Ubiquitin-dependent Degradation of Interferon Regulatory Factor-8 Mediated by Cbl Down-regulates Interleukin-12 Expression*

Interferon regulatory factor (IRF)-8/interferon consensus sequence-binding protein is regulated by both transcription and degradation. IRF-8 induced in peritoneal macrophages by interferon-γ and lipopolysaccharide was degraded rapidly, and degradation of IRF-8 was blocked by MG132, the proteasome inhibitor, but inhibitors of calpain and lysosomal enzymes had no effect. The ubiquitination of IRF-8 was shown by co-immunoprecipitation from RAW264.7 macrophages retrovirally transduced with IRF-8 and hemagglutinin-ubiquitin. The dominant negative ubiquitin mutants K48R and K29R inhibited IRF-8 degradation in 293T cells, confirming the relationship between ubiquitination of IRF-8 and its degradation. IRF-8 carboxyl-terminal truncation mutants were not ubiquitinated and were consequently stable, indicating that the carboxyl-terminal domain of IRF-8 controls ubiquitination. The ubiquitin-protein isopeptide ligase (E3) that ubiquitinated IRF-8 was likely to be Cbl, which formed a complex with IRF-8, demonstrable by both immunoprecipitation and gel filtration. Furthermore, IRF-8 stability was increased by dominant negative Cbl, and IRF-8 ubiquitination was significantly attenuated in Cbl–/– cells. Reflecting increased stability and expression, the IRF-8 carboxyl-terminal deletion mutant induced interleukin (IL)-12 p40 promoter activity much more strongly than IRF-8 did. Furthermore, IRF-8-induced IL-12 p40 synthesis in RAW264.7 cells was enhanced by dominant negative Cbl, and peritoneal macrophages from Cbl–/– mice showed increased IL-12 p40 protein production. Taken together, these results suggest that the proteasomal degradation of IRF-8 mediated by the ubiquitin E3 ligase Cbl down-regulates IL-12 expression.

IRF-8 was originally identified as a negative regulator of promoters containing the IFN-␥-stimulated response element (6,15,16). IRF-8 inhibits IRF-1-mediated transcriptional activation and acts as co-repressor with IRF-2. The domain responsible for IRF-8 repression has been mapped to the carboxyl terminus (17). However, IRF-8 can also activate gene expression required for the development and function of macrophages (5,11,18). IRF-8 and IRF-1 synergistically activate the expression of IL-12 (18), which is a pivotal cytokine driving the immune system toward a T help type (Th) 1 response (19). Both IRF-8-and IRF-1-deficient mice are highly susceptible to several pathogens, including Listeria monocytogenes, Yersinia enterocolitica, and Leishmania major (20 -22), due to the inability of these mice to generate a Th1 cell-mediated response directed against intracellular pathogens. Macrophages from these mice do not produce IL-12 in response to IFN-␥ and LPS. These results indicate the essential nature of IRF-8 and IRF-1 for the generation of a Th1 immune response.
It has been clearly demonstrated that an unregulated Th1 immune response and subsequent overproduction of Th1 cytokines lead to autoimmune diseases such as inflammatory bowel disease, autoimmune arthritis, experimental autoimmune encephalomyelitis, and type I diabetes (23)(24)(25)(26)(27). There is thus intense interest in understanding the regulation of the Th1 immune response, with the goal of developing therapeutic strategies. The regulated expression of IRF-8 and IRF-1 controls the character of immune responses, and the expression * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. level of a transcription factor reflects the balance between synthesis and degradation. Although much is known about the mechanism regulating the synthesis of IRF-8 and IRF-1, little is known about the mechanism of degradation of IRF-8 and IRF-1 and its role in regulating IL-12 expression. In the present study we report that ubiquitination of IRF-8 by the E3 ligase Cbl targets IRF-8 for degradation, which results in down-regulation of IL-12 expression.
