Abstract
IFN-α inhibits B lymphocyte development, and the nuclear protein Daxx has been reported to be essential for this biological activity. We show in this study that IFN-α inhibits the clonal proliferation of B lymphocyte progenitors in response to IL-7 in wild-type, but not in tyk2-deficient, mice. In addition, the IFN-α-induced up-regulation and nuclear translocation of Daxx are completely abrogated in the absence of tyk2. Therefore, tyk2 is directly involved in IFN-α signaling for the induction and translocation of Daxx, which may result in B lymphocyte growth arrest and/or apoptosis.
Interferon-α mediates potent antiviral and antiproliferative activities on target cells through its interaction with IFN-α receptors. Janus kinase (Jak)3 1 and Tyk2, the nonreceptor tyrosine kinases of the Jak family, are associated with IFN-α receptors and play a pivotal role in transducing IFN-α signals (1, 2, 3). Activated Jaks phosphorylate tyrosine residues on IFN-α receptors, thereby recruiting Stat1 and 2, and other signaling molecules to the activated receptor complex. Stat1 and 2 are in turn phosphorylated by Jaks, and these activated Stats subsequently associate to form either Stat1 homodimers or the transcription factor IFN-stimulated gene factor-3, which then translocates to the nucleus to regulate gene expression (4, 5). Tyk2 was identified as a novel protein kinase, based on its ability to rescue signaling in mutated fibroblasts, nonresponsive to IFN-α (6). However, we and others have shown using tyk2-deficient mice that tyk2 has a restricted function and does not play a major role in IFN-α signaling (7, 8). In contrast, Jak 1-null cells fail to respond to IFN-α (9). As for Stat function, in Stat1-null mice all the examined IFN-α-induced responses, including the induction of MHC class I molecules in T cells, NO production by macrophages, and antimicrobial or antiviral activities were rendered defective in the absence of Stat1 (10, 11). Stat2-null mice also showed an increased susceptibility to viral infection (12). Accordingly, the Jak1-Stat signaling pathway is thought to be essential for IFN-α signaling.
IFN-α is a potent inhibitor of IL-7-dependent growth of early B cell lineage progenitors, effectively aborting further B cell lineage differentiation at the pro-B cell stage (13). Although many IFN-α-induced responses are abrogated in Stat1-deficient mice (10, 11), the inhibition of IL-7-dependent B lymphopoiesis by IFN-α is unaffected in the absence of Stat1 (14).
Using previously generated tyk2-deficient mice, we show in this study that tyk2 is essential for IFN-α-induced B lymphocyte growth inhibition, through the up-regulation and nuclear translocation of Daxx, which was recently identified as playing an important role in the IFN-α-mediated inhibition of B lymphopoiesis (15).
Materials and Methods
Mice
The generation of tyk2-deficient mice has been previously described (7). Mice were housed and bred in the Kyushu University Animal Center (Fukuoka, Japan).
Abs and cytokines
Anti-Daxx Ab (M-112) was purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Murine natural IFN-α was obtained from Hayashibara Biochemical Institute (Okayama, Japan).
CFU-IL7 colony assay
Bone marrow cells were prepared and suspended in a 1-ml assay medium as previously described (16, 17). The semisolid agar assay for CFU-IL-7 was done with 1 ng/ml IL-7 with or without 1000 U/ml IFN-α. Numbers of colonies were counted after 6 days of culture.
In vitro bone marrow cultures
Bone marrow cells from wild-type or tyk2-deficient mice were prepared and stimulated by 1000 U/ml IFN-α for the indicated times. Cells were harvested and analyzed for the expression of Daxx.
Generation of B cell precursors on ST-2 cells
We generated B cell precursors from bone marrow cells by coculturing with ST-2 cells (18). ST-2 is a bone marrow-derived stromal cell line that can support the generation of B lymphocyte progenitor cells. ST-2 cells produce IL-7, and the production of IL-7 from ST-2 is enhanced by close association with hemopoietic cells. ST-2 cells were grown to ∼80% confluence in a 10-cm dish. Bone marrow cells (1 × 106) from wild-type or tyk2-deficient mice were cultured on ST-2 cells with 10 ml of RPMI containing 10% FCS and 5 × 10−5 mol/L 2-ME. Ten days after coculture, 1000 U/ml IFN-α was added to the dish. Cells were harvested at the indicated times after adding IFN-α.
