF-10 probes. The sample from CSV-infected, IFN-{gamma}-treated CEF cultures (bottom, lane 5) exhibits a defect when hybridized with an MHC class I probe. Northern blots hybridized with the IRF-10 probe show endogenous IRF-10 mRNA (1.8 to 1.9 kb) in CSV-infected CEF cultures (lanes 1 to 6), while two retroviral REV-IRF-10 RNAs (approximately 6 and 3 kb) are shown in blots from REV-IRF-10-infected CEF cultures (lanes 7 to 12). The hybridized blots showing endogenous IRF-10 mRNA were exposed to film six times longer than the blots with REV-IRF-10 RNA. The intensity of rRNA staining with ethidium bromide is shown below for each Northern blot (rRNA).

 
Expression of IRF-10 in transformed cell lines. Several transformed cell lines were analyzed for expression of IRF-10 mRNA (Fig. 10A). Bursal B-cell lines DT40 and DT95 failed to express IRF-10. The expression pattern in T cells transformed by Marek's disease virus, RP-1 and MSB-1, was variable. IRF-10 was strongly expressed in RP-1, but its expression in MSB-1 cells was very weak. IRF-10 expression correlated with the expression of MHC class II in these cell lines; MSB-1 cells express very low levels of MHC class II, while RP-1 express high levels (79; data not shown), suggesting that IRF-10 may contribute to regulation of the expression of the gene for MHC class II. Strong expression of IRF-10 was also detected in the BM-2 macrophage cell line and an AEV-1 erythroblastoid cell line.



View larger version (56K):
[in this window]
[in a new window]
  		FIG. 10. Constitutive and induced expression of IRF-10 mRNA in transformed cell lines and CEF. Total RNA (10 µg per lane) was subjected to Northern analysis and hybridized with IRF-10, cIRF-3, and c-rel probes. The intensity of rRNA staining with ethidium bromide is shown at the bottom (rRNA). (A) Constitutive IRF-10 expression in B-cell lines DT40 and DT95, T-cell lines MSB-1 and RP-1, myeloblastoid cell line BM-2, erythroblastoid cell line AEV-1, and v-rel-transformed cell lines with the B-cell (123/12), T-cell (160/2), and non-B-, non-T-cell phenotypes (123/6T) was analyzed. (B) Constitutive expression of IRF-10 in bursal lymphocytes (BL) and v-rel-transformed cell lines of bursal origin (TWB5, TWB2, and TWB4). The expression of v-rel from a retroviral vector is shown in the middle. Endogenous c-rel mRNA is not visible at this exposure. (C) Induction of IRF-10 expression in DT40 and DT95 B-cell lines by v-Rel, c-Rel, and c-Rel{Delta}Xba and comparison with a CSV control. The expression of rel genes from retroviral vectors is shown in the middle. Endogenous c-rel mRNA is at the detection threshold at this exposure and is visible in lanes 1, 4, and 6. (D) Cooperation of IRF-4 with v-Rel in the induction of IRF-10 and cIRF-3 in CEF. Secondary cultures of CEF were infected with REV-IRF-4 (IRF-4), REV-IRF-4{Delta}E6 (IRF-4{Delta}E6), or DSv-Rel (v-Rel) or coinfected with DSv-Rel and REV-IRF-4 or REV-IRF-4{Delta}E6 at a multiplicity of infection of 3. Control cells were left uninfected or infected with CSV. The expression of IRF-4 and v-Rel proteins in these fibroblasts was confirmed by Western blot analysis at 3 weeks after infection as described previously (39), and the expression of IRF-10 and cIRF-3 mRNA was then analyzed by Northern blot analysis as described in the legend to Fig. 4.

 
The highest level of IRF-10 expression was detected in v-rel-transformed cell lines of splenic origin, irrespective of their phenotype; IgM-expressing line 123/12, TCR2-expressing 160/2, and 123/6T, which expresses macrophage markers CMTD I and II, all had high levels of IRF-10 mRNA (42). Because IRF-10 levels were low in bursal lymphocytes and were not detected in bursal ALV-derived cell lines DT40 and DT95, it was of interest to determine whether IRF-10 is expressed in v-rel-transformed cell lines of bursal origin. The expression of IRF-10 in three v-rel-transformed cell lines derived from the bursa was examined (Fig. 10B). In contrast to bursal lymphocytes, all of these cell lines expressed high levels of IRF-10 mRNA. In summary, IRF-10 is expressed in a variety of hematopoietic transformed cell lines but the highest expression level was detected in v-rel-transformed cell lines.