Plasmid Constructs-Full-length IRF-8 cDNA and IRF-8 carboxylterminal deletion mutants were inserted into the mammalian expression vector pcDNA3.1. The IRF-1 expression plasmid was a gift from T. Taniguchi (University of Tokyo, Tokyo, Japan). The HA-ubiquitin plasmid and HA-ubiquitin mutant plasmids K48R, K29R, and K63R were kindly provided by Dr. Ze'ev Ronai (Mount Sinai School of Medicine). The Cbl dominant negative plasmid was a gift from Dr. Reuben Siraganian (National Institutes of Health).
Cell Lines-The RAW 264.7 murine macrophage cell line (American Type Culture Collection) was maintained in DMEM supplemented with 10% fetal bovine serum and antibiotics. 293T cells were maintained in DMEM supplemented with 10% fetal bovine serum and antibiotics. Cblϩ/ϩ and CblϪ/Ϫ fibroblast cells were obtained from Dr. H. Band (Harvard Medical School) and maintained in DMEM with 10% fetal bovine serum and antibiotics.
IL-12 p40 Promoter Assay-RAW264.7 cells were transiently transfected using Superfect (Qiagen). For each transfection, 2.5 g of plasmid was mixed with 100 l of DMEM (without serum and antibiotics) and 10 l of Superfect reagent. After 10 min at room temperature, 600 l of DMEM complete medium was added, and the mixture was added to cells in 6-well plates. Luciferase activity was measured 16 -24 h later. When indicated, murine IFN-␥ (10 ng/ml) or LPS (1 g/ml) was added to the culture for 6 -12 h before harvest. The cells were harvested with reporter lysis buffer (Promega), and 20 l of extract was assayed for luciferase as described previously (11). Cells were co-transfected with a constitutively active cytomegalovirus promoter-␤-galactosidase reporter plasmid to normalize experiments for transfection efficiency.
Retroviral Transduction-For retroviral transduction we used a derivative of the Moloney murine leukemia virus vector pMMP412 (29). To prepare pseudotyped virus, human 293 EbnaT cells were seeded at a density of 4 ϫ 10 6 cells in a 10-cm dish. The next day, cells were transfected using DNA-calcium phosphate precipitate containing 2.5 g of plasmid pMD.G encoding vesicular stomatitis virus G protein, 7.5 g of plasmid pMD.OGP encoding gag-pol, and 10 g of the retroviral expression construct. 48 h after transfection, the viral supernatant was collected, centrifuged at 800 ϫ g, and used to infect cells. RAW264.7 cells (5 ϫ 10 6 ) were resuspended in 12 ml of viral supernatant in the presence of Polybrene (4 g/ml), aliquoted into 6-well plates, and spun at 800 ϫ g in a microtiter rotor at room temperature for 1 h.
Co-immunoprecipitation-Cells were washed with cold phosphatebuffered saline and lysed for 15 min on ice in 0.5 ml of lysis buffer (50 mM Tris-HCl, pH 8.0, 280 mM NaCl, 0.5% Nonidet P-40, 0.2 mM EDTA, 2 mM EGTA, 10% glycerol, and 1 mM dithiothreitol) containing protease inhibitors. Cell lysates were clarified by centrifugation (4°C, 15 min, 14,000 rpm), aliquots (500 g) were incubated with 5 g of normal goat IgG for 4 h, and IgG was collected with 20 l of protein G-Sepharose for 2 h. After centrifugation, the supernatant was collected and incubated with 2 g of IRF-8 antibody (Santa Cruz) overnight at 4°C with gentle rocking, after which immune complexes were collected as described before with 20 l of protein G-Sepharose. After washing five times with lysis buffer, immunoblotting was performed as described.