RT-PCR analysis
RNA was isolated from freshly harvested cells using Isogen (Nippon Gene, Tokyo, Japan) and dissolved in diethyl pyrocarbonate-treated water. First-strand cDNA synthesis from RNA was conducted using poly(dT) and reverse transcriptase, following protocols supplied by the manufacturer (Takara Biomedicals, Tokyo, Japan). Target RNA (1 μg) was reverse-transcribed using 0.25 U AMV Reverse Transcriptase XL (Takara Biomedicals) at 42°C for 30 min in the presence of 50 mM KCl, 10 mM Tris-HCl (pH 8.3), 5 mM MgCl2, 1 mM dNTPs, 0.25 U RNase inhibitor, and 0.125 μM oligo(dT)-adaptor primer. Each 20-μl PCR contained 1 μl of first-strand cDNA, 1 μM each primer (sense and antisense), 0.2 mM dNTPs, 1.5 mM MgCl2, 1× polymerase buffer, and 2 U Taq polymerase (Takara Biomedicals). For evaluation of Daxx expression, forward, 5′-CCCATGGCCACCGATGACAGCAT-3′, and reverse, 5′-AGGGTTAGGGCCCGACGCCTCACT-3′, primers were used. Expression of β-actin was examined by using forward, 5′-TAGACTTCGAGCAGGAGGAGATG-3′, and reverse, 5′-CGTACTCCTGCTTGCTGATCCA-3′, primers. The PCR mixture was denatured in a thermal cycler at 94°C for 3 min and then 25 cycles were performed, each consisting of denaturing at 94°C for 1 min, annealing at 60°C for 2 min, and extension at 72°C for 3 min (15).
Real-time PCR assay
Relative quantification of Daxx in cells was performed using real-time quantitative PCR, by a TaqMan assay on an ABI 7000 system. Real-time quantitative PCR used a cDNA template with the appropriate primers as follows: primers used to amplify murine Daxx were 5′-CATGAACTGGTGACCAGCTCTCT-3′ and 5′-TGGCCACACTGGTCTTATAAATACA-3′, with a murine Daxx probe, FAM-5′-TCCATCCCTGCTTCTCCAGACACCC-3′-TAMRA. Primers used to amplify murine GAPDH were 5′-ACGGCAAATTCAACGGCA-3′ and 5′-AGATGGTGATGGGCTTCCC-3′, with a murine GAPDH probe, FAM-5′-AGGCCGAGAATGGGAAGCTTGTCATC-3′-TAMRA. PCR amplifications were performed in a 50-μl volume, containing 1 μl of cDNA template, 50 mM KCl, 10 mM Tris-HCl (pH 8.3), 10 mM EDTA, 200 μM dNTPs, 3 mM MgCl2, 200 nM of each primer, 0.625 U AmpliTaq Gold, and 0.25 U AmpErase uracil N-glycosylase. Each amplification also contained 100 nM of the appropriate detection probe. Each PCR amplification was performed in duplicate, using the following conditions: 50°C for 2 min preceding 95°C for 10 min, followed by 40 cycles of amplification (95°C for 15 s, 60°C for 1 min). In each reaction, GAPDH was amplified as a housekeeping gene to generate a standard curve and correct for variations in target sample quantity. Relative copy number was calculated for each sample based on the standard curve after normalization to GAPDH content.
Western blotting
Cells from wild-type or tyk2-deficient mice were lysed in lysis buffer as previously described (19). Cell lysates were centrifuged at 12,000 × g for 15 min to remove debris. Total cell lysates were resolved by SDS-10% PAGE and transferred to a nitrocellulose membrane. Membranes were probed using anti-Daxx and anti-Stat1 Abs and visualized with the ECL detection system (Amersham, Uppsala, Sweden).
Immunocytochemistry
Cells centrifuged onto glass slides were fixed in cold methanol for 20 min and washed extensively in PBS before blocking with 10% normal goat serum and incubating overnight at 4°C either with rabbit polyclonal anti-Daxx Abs (Santa Cruz Biotechnology) or with nonimmune rabbit serum as a negative control. The slides were then incubated for 1 h with an Alexa Fluor 488-conjugated goat anti-rabbit IgG Ab. After washing in PBS, slides were mounted with coverslips in glycerol containing antifade reagent and propidium iodide, and were examined by confocal microscopy (LSM-GB2000; Olympus, Tokyo, Japan) (15).