v-Rel and c-Rel induce IRF-10 expression in transformed lymphoid cell lines. The high level of IRF-10 expression in v-rel-transformed cell lines, especially in bursal cell lines, suggested that IRF-10 may be regulated, in part, by Rel/NF-{kappa}B transcription factors. Therefore, the expression of IRF-10 in the DT40 and DT95 cell lines and primary CEF expressing v-Rel or c-Rel was compared to the expression of IRF-10 in parental cells (Fig. 10C and D). The expression of IRF-10 was strongly elevated in the DT40 cell line expressing v-Rel in comparison to that in the parental cell line, but it was not detected in DT40 cells expressing c-Rel (Fig. 10C). However, endogenous c-Rel, unless activated by an external signal, resides predominantly in the cytoplasm (41). Deletion of the cytoplasmic anchoring sequence located in the C terminus in mutant c-Rel proteins allows nuclear access in the absence of an appropriate external signal. These mutant proteins would be expected to be better transcriptional activators than wild type c-Rel (75). Therefore, c-Rel{Delta}Xba, a transforming mutant form of c-Rel lacking the C terminus, was evaluated for induction of IRF-10 in DT40 cells. While c-Rel{Delta}Xba induced high levels of IRF-10, this level did not reach the level induced by v-Rel even though c-Rel{Delta}Xba was more abundant in cells than was v-Rel. Similarly, in the DT95 cell line expressing v-Rel, the level of IRF-10 was higher than in c-Rel-expressing DT95 cells although the level of v-Rel was lower than that of c-Rel in these cells (Fig. 10C). Surprisingly, the expression of IRF-10 was not elevated in fibroblast cultures overexpressing v-Rel or c-Rel (Fig. 10D, data not shown). These results indicate that IRF-10 expression is inducible by v-Rel and a to lesser degree by c-Rel in lymphoid cells.

IRF-4 induction of IRF-10 is enhanced by v-Rel. The failure of v-Rel to induce IRF-10 in fibroblasts suggests that some cell-specific transcription factor(s) necessary for IRF-10 induction may be absent in these cells. Fibroblasts and the B-cell lines DT40 and DT95 differ in the level of expression of IRF-4. The endogenous level of IRF-4 mRNA in fibroblasts is very low; by contrast, the levels of IRF-4 in DT40 and DT95 are high (39). To test the hypothesis that the difference in the levels of IRF-4 is responsible for the tissue-specific induction of IRF-10 expression by v-Rel, fibroblasts were infected with a retrovirus expressing v-Rel, IRF-4, or IRF-4 splice variant IRF-4{Delta}E6 or coinfected with retroviruses expressing v-Rel and IRF-4 or v-Rel and IRF-4{Delta}E6 (Fig. 10D). In contrast to v-Rel, IRF-4 alone induced the expression of IRF-10 in fibroblasts. However, when v-Rel was coexpressed with IRF-4 or IRF-4{Delta}E6 in fibroblasts, expression of IRF-10 mRNA was highly elevated, suggesting that IRF-4 and v-Rel cooperate in the induction of IRF-10. Similar cooperation between v-Rel and IRF-4 was observed in induction of the expression of cIRF-3 but not with other IRF family members (Fig. 10D; data not shown).


   DISCUSSION
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Evolution of the IRF family. This report describes the characterization of a novel member of the IRF family, IRF-10. IRF-encoding genes are known only in vertebrates. All IRF family members are characterized by a DBD with similarly spaced tryptophan residues, as in the DBD of the myb transcription factors (99). The myb family functions in proliferation, and myb homologues have been identified in both vertebrates and invertebrates (59). The relatively recent evolution of the IRF family, therefore, likely occurred from a prototypical protein with a myb-like DNA-binding motif. All IRFs, except the members of the IRF-1 subfamily, also contain an IAD protein-protein interaction domain, which has a high level of similarity to the transactivation domain of Smad morphogens (19). The acquisition of this protein-protein interaction module of the Smad family of transcription factors resulted in further branching of the IRF network.