IRF-8 in Peritoneal Macrophages
Is Degraded by the Proteasome-We have shown previously that IFN-␥ and LPS synergistically induce IRF-8 protein expression in macrophages (11), and we now examine the stability of IRF-8. Thioglycollateelicited macrophages from C3H/OUJ mice activated for 6 h with IFN-␥ (10 ng/ml) and LPS (1 g/ml) were incubated with fresh medium in the presence or absence of the proteasome inhibitor MG132 (10 M), and at various time intervals cell lysates were assayed by immunoblotting for IRF-8. The expression of IRF-8 decreased with a t1 ⁄2 between 2 and 5 h, but in the presence of proteasome inhibitor MG132, IRF-8 became stable (Fig. 1A). Equivalent results were obtained with cells pretreated with cycloheximide (10 M) for 30 min (Fig. 1B). We have also examined two other members of the IRF family, IRF-1 and IRF-4. IRF-1 was degraded in a pattern similar to that of IRF-8, and the degradation was again blocked by MG132 (Fig. 1B). However, IRF-4, which is constitutively expressed in macrophages, was stable (Fig. 1B). Inhibitors of calpains and lysosomal enzymes had no effect on IRF-8 degradation (Fig. 2). Notably, lactacystin, a more specific proteasome inhibitor than MG132, blocked degradation of IRF-8 to the same extent as MG132 (Fig. 2). These results indicate that the proteasome is involved in the degradation of IRF-8 and IRF-1.
Ubiquitination of IRF-8 in RAW264.7 Murine Macrophages-We next addressed the pathway leading to proteasomal degradation. Ubiquitin has diverse cellular functions, including marking proteins for degradation by the proteasome. As a model for peritoneal macrophages, we used the macrophage cell line RAW264.7, in which IRF-8 is also degraded rapidly (Fig. 3A). Cell lysates from IFN-␥and LPS-activated RAW264.7 cells were immunoprecipitated with IRF-8 antibody and immunoblotted with anti-ubiquitin antibody. A high molecular mass smear typical of multi-ubiquitinated proteins was detected, which was increased following incubation with MG132 (Fig. 3B). In contrast, IRF-4, which is stable, was not ubiquitinated. To confirm the ubiquitination of IRF-8 in RAW264.7 cells, 48 h after co-transduction with retroviruses encoding IRF-8 and HA-ubiquitin, cell lysates were immunoprecipitated with anti-IRF-8 antibody and immunoblotted with anti-HA antibody. Ubiquitinated protein was present only in samples in which IRF-8 was expressed (Fig. 3C). These data indicate that both endogenously and exogenously expressed IRF-8 in RAW264.7 cells are subject to ubiquitination.
The Carboxyl-terminal Region of IRF-8 Controls Degradation-The stability of IRF-1 is controlled by carboxyl-terminal domain (30). The homology between IRF-1 and IRF-8 carboxylterminal domains suggests that the stability of IRF-8 might also be controlled by the carboxyl terminus. To examine this, we first had to show that IRF-8 degraded in 293T cells comparably to macrophages, which was found to be the case (Fig. 4A), and IRF-4 was stable in both cell types (Fig. 4A). Ubiquitina-tion of IRF-8 in 293T cells transfected with IRF-8 and HAubiquitin was then analyzed by immunoprecipitation with anti-IRF-8 antibody and immunoblotting with anti-HA mAb. A high molecular mass smear typical of ubiquitinated protein (Fig. 3B) was detected only in samples co-transfected with IRF-8 and HA-ubiquitin (Fig. 4B). Conversely, immunoprecipitation of the same lysates with anti-HA antibody and immunoblotting with anti-IRF-8 antibody also showed association of ubiquitin with IRF-8 (data not shown). These results confirm that macrophages and 293T cells possess the same cellular machinery responsible for ubiquitination and degradation of IRF-8.
Next we constructed two amino-terminal FLAG epitopetagged carboxyl-terminal truncation expression plasmids and transfected them into 293T cells (Fig. 5A). Both carboxyl-terminal truncation mutants were stable, showing that the carboxyl-terminal 34 amino acids are required for rapid IRF-8 degradation (Fig. 5B). We then co-transfected 293T cells with full-length IRF-8 or the two carboxyl-terminal truncated expression plasmids and HA-ubiquitin. Co-immunoprecipitation experiments of 293T cells co-transfected with IRF-8 constructs and HA-ubiquitin showed that the IRF-8 truncation mutants were not ubiquitinated (Fig. 5C). The stability and lack of ubiquitination of IRF-8 carboxyl-terminal truncation mutants indicate that elements in this domain are required for E3 recognition and targeting for proteasome.