Results
Tyk2 is required for IFN-α mediated inhibition of IL-7-dependent B lymphopoiesis
IL-7 induces the proliferation of B cell precursors. We used a colony assay to evaluate the role of tyk2 on IL-7-dependent growth of B lymphocytes. The absence of tyk2 did not have any effect on the number of IL-7-induced colonies of B lymphocytes from bone marrow cells (Fig. 1⇓). IFN-α suppressed the CFU-IL-7 from wild-type mice, but IFN-α-induced B lymphocyte growth suppression was not observed in tyk2-deficient bone marrow cells (Fig. 1⇓).
IFN-α inhibition of IL-7-dependent B lymphopoiesis in wild-type and Tyk2-deficient mice. Adult bone marrow cells from wild-type or tyk2-deficient mice (5 × 104/plate) were plated in semisolid agar cultures supplemented with IL-7 in the presence or absence of IFN-α. After 6 days of culture, the total number of colonies (>50 cells) was enumerated.
Tyk2 is required for the induction of Daxx expression by IFN-α
We next examined the expression and localization of Daxx following IFN-α treatment, as Daxx has recently been reported to be important in IFN-α-mediated inhibition of B lymphopoiesis (15).
First, we examined Daxx expression in bone marrow cells before and after IFN-α stimulation. RT-PCR assessment of Daxx transcript levels showed that Daxx expression was increased in bone marrow cells from wild-type mice by treatment with IFN-α for 8 h (Fig. 2⇓A). In contrast, Daxx transcript levels only increased slightly after IFN-α treatment of bone marrow cells from tyk2−/− mice (Fig. 2⇓A). Analysis of expression by real-time quantitative TaqMan PCR revealed that Daxx mRNA levels in bone marrow cells from wild-type mice were significantly increased by IFN-α stimulation (Fig. 2⇓B). In contrast to wild-type mice, IFN-α increased Daxx transcript levels more modestly in bone marrow cells from tyk2-deficient mice (Fig. 2⇓B).
Effect of IFN-α on Daxx expression in wild-type and tyk2-deficient bone marrow cells. A, Daxx and β-actin transcripts were analyzed by RT-PCR in bone marrow cells from wild-type or tyk2-deficient mice before and 8 h after treatment with IFN-α. B, Real-time RT-PCR analysis of Daxx expression in bone marrow cells from wild-type or tyk2-deficient mice before and 8 h after treatment with IFN-α. Results have been normalized by comparison to GAPDH mRNA expression. C, Western blot analysis of Daxx and Stat1 levels in wild-type and tyk2-deficient bone marrow cells either untreated or treated with IFN-α for 24 h.
Next, we examined intracellular Daxx protein levels in bone marrow cells before and after IFN-α treatment. The expression of Daxx protein in bone marrow cells from wild-type mice was enhanced by treatment with IFN-α for 24 h. In contrast to wild-type mice, IFN-α did not increase Daxx protein levels in bone marrow cells from tyk2-deficient mice (Fig. 2⇑C).
As IFN-α-mediated inhibition of B lymphocyte progenitor clonal proliferation in response to IL-7 was not observed in tyk2-deficient mice, we enriched B cell precursors by coculturing bone marrow cells with the IL-7-producing stromal cell line ST-2. Ten days after coculture, >95% of cells were positive for B220 (data not shown). By using enriched B cell precursors from wild-type mice or tyk2-deficient mice, we examined the time course of induction of Daxx expression by IFN-α (Fig. 3⇓A). The induction of Daxx mRNA was rapid, and began to decrease after 17 h of IFN-α stimulation. There was no difference in the time course of Daxx induction by IFN-α in B cell precursors from wild-type and tyk2-deficient mice. The increase in Daxx transcript levels by IFN-α in B cell precursors from tyk2-deficient mice, however, was much less than that from wild-type mice.