The phylogenetic analysis presented in this report suggests that the family of IRF transcription factors appeared early in vertebrate evolution and was followed by rapid diversification since all four IRF subfamilies are present in Teleostei fish. The majority of the IRF family members identified in nonmammalian vertebrates are orthologues of the mammalian IRFs. However, the fish IRF-1 subfamily members (P. olivaceus IRF and Takifugu rubipes IRF) and avian cIRF-3 differ in sequence to such an extent that it is difficult to consider them orthologues of any specific mammalian IRF (29, 82, 106). A detailed analysis of the sequences of P. olivaceus IRF and T. rubipes IRF suggests greater similarity to mammalian IRF-1 than to IRF-2. Moreover, a fish IRF with strong homology to IRF-2 has been recently identified (GenBank accession number AF146733.1). Therefore, it is unlikely that P. olivaceus IRF and T. rubipes IRF is a common precursor of IRF-1 and IRF-2, suggesting that the difference between P. olivaceus IRF or T. rubipes IRF and mammalian IRF-1 is the result of extensive diversification. The situation is different with cIRF-3. The gene with strong homology to avian cIRF-3 has been identified in fish (GenBank accession number BE605317.1), but a cIRF-3 homologue is not present in the human genome. By contrast, the orthologues of mammalian IRFs most closely related to cIRF-3 (IRF-3 and IRF-7) have not been identified in nonmammalian vertebrates (data not shown). Therefore, it is plausible that IRF-3 and IRF-7 evolved from a cIRF-3-like precursor by duplication and subsequent diversification. The appearance of most of the IRF family members in lower vertebrates correlates with the appearance of adaptive immunity in these species. However, the evolution by duplication and diversification further modified the IRF family members in individual vertebrate branches.

IRF-10 evolved as one of the last IRF family members from a precursor it has in common with IRF-4. Comparison of the lengths of the branches of the IRF phylogenetic tree suggests that unless the IRF-10 diversification rate was substantially higher than for other IRF family members, IRF-10 evolved in the ancestors of Sauropsida and mammals. Its origin may even predate the divergence of Teleostei fish and tetrapods. Although we were unable to identify an IRF-10 protein in fish, we have identified a tentative P. olivaceus IRF-4 orthologue (dbj|AU091159.1), suggesting that IRF-10 may exist in recently evolved fish. More confusing is the fate of mammalian IRF-10. Last year, sequences from two almost complete human genome projects became available (56, 100). All nine human IRF family members can be identified in both databases. Only one additional IRF-like sequence was found close to the centromeric region of chromosome 20 (hypothetical protein products of this gene have GenBank accession numbers CAC28981 and CAC28982). This sequence has the greatest homology to IRF-10, but the locus diverged to such an extent that it cannot produce a full-length protein. Accordingly, we were not able to detect mRNA (in human spleen or in HeLa cells in response to human IFN-{gamma}) with probes or primers representing the region, which contains the DBD in other IRF family members (data not shown). Recently, a cDNA clone from a carcinoma cell line was identified that contains mRNA formed by splicing of the last three exons of the putative human IRF-10-encoding locus (GenBank accession number AK000652). However, the predicted ORF is incomplete and contains at least one interruption; therefore, it is unlikely that a functional IRF protein could be translated from that mRNA. Our current hypothesis is that IRF-10 is not expressed in the more recently evolved mammals. We hypothesize that elimination of IRF-10 expression in mammals is connected with the branching of cIRF-3 into two closely related but functionally different IRF-3 and IRF-7 family members. IRF-3 is considered to be the first IRF to function in response to viral infection whereby the protein located in the cytoplasm translocates to the nucleus and activates the expression of IFN-ß (62). Subsequently, the transcriptional activation of IRF-7 by IFN-ß serves to amplify the expression of IFN-ß and activates the IFN-{alpha} genes in lymphoid cells (107). This two-step induction model may be similar to the avian system, with IRF-10 employed in the delayed phase.

IRF-10 has a unique function in the immune response. IRF-10 differs in constitutive and inducible expression from its closest relatives, IRF-4 and IRF-8, suggesting a different biological function. In contrast to that of IRF-4, expression of IRF-10 is induced by IFN type I. IFN type I-inducible IRFs, like IRF-1, play a role in the protection of cells against most viruses by regulating expression of IFN target genes (84). The IFN target gene that encodes GBP is regulated by mammalian IRF-1, but overexpression of avian IRF-1 has no effect on GBP levels (9, 111). Therefore, it is possible that IRF-10, which enhanced the induction of GBP by IFN1, may compensate for avian IRF-1. IRF-1 is also involved in the regulation of MHC class I expression in both avian and mammalian cells (38, 111). Similarly, IRF-10 is able to induce the expression and enhance the induction of MHC class I in fibroblasts exposed to type I IFN. The genes for GBP and MHC class I are induced by both types of IFNs, which is in contrast to the genes for OAS and Mx, which are only induced by type I IFN (15). It is interesting that OAS expression is reduced after treatment of IRF-10-espressing cells with IFN1. The ability of IRF-10 to regulate the expression of GBP and MHC class I and interfere with OAS expression suggests that IRF-10 may selectively regulate type II over type I IFN target genes.