Dominant Negative Ubiquitin Mutants Block IRF-8 Degradation-Although ubiquitin Lys-48 is the primary site of isopeptide polyubiquitination required to target protein for degradation (31), polyubiquitin chains linked through Lys-29 have also been identified both in cells and in vitro and also target proteins for degradation (32). In contrast, ubiquitin Lys-63 is involved in endosomal trafficking and in activation of TLR pathways. K48R and K29R ubiquitin mutants act as ubiquitin dominant negatives, resulting in premature termination of ubiquitin chains and inhibition of degradation by the proteasome. We have shown that degradation of IRF-8 is proteasomedependent, whereas IRF-4 is not ubiquitinated proteasomeindependent. To confirm that ubiquitination is required for degradation, we co-transfected into 293T cells IRF-8 and IRF-4 together with ubiquitin mutants K48R, K29R, and K63R. Cotransfection of K48R or K29R ubiquitin mutants resulted in higher expression of IRF-8 but had no effect on IRF-4 (Fig. 6). There was no change of glyceraldehyde-3-phosphate dehydrogenase, which is degraded by the lysosomal pathway. These data confirm that ubiquitination of IRF-8 is required for degradation.
Cbl Is an E3 Ligase That Targets IRF-8 Degradation-Members of the Cbl family are ubiquitin ligases that function as negative regulators of activated tyrosine kinase-coupled receptors. Cbl proteins contain, in addition to a typical E3 ring domain, a conserved amino-terminal phosphotyrosine-binding PTB domain, which targets the ubiquitination of activated protein tyrosine kinases and other phosphotyrosine-containing proteins (33,34). Because it has been reported that IRF-8 is constitutively tyrosine-phosphorylated in vivo (17), we speculated that Cbl may also be involved in IRF-8 degradation. We first confirmed that Cbl was expressed in macrophages and 293T cells by immunoblotting (Fig. 7A). Co-transfection of IRF-8 and Cbl dominant   FIG. 1. IRF-8 is subject to proteasomal degradation in peritoneal macrophages. A, thioglycollate-elicited peritoneal macrophages were activated with IFN-␥ (10 ng/ml) and LPS (1 g/ml) for 6 h, washed three times, and incubated with fresh medium in the presence or absence of MG132 (10 M). Cell lysates were subjected to 10% SDS-PAGE, and after transfer to nitrocellulose, the blots were immunostained for IRF-8 and ␤-actin. B, thioglycollate-elicited peritoneal macrophages were activated with IFN-␥ (10 ng/ ml) and LPS (1 g/ml) for 6 h, cycloheximide (10 g/ml) was added throughout the incubation, and the cells were cultured in the presence or absence of MG132 (10 M). Cell lysates were analyzed as described before for IRF-8, IRF-1, IRF-4, and ␤-actin.

FIG. 2. IRF-8 degradation is blocked by proteasome inhibitors.
Thioglycollate-elicited peritoneal macrophages were activated with IFN-␥ (10 ng/ml) and LPS (1 g/ml) for 6 h, cycloheximide (10 g/ml) was added, and the cells were incubated with fresh medium in the presence of calpastatin ( negative plasmids into 293T cells resulted in increased expression of IRF-8 as predicted if Cbl were the E3 ligase mediating ubiquitination of IRF-8 (Fig. 7B). If Cbl is involved in the degradation of IRF-8, there should be direct physical interaction between IRF-8 and Cbl. To test this, we co-transfected FLAGtagged IRF-8 or IRF-8 carboxyl-terminal deletion mutants (IRF-8-(1-390) and IRF-8-(1-356)) with Cbl into 293T cells. Immunoprecipitation of FLAG-IRF-8 from cell lysates resulted in co-precipitation of Cbl, demonstrating that a complex was formed between the two proteins. However, neither IRF-8 carboxyl-terminal truncation mutant interacted with Cbl (Fig. 7C). We further analyzed the interaction between Cbl and IRF-8 by Superose 6 gel filtration chromatography. Gel filtration of lysates from 293T cells co-transfected with Cbl and IRF-8 showed two proteins co-eluting over a wide range of molecular mass (Fig. 7D).
