Failure of Measles Virus to Activate Nuclear Factor-kB in Neuronal Cells: Implications on the Immune Response to Viral Infections in the Central Nervous System1 Suhayl Dhib-Jalbut,2 Jane Xia, Himabindu Rangaviggula, Yu-Yan Fang, and Terry Lee Neurons are postmitotic cells that foster virus persistence. These cells lack the HLA class I molecules required for clearance of infected cells. Previously, we showed that HLA class I is induced by measles virus (MV) on glial cells, which is primarily mediated by IFN-b. In contrast, MV was unable to induce HLA class I or IFN-b in neuronal cells. This failure was associated with lack of NF-kB binding to the positive regulatory domain II element of the IFN-b promoter, which is essential for virus-induced IFN-b gene activity. In this study, we demonstrate that the failure to activate NF-kB in neuronal cells is due to the inability of MV to induce phosphorylation and degradation of IkB, the inhibitor of NF-kB. In contrast, TNF-a induced degradation of IkBa in the neuronal cells, suggesting that failure to induce IkBa degradation is likely due to a defect in virus-mediated signaling rather than to a defect involving neuronal IkBa. Like MV, mumps virus and dsRNA failed to induce IkBa degradation in the neuronal cells, suggesting that this defect may be specific to viruses. Autophosphorylation of the dsRNA-dependent protein kinase, a kinase possibly involved in virus-mediated IkBa phosphorylation, was intact in both cell types. The failure of virus to induce IkBa phosphorylation and consequently to activate NF-kB in neuronal cells could explain the repression of IFN-b and class I gene expression in virus-infected cells. These findings provide a potential mechanism for the ability of virus to persist in neurons and to escape immune surveillance. The Journal of Immunology, 1999, 162: 4024 – 4029.
N
eurons are postmitotic cells that lack the ability to regenerate. Therefore, mechanisms may have evolved to escape the neuronal damage caused by viral infection and immune-mediated cell death. These protective mechanisms may include the presence of a blood brain barrier and very low levels of MHC (HLA) class I on neurons that allow these cells to escape recognition by MHC class I-restricted CTL (1, 2). Previously, we demonstrated that measles virus (MV)3 infection of human glioma cell lines resulted in the up-regulation of HLA class I molecules, which was mediated primarily by IFN-b (3). In contrast, infection of human neuronal cell lines with MV failed to induce HLA class I molecules and IFN-b gene expression, indicating a defect in IFN-b gene regulation by virus (4). The lack of IFN-b induction by MV in neuronal cells was not due to a deficiency in MV receptor expression or MV replication in these cells (4). Similarly, HLA class I and IFN-b were not expressed in MVinfected neurons in subacute sclerosing panencephalitis (SSPE), a persistent MV infection in humans, or in rat subacute measles encephalitis, an animal model of SSPE (5). Collectively, these in
Department of Neurology, University of Maryland at Baltimore, Baltimore, MD 21201; and Department of Veterans Affairs, Baltimore, MD 21201 Received for publication October 5, 1998. Accepted for publication December 22, 1998. 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. 1 This work was supported by Grant 2P50 NS20022-09A1 from the National Institutes of Health and by a Merit Review Grant from the Department of Veterans Affairs. 2 Address correspondence and reprint requests to Dr. Suhayl Dhib-Jalbut, Department of Neurology, University of Maryland Hospital, Rm. N4W46, 22 S. Greene Street, Baltimore, MD 21201. E-mail address:
[email protected] 3 Abbreviations used in this paper: MV, measles virus; AEBSF, 4-(2-aminoethyl)benzene sulfonyl fluoride; CIP, calf intestinal phosphatase; IkB, inhibitor kB; MPS, mumps virus; PIPC, polyinosinic polycytidylic acid; PKR, PRDII, positive regulatory domain II; dsRNA-dependent protein kinase; SSPE, subacute sclerosing panencephalitis.
Copyright © 1999 by The American Association of Immunologists
vitro and in vivo findings suggested that the failure to induce HLA class I and IFN-b genes in neuronal cells after viral infection may contribute to virus persistence by escaping CTL recognition and the absence of the antiviral effect of IFN-b normally produced in infected cells. Demonstration of failure to induce IFN-b in neuronal but not glial cells was associated with lack of NF-kB binding to the positive regulatory domain II (PRDII) element of the IFN-b promoter (4) necessary for virus inducibility of the IFN-b gene promoter activity (6). This defect is virus specific as TNF-a, but MV, mumps virus (MPS), and dsRNA were all unable to induce NF-kB DNA-binding activity (4). NF-kB, a member of the Rel family of proteins, is involved in the regulation of both HLA class I and IFN-b gene expression (6 – 8). The Rel family of proteins includes p50 (NF-kB1), p52 (NF-kB2), and p65 (RelA) and form homodimeric or heterodimeric complexes of which the p50/p65 dimer is most abundant and is strongly transactivating. In unstimulated cells, NF-kB is sequestered in the cytoplasm by complexing with IkB, a family of proteins that includes IkBa, IkBb, Bcl-3, and p105. IkBa predominantly inhibits NF-kB. IkBa is a constitutively phosphorylated protein and hyperphosphorylation induced by a variety of agents including cytokines, virus, or dsRNA signals ubiquitination and subsequent degradation by the 26S proteasome complex (9 –16). This allows NF-kB to translocate into the nucleus and to bind the target kB site. Phosphorylation of IkBa by the dsRNA-dependent protein kinase (PKR) is believed to be involved in the virus-induced activation of NF-kB (13). PKR is a serine-threonine kinase involved in growth inhibition and NF-kB activation through phosphorylating eIF2a and IkBa, respectively (17). PKR functions as a selfassociating dimer with an N-terminal dsRNA-binding domain that requires autophosphorylation for its activity. Autophosphorylation but not transphosphorylation of PKR requires dsRNA. PKR is induced by IFNs and is believed to mediate some of the antiviral and antiproliferative effects of these cytokines (17). PKR activity can 0022-1767/99/$02.00
The Journal of Immunology also be inhibited by some cellular and viral gene products (18, 19). Studies in PKR knockout mice showed that the induction of type I IFN by dsRNA was unimpaired in these mice and that the antiviral response induced by dsRNA and IFN-g was reduced. However, in embryo fibroblasts from these mice, the induction of type I IFN and NF-kB activation by dsRNA were strongly impaired but partially restored with IFN (20). Thus, although PKR is involved in dsRNA-induced activation of NF-kB and type I IFN expression, PKR activity may not be required for virus to induce IFN-ab and NF-kB. Because NF-kB is involved in the regulation of both IFN-b and class I genes, and because both genes are suppressed in virusinfected neuronal cells, we investigated the mechanism responsible for the failure of MV to activate NF-kB in these cells. In this study, we demonstrate that the lack of NF-kB activation in neuronal cells by MV is due to a defect in the phosphorylation and degradation of the NF-kB inhibitor, IkB.