The role of IRF-10 in the induction of GBP and MHC class I resembles the function of mammalian IRF-1. There are, however, two major differences between IRF-10 and IRF-1. In contrast to IRF-10, IRF-1 is the first IRF induced in response to the activators whereas IRF-10 has a very delayed kinetics of induction. The other difference is in their effects on cell proliferation. Both fibroblasts and lymphoid B-cell lines support the propagation of retrovirus overexpressing IRF-10 but not IRF-1. IRF-1 is known to induce a cytotoxic effect in several cell types and upregulates the expression of proapoptotic genes (84). In conclusion, IRF-10 has a unique role in regulating the switch from an innate to an adaptive immune response against virus without an antiproliferative effect.

The Rel/NF-{kappa}B and IRF families regulate IRF-10 expression. v-Rel, c-Rel, and IRF-4 regulate IRF-10 expression in a cell-specific manner. The expression of other avian IRFs, including IRF-4, IRF-8, cIRF-3, and IRF-1, is also induced by v-Rel and c-Rel (39, 40). Several lines of evidence indicate that IRF-encoding genes are direct transcriptional targets of Rel/NF-{kappa}B and IRF proteins. Rel/NF-{kappa}B factors regulate IRF-4 and IRF-1 at the promoter level (32, 44). The cIRF-3 promoter contains {kappa}B sites responsible for dsRNA inducibility (66). Furthermore, both the avian and mammalian IRF-8 promoters contain a {kappa}B site and the {kappa}B site in the mammalian promoter is bound by Rel/NF-{kappa}B factors (16, 50). Similarly, the regulation of several IRFs by IFNs is mediated directly through IRF-binding sites in their promoters (16, 35, 60, 66). Therefore, it is likely that IRF-10 regulation by v-Rel, c-Rel, and IRF-4 also takes place at the promoter level.

IRF-10 expression was strongly induced by v-Rel in lymphoid cells. However, wild type c-Rel induced IRF-10 expression considerably less efficiently than did v-Rel. c-Rel is a critical Rel/NF-{kappa}B transcriptional regulator in B cells (33). Therefore, the low efficiency with which c-Rel induces the expression of IRF-10 may be one reason why IRF-10 is present at such low levels in bursal cells. v-Rel is a highly mutated form of c-Rel (27). v-Rel has a C-terminal deletion that removes a cytoplasmic retention domain, allowing nuclear translocation in the absence of an appropriate exogenous signal (12). By contrast, c-Rel is retained in the cytoplasm until cells receive appropriate signals. A c-Rel C-terminal deletion mutant protein that is expressed predominantly in the nucleus induces IRF-10 expression to significantly higher levels than does wild-type c-Rel. Therefore, c-Rel, if activated and translocated to the nucleus, is also capable of efficiently inducing IRF-10 expression. However, v-Rel apparently is still a more efficient inducer of IRF-10 expression than C-terminally truncated c-Rel. This is likely due the difference in the DNA-binding specificities of c-Rel and v-Rel, which, in turn, also contribute to their different transcriptional activities (74). Induction of IRF-10 expression by c-Rel, its C-terminally deleted mutant, and v-Rel correlates with their transformation potential, suggesting a possible role of IRF-10 in v-Rel transformation (39).

The IRF and Rel/NF-{kappa}B families participate in regulation of the immune response, and certain members of these two families cooperate in the regulation of the transcription of some genes (24, 85). v-Rel failed to induce the expression of IRF-10 in fibroblasts, but when IRF-4 or IRF-4{Delta}E6 was expressed at high levels together with v-Rel, expression of IRF-10 was induced in a synergistic manner. Synergism between the IRF-4 lymphocyte-specific factors and Rel/NF-{kappa}B family members may explain why IRF-10 is predominantly expressed constitutively in hematopoietic cells. The same synergism was also observed in the regulation of cIRF-3 by these factors.