To demonstrate the importance of Cbl in IRF-8 degradation, we co-transduced retrovirus-encoded IRF-8 and HA-ubiquitin into Cblϩ/ϩ and CblϪ/Ϫ fibroblasts. After 2 days, the cells were treated with or without MG132 (10 M) for 4 h, and whole cell lysates were then immunoprecipitated with an anti-IRF-8 antibody and immunoblotted with anti-HA antibody. Incubation with MG132 clearly increased IRF-8 ubiquitination in Cblϩ/ϩ cells but had no effect in CblϪ/Ϫ cells (Fig. 8A). Although there is some degradation of IRF-8 in CblϪ/Ϫ macrophages, the protein is relatively more stable than in wild type macrophages (Fig. 8B). Taken together, these results show that Cbl serves an E3 ligases mediating IRF-8 degradation.
IRF-8 Degradation Regulates IL-12 p40 Expression-Studies with IRF-8Ϫ/Ϫ mice revealed that IRF-8 is an obligatory activator of several macrophage genes (20,35,36) and , together with IRF-1, plays a central role in regulation of IFN-␥-stimulated IL-12 p40 synthesis. Our data indicate that the ubiquitinproteasome pathway is involved in IRF-8 degradation and that Cbl serves as an E3 ligase for IRF-8. We would like to test the hypothesis that the degradation of IRF-8 affects IL-12 expression. To approach this question, RAW 264.7 cells were cotransfected for 30 h with an IL-12 p40 promoter reporter and IRF-8 or the IRF-8 truncation mutant IRF-8-(1-390), which is resistant to degradation. Transfection of IRF-8 mutant IRF-8-(1-390) resulted in significantly greater IL-12 p40 promoter activation than IRF-8 (Fig. 9A), suggesting that IRF-8 degradation regulates IL-12 expression. Because Cbl serves as an E3 ligase for IRF-8 degradation, we analyzed IL-12 p40 promoter activation in CblϪ/Ϫ fibroblasts. Both Cblϩ/ϩ and CblϪ/Ϫ cells were co-transfected with IL-12 p40 promoter reporter and IRF-8 plasmids for 30 h. IL-12 p40 promoter activity was significantly up-regulated in CblϪ/Ϫ cells compared with Cblϩ/ϩ cells (Fig. 9B). In addition, we co-transduced into RAW264.7 cells retrovirus-encoded IRF-8 and dominant negative Cbl or green fluorescent protein as a control and found IL-12 p40 protein production was significantly enhanced in cells co-transduced with IRF-8 and dominant negative Cbl after activation with IFN-␥ and LPS for 24 h (Fig. 9C). To confirm these experiments, we stimulated peritoneal macrophages from both wild type and CblϪ/Ϫ mice, and the data showed that IL-12 p40 production was significantly higher in CblϪ/Ϫ cells (Fig.  9D). Taken together, the results suggest IRF-8 degradation mediated by Cbl down-regulates IL-12 expression. DISCUSSION This study describes the relationship between the degradation of IRF-8 and IL-12 gene expression. We have shown that the proteasome inhibitor MG132 blocks the rapid degradation of IRF-8 in primary macrophages, macrophage cell lines, and 293T cells. Ubiquitination of IRF-8 was demonstrated by retroviral co-transduction of RAW 264.7 cells with IRF-8 and HA-ubiquitin. Furthermore, dominant negative ubiquitin mutants K48R and K29R clearly increase the expression of IRF-8. Several lines of experimentation show that E3 ligase Cbl co-precipitates IRF-8, and expression of IRF-8 is increased by expression of a Cbl dominant negative, showing a physical interaction. In contrast to IRF-8, carboxyl-terminal truncations of IRF-8 are more stable and are not ubiquitinated, and they fail to interact with Cbl, indicating the importance of carboxyl-terminal domain. Cbl is an E3 ligase that interacts with IRF-8, and IRF-8 ubiquitination is significantly attenuated in CblϪ/Ϫ cells. The stable carboxylterminal truncation mutant more strongly induces IL-12 p40 promoter activation than IRF-8, and similarly, IRF-8-induced IL-12 p40 promoter activity is stronger in CblϪ/Ϫ cells than in Cblϩ/ϩ cells. Co-transduction of dominant negative Cbl significantly increased IL-12 p40 protein production induced by IFN-␥ and LPS. Furthermore, macrophages from CblϪ/Ϫ mice have enhanced IL-12 p40 production compared with wild type mice. These results indicate that the degradation of IRF-8 mediated by Cbl down-regulates IL-12 expression, suggesting a novel mechanism for the regulation of cytokine gene expression by the ubiquitin-proteasome pathway.