Time course of Daxx expression following IFN-α stimulation in wild-type and tyk2-deficient B cell precursors. A, B cell precursors from wild-type or tyk2-deficient mice were stimulated with IFN-α for the indicated times. RNA detection and quantification was performed using RT and real-time TaqMan PCR. Quantification of Daxx mRNA was based on expression relative to the computed amount of GAPDH mRNA. B, Western blot analysis of Daxx and Stat1 levels in wild-type and tyk2-deficient B cell precursors treated with IFN-α for the indicated time.
IFN-α augmented the expression of Daxx protein in B cell precursors from wild-type mice, with maximal induction of Daxx protein observed after 48 h of IFN-α stimulation (Fig. 3⇑B). Although IFN-α also augmented the expression of Daxx protein in B cell precursors from tyk2-deficient mice, the level of induction was much less than from wild-type mice.
Tyk2 is required for IFN-α-mediated nuclear localization of Daxx
The IFN-α-triggered pathway of apoptosis involves translocation of Daxx to nuclear bodies and the related suppression of cell-cycle-related genes. To evaluate changes in the intracellular localization of Daxx protein, B cell precursors obtained from coculturing bone marrow cells with ST-2 cells were examined before and after IFN-α treatment by immunofluorescence confocal microscopy with polyclonal anti-Daxx Abs (Fig. 4⇓). Daxx nuclear body translocation was induced by IFN-α in wild-type B cell precursors (Fig. 4⇓B) as previously reported. In contrast, Daxx protein was not concentrated within nuclear bodies in tyk2-deficient B cell precursors following IFN-α treatment (Fig. 4⇓D).
Localization of Daxx protein after IFN-α treatment in wild-type and tyk2-deficient mice. B cell precursors from wild-type (A and B) and tyk2-deficient (C and D) mice were cultured with (B and D) or without (A and C) 1000 U/ml IFN-α for 16 h. Daxx Ab stains green and nuclear staining with the fluorescent red propidium iodide is visualized by confocal microscopy.
Discussion
When cells are treated with IFN-α, Jak1 and tyk2 are initially phosphorylated, followed by Stat1 and Stat2 activation. The activated Stats associate to form either Stat1 homodimers or IFN-stimulated gene factor-3, and then bind to the IFN-stimulated response element of IFN-α-inducible genes. Although tyk2 was initially cloned as a molecule essential for transducing IFN-α signals (6), we and others have shown that the absence of tyk2 either did not affect the activation of Stat1 by IFN-α, or only affected some IFN-α-induced biological effects, such as NO production from macrophages (7, 8). Moreover, most IFN-α-induced biological functions such as anti-viral capacity and MHC class I expression were not affected in the absence of tyk2 (7, 8). In contrast, all IFN-α-induced responses examined were rendered defective in the absence of Stat1 (10, 11). Accordingly, the Jak1-Stat1 signaling pathway is thought to be essential for IFN-α signaling.
IL-7, an essential growth factor for B and T cell lymphopoiesis in mice, stimulates the formation of B lymphocyte colonies from bone marrow cells in vitro (20). There was no difference in the number of CFU-IL-7 in bone marrow cells from wild-type and tyk2-deficient mice, indicating that the number of B lymphocyte progenitor cells was not affected in tyk2-deficient mice. IFN-α inhibited the IL-7 promoted growth of early B cell lineage cells (13). The number of CFU-IL-7 from wild-type bone marrow cells was decreased by the addition of IFN-α. The inhibition of IL-7-dependent B lymphopoiesis by IFN-α was almost abrogated in the absence of tyk2 (Fig. 1⇑), although it was reported to be unaffected in Stat1-deficient mice (14). Accordingly, other signaling molecules, besides Stat1, which is phosphorylated by IFN-α regardless of the absence of tyk2 (7), are thought to transduce the IFN-α signal inhibiting B lymphocyte growth.