In summary, a novel IRF family member has been identified and characterized. Although it is most closely related to IRF-4 and IRF-8, IRF-10 has a distinct constitutive and inducible expression pattern. Unlike that of other IRF family members, the induction of IRF-10 is delayed and requires protein synthesis. IRF-10 functions in the later stages of an antiviral defense by regulating IFN-{gamma} target genes.


   ACKNOWLEDGMENTS
 
We are grateful to many colleagues for providing chicken cDNA clones. We thank to Robin Morgan (University of Delaware, Newark) for cDNA clone ptr1c.pk002.b9, which was isolated as a part of the University of Delaware chicken EST project. We thank Igor Dawid (National Institutes of Health) for IRF-1 and IRF-8, Roger Deeley and Caroline Grant (Queen's University, Kingston, Ontario, Canada) for cIRF-3, Thomas Gilmore (Boston University, Boston, Mass.) for c-rel, Christoph Jungwirth (Universität Würzburg, Würzburg, Germany) for IRF-2, Christine Sick and Peter Staeheli (Universität Freiburg, Freiburg, Germany) for IFN1 and IFN-{gamma}, Kirsten Weining (Universität Freiburg) for Mx and GBP, and Y. Xu (Iowa State University, Ames) for MHC class II ß-chain cDNA clones. We thank Emin Ulug (University of Texas, Austin) for the kind gift of VSV.

This study was supported by Public Health Service grant CA33192 from the National Cancer Institute.


   FOOTNOTES
 
* Corresponding author. Mailing address: Section of Molecular Genetics and Microbiology, The University of Texas at Austin, Austin, TX 78712-1095. Phone: (512) 471-5525. Fax: (512) 471-2130. E-mail: bose{at}mail.utexas.edu. Back


   REFERENCES
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 
1. Akiyama, Y., and S. Kato. 1974. Two cell lines from lymphomas of Marek's disease. Biken J. 17:105-116.[Medline]

2. Altschul, S. F., W. Gish, W. Miller, E. W. Myers, and D. J. Lipman. 1990. Basic local alignment search tool. J. Mol. Biol. 215:403-410.[CrossRef][Medline]

3. Altschul, S. F., T. L. Madden, A. A. Schaffer, J. Zhang, Z. Zhang, W. Miller, and D. J. Lipman. 1997. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res. 25:3389-3402.[Abstract/Free Full Text]

4. Au, W. C., P. A. Moore, W. Lowther, Y. T. Juang, and P. M. Pitha. 1995. Identification of a member of the interferon regulatory factor family that binds to the interferon-stimulated response element and activates expression of interferon-induced genes. Proc. Natl. Acad. Sci. USA 92:11657-11661.[Abstract/Free Full Text]

5. Baba, T. W., B. P. Giroir, and E. H. Humphries. 1985. Cell lines derived from avian lymphomas exhibit two distinct phenotypes. Virology 144:139-151.[CrossRef][Medline]

6. Bernasconi, D., U. Schultz, and P. Staeheli. 1995. The interferon-induced Mx protein of chickens lacks antiviral activity. J. Interferon Cytokine Res. 15:47-53.

7. Boehm, U., T. Klamp, M. Groot, and J. C. Howard. 1997. Cellular responses to interferon-{gamma}. Annu. Rev. Immunol. 15:749-795.[CrossRef][Medline]

8. Bose, H. R., Jr. 1992. The Rel family: models for transcriptional regulation and oncogenic transformation. Biochim. Biophys. Acta 1114:1-17.[Medline]

9. Briken, V., H. Ruffner, U. Schultz, A. Schwarz, L. F. Reis, I. Strehlow, T. Decker, and P. Staeheli. 1995. Interferon regulatory factor 1 is required for mouse Gbp gene activation by gamma interferon. Mol. Cell. Biol. 15:975-985.[Abstract]

10. Burysek, L., W.-S. Yeow, B. Lubyová, M. Kellum, S. L. Schafer, Y. Q. Huang, and P. M. Pitha. 1999. Functional analysis of human herpesvirus 8-encoded viral interferon regulatory factor 1 and its association with cellular interferon regulatory factors and p300. J. Virol. 73:7334-7342.[Abstract/Free Full Text]

11. Bushdid, P. B., D. M. Brantley, F. E. Yull, G. L. Blaeuer, L. H. Hoffman, L. Niswander, and L. D. Kerr. 1998. Inhibition of NF-{kappa}B activity results in disruption of the apical ectodermal ridge and aberrant limb morphogenesis. Nature 392:615-618.