IRF-8 is a transcription factor that plays an important role in the immune system (5). The IRF-8 promoter has both a GAS sequence, induced by IFN-␥, and a nuclear factor-B site, induced by LPS (10). Interaction of these two sites accounts for the synergistic induction of IRF-8 by IFN-␥ and LPS in macrophages. In vivo experiments showed that IRF-8Ϫ/Ϫ mice are deficient in synthesis of IL-12 and IFN-␥ and, as a consequence, are unable to mount a Th1-mediated immune response (20,35). Additional studies reported that IRF-8 binds to the IFN-␥stimulated response element site of the IL-12 promoter and is directly involved in the activation of IL-12 p40 gene expression (18,37). In addition, IRF-8 controls other proteins important in inflammation, such as inducible nitric-oxide synthase, IL-18, and IL-1 (11,38,39). Although the Th1 immune response is important for host defense against microbial pathogens, an unregulated Th1 immune response with enhanced production of Th1 cytokines is found in many autoimmune or inflammatory diseases. Therefore, the balance between the synthesis and degradation of IRF-8 is important to the host.
Like IRF-8, we also found that IRF-1 was degraded rapidly, consistent with a recent report that rapid degradation of IRF-1 protein in COS7 cells was inhibited by MG132 (30). IRF-3, another IRF family member, which regulates the expression of the interferon-␤ gene, is also degraded by the proteasome pathway, but only when phosphorylation on serine and threonine residues has been induced by virus infection (40). Therefore, it appears that the proteasome pathway is vital for the degradation of several IRF family members. Interestingly, IRF-4, another member of IRF family that is homologous to IRF-8, was quite stable in macrophages and 293T cells. The reasons for the differences in protein stability among IRF family members are not clear.
Targeting of cellular proteins for proteolysis by the proteasome is a highly complex and tightly regulated process (31). Ubiquitin, an evolutionarily conserved protein of 76 residues, has diverse cellular functions, including marking proteins for degradation by the proteasome or trafficking to the lysosomal compartment and activation of signaling pathways. However, because some non-ubiquitinated proteins are degraded by the proteasome, and in other cases ubiquitination serves other functions than as a degradation signal, ubiquitination alone does not allow one to conclude that a protein is degraded by the proteasome (41,42). In the present study, we show that the dominant negative ubiquitin mutants K48R and K29R significantly increase the stability of IRF-8, confirming a link between ubiquitination and degradation of IRF-8.