Daxx was reported to be involved in the Fas and TGF-β apoptosis-signaling pathway (21, 22). Daxx was originally cloned as a FasR-associated protein and binds specifically to the death domain of the FasR, although Daxx itself lacks a death domain (21, 23, 24). Fas belongs to the TNFR superfamily and induces apoptosis upon receptor oligomerization (25). There are two independent signaling pathways downstream of the FasR (24), involving the adapter protein Fas-associated death domain (25) and Daxx (23). The activation of Fas-associated death domain induces a protease cascade (26, 27), while that of Daxx induces c-Jun N-terminal kinase activation, leading to apoptosis (23). Overexpression of Daxx enhances Fas-induced apoptosis (23), and the targeted disruption of the Daxx gene in mice results in embryonic lethality (28). Daxx is also involved in coupling TGF-βR signaling with components of the apoptotic machinery (22). TGF-β induces apoptosis in primary and cultured lymphocytes (29). Daxx associates with the cytoplasmic domain of the type II TGF-βR and transduces apoptotic signals by TGF-β (22). The C-terminal portion of Daxx (resides 626–739) acts as a dominant negative inhibitor of Fas-mediated apoptosis, presumably by competing with endogenous Daxx for binding to Fas. The expression of the C-terminal portion of Daxx suppresses TGF-β-mediated apoptosis, as well as Fas-mediated apoptosis (22).
Recently, Daxx was also reported to be essential for inhibition of B lymphopoiesis by IFN-α (15). IFN-α enhances Daxx expression, with concomitant increases in Daxx protein levels and nuclear body translocation. Moreover, Daxx antisense oligonucleotides rescue IFN-α-treated pro-B cells from growth arrest and apoptosis. Therefore, we examined the expression of Daxx mRNA and protein in tyk2-deficient cells. Strikingly, the augmentation of Daxx expression by IFN-α, which is observed in wild-type cells, was drastically reduced in tyk2-deficient cells (Figs. 2⇑ and 3⇑). Moreover, the nuclear translocation of Daxx by IFN-α was also abrogated in the absence of tyk2 (Fig. 4⇑). Our study demonstrates that tyk2 is essential for the transduction of IFN-α-induced B lymphocyte growth arrest signals through the activation of some signaling molecule other than Stat1, followed by the up-regulation and nuclear translocation of Daxx. To clarify the biochemical interactions between tyk2 and Daxx, we examined whether tyk2, which is a tyrosine kinase and activates Stats through phosphorylation, also phosphorylated Daxx. Daxx was not phosphorylated in B lymphocytes stimulated with IFN-α (data not shown). In addition, no changes in the phosphorylation pattern of Daxx were detected upon stimulation with TGF-β or Fas (22).
The apoptotic signaling pathway downstream of Daxx is as yet unknown. Because Daxx associates with Fas and TGF-βRII, it may also associate with the IFN-αR, tyk2, or with the IFN-αR/tyk2 complex. The presence of Daxx within the receptor complex may be important for the recruitment and activation of the next substrate downstream of Daxx in the Fas, TGF-β, and IFN-α apoptosis-signaling pathway. Alternatively, the localization of Daxx may be important. Although Daxx is an adapter protein and is associated with Fas and TGF-βRII, it can localize to the nucleus of some cells (30). The localization of Daxx in either the cytoplasmic or nuclear compartment was reported to be dependent upon the cell type and/or its functional status (31). The subcellular distribution of the Daxx protein was not changed even after exposure of the cells to TGF-β (22). In contrast, the IFN-α-triggered pathway of apoptosis involved the translocation of Daxx to nuclear bodies and the related suppression of cell cycle-related genes (15). The requirement of tyk2 for IFN-α-mediated apoptosis of B lymphocytes may be a direct result of the inhibition of Daxx nuclear translocation by IFN-α (Fig. 4⇑).
Acknowledgments
We thank A. Tomioka and M. Sato for their excellent technical assistance.
Footnotes
-
↵1 This work was supported in part by a Grant of the Welfide Medical Research Foundation and Grants-in-Aid for Scientific Research (nos. 11770577, 11307015, and 13218096) from the Ministry of Education, Science, Sports, and Culture in Japan.
-
↵2 Address correspondence and reprint requests to Dr. Kazuya Shimoda, First Department of Internal Medicine, Faculty of Medicine, Kyushu University, 3-1-1 Maidashi, Higashi-ku, Fukuoka, Fukuoka 812-8582, Japan. E-mail address: kshimoda{at}intmed1.med.kyushu-u.ac.jp
-
↵3 Abbreviation used in this paper: Jak, Janus kinase.
- Received May 20, 2002.
- Accepted September 4, 2002.
- Copyright © 2002 by The American Association of Immunologists
References
In this issue