Materials and Methods Cell cultures The neuroblastoma cell line IMR-32 clone was a gift from Dr. Richard J. Ziegler (University of Minnesota, Duluth, MN). This cell line retains a number of neuronal characteristics such as expression of nicotinic and muscarinic cholinergic receptors and neurotransmitter expression (21). The neuroblastoma cell line CHP-126 (22) was used in limited experiments and was a gift from Dr. Lois A. Lampson (Harvard Medical School, Boston, MA). The neural cell lines were grown in MEM medium supplemented with 5% FCS, glutamine, HEPES buffer, penicillin, streptomycin, and gentamicin. The human astrocytoma cell line U-251 MG was a gift from Dr. Darryl Bigner (Duke University, Durham, NC). The characteristics of this cell line and the culture conditions have been previously described (23).
Reagents The Edmonston strain of MV and MPS were obtained from the American Type Culture Collection (ATCC, Manassas, VA), grown in confluent vero cells (monkey kidney fibroblasts from the ATCC), and titrated by plaque assay according to standard methods. The stock titer was 5 3 107 plaque forming units/ml for MV and 108 plaque forming units/ml for MPS. When monolayers of neuronal and glial cells were ;80% confluent, cells were stimulated with varying doses of MV or MPS in serum-free medium for the indicated time periods. In some experiments, MV was inactivated by UV irradiation (400 mW/cm2 3 8 min). dsRNA polyinosinic polycytidylic acid (PIPC) was from Pharmacia (Piscataway, NJ). Human recombinant TNF-a was from Genzyme (Cambridge, MA). Rabbit polyclonal IgG to p65 NF-kB subunit, IkBa, and PKR were from Santa Cruz Biotechnology (Santa Cruz, CA). Calf intestinal phosphatase (CIP) was from Sigma (St. Louis, MO).
Immunoprecipitation and Western blot analyses Monolayer cells were grown in 75-cm2 flasks to ;80% confluency (;2 3 107 cells/flask). Cells were washed twice with PBS, and complete medium was added before infection with MV at a moi of 2.0 or treatment with 500 U/ml of TNF-a for the indicated time periods. After washing four times in ice-cold PBS, cells were lysed in 1 ml of TNT-E (Tris-NaCl-Triton X-100EDTA) lysis buffer (12) containing the following protease inhibitors: 50 mM NaF, 0.1 mM Na orthovanaolate, 20 mM b-glycosophosphate, 10 mM molybdic acid, 21 mg/ml aprotinin, and 0.2 mM 4-(2-aminoethyl)-benzene sulfonyl fluoride (AEBSF; Sigma). Cell lysates were centrifuged at 10,000 3 g for 10 min at 4°C. The supernatants were mixed with 50 ml of protein A-Sepharose (Pharmacia) for 1 h at 4°C. After centrifugation at 10,000 3 g for 10 min at 4°C, the clarified lysates were incubated with the primary or control Abs for 16 h at 4°C. Fifty microliters of protein ASepharose was then added to each tube and incubated with constant shaking for 2 h at 4°C. The tubes were spun again, and the supernatants were discarded. The precipitates were then washed three times with TNT-E buffer containing protease inhibitors. Fifty microliters of 23 sample buffer (Promega, Madison, WI) was then added to the precipitate and boiled for 5 min. The immunoprecipitates were then electrophoresed on a 15% SDS/ polacrylamide gel at 75 V for 16 h. The proteins were then transferred to a Hybond ECL membrane at 30 V overnight. Filter strips were blocked with 5% milk-TBS at 4°C overnight and then incubated with .5 mg/ml primary Ab of milk-TBS for 1 h at room temperature. The blots were
4025 washed three times with TBS-milk, then incubated with anti-IgG HRP conjugate (1:10,000 dilution) for 1 h. After three washes with TBS, the blots were developed using the Amersham enhanced chemiluminescence system (Arlington Heights, IL).
PKR autophosphorylation assay Autophosphorylation of PKR was determined as previously described (13) with the following modifications. Neuronal and glial cells (107 cells/75cm2 flask) were infected with 2 moi of MV for 20 h. Cells were washed with ice-cold PBS and lysed with TNT-E buffer containing AEBSF. PKR was then immunoprecipitated with anti-PKR mAb on protein A-Sepharose beads and washed with kinase reaction buffer (20 mm HEPES (pH 7.5), 50