Several degradation signals in protein substrates have been implicated to control the ubiquitin-proteasome pathway (43), including PEST sequences, degradation signals of a hydrophobic nature, phosphorylation-dependent signals, and the carboxyl-terminal sequence (44 -47). The IRF-1 carboxyl-terminal 39 residues are thought to control the stability of IRF-1 protein, and the hydrophobic nature of the carboxyl-terminal of IRF-1 was proposed as a degradation signal (30). Phosphorylation of IRF-3 carboxyl-terminal region is a signal for its subsequent degradation, implying that this carboxyl-terminal region may function as the phosphorylation-dependent degradation signal recognized by an E3 ligase complex (40). Our experiments using the IRF-8 carboxyl-terminal truncated mutants indicate that the carboxyl-terminal 390 -424 residues of IRF-8 are required for ubiquitination and degradation. The carboxyl-terminal region does not have typical PEST sequences but is hydrophobic, in comparison with other regions of IRF-8.
Our results show that Cbl is an E3 ligase for IRF-8. The RING finger domain and phosphotyrosine-binding domain (PTB) of Cbl are required for ubiquitination in vitro and in vivo (48). Biochem- ical mutational analysis and crystal structure data demonstrate that the RING finger mediates binding to E2 enzymes. We find that the residues 390 -424 at the carboxyl terminus of IRF-8 are required both for physical interaction with Cbl and for degradation, but this sequence does not contain any tyrosine residues. However, there is a tyrosine residue that is very close to the carboxyl terminus that might serve as a PTB docking site in conjunction with more carboxyl-terminal residues.
Although our data showed that Cbl is involved in IRF-8 degradation, the fact that IRF-8 is still degraded in cCbl-deficient macrophages suggests that some other E3 ligases may also contribute to the control of IRF-8 degradation. Our future study will address two issues: 1) identification of other E3 ligases responsible for IRF-8 stability, and 2) detailed dissection of Cbl in the control of protein stability of other IRF family members including IRF-1.
The IL-12 p40 gene, reflecting its importance in the generation of a Th1 versus a Th2 immune response, is subject to complex transcriptional regulation by multiple transcription factors. Nuclear factor-B, CAAT/enhancer-binding protein, AP-1, IRF-1, and IRF-8 have all been reported as important activators of IL-12 p40 transcription (49 -53). Although these studies suggest that these transcription factors are activators of IL-12 p40 gene, how the degradation of transcription factors affects IL-12 p40 promoter activation is incompletely understood. In the present study, we found that an IRF-8 carboxyl-terminal truncation mutant much more strongly activates IL-12 p40 promoter and that a dominant negative Cbl mutant significantly enhanced IRF-8induced IL-12 p40 promoter activation. Furthermore, IRF-8-induced IL-12 p40 promoter activation was up-regulated in CblϪ/Ϫ cells in contrast to Cblϩ/ϩ cells, and co-transduction of dominant negative Cbl significantly enhanced IRF-8-induced IL-12 p40 production. These results indicate that the ubiquitin-proteasome pathway is involved in the regulation of IL-12 gene expression, suggesting that balanced control of transcription factor expression level is an important mechanism for the modulation of target gene expression.
Regulated protein degradation is critical for many cellular processes, including control of the cell cycle and inflammation. Deeper understanding of the mechanism for IRF-8 degradation may facilitate the modulation of this transcription factor, which is central to a competent Th1 immune response. Recent studies revealed that IRF-1 and IRF-3 are subject to proteasomal degradation, implying conservation of the signals controlling degradation among various members of the IRF family. A future study will address identification and characterization of the specific ubiquitin-conjugating enzymes involved in the degradation of IRF family members, which will help understand the consequence of degradation of transcription factors in immune responses.
FIG. 8. Cbl controls IRF-8 ubiquitination and degradation. A, Cblϩ/ϩ or CblϪ/Ϫ fibroblasts were co-transduced with retrovirus-encoded IRF-8 and HA-ubiquitin for 48 h. Cell lysate was immunoprecipitated with anti-IRF-8 antibody and immunoblotted with anti-HA mAb. B, peritoneal macrophages from wild type and cCblϪ/Ϫ mice were activated with IFN-␥ and LPS for 6 h, cycloheximide (10 g/ml) was added, and the cells were incubated with in the presence or absence of MG132 (10 M) as indicated. Cell lysates were analyzed as described for IRF-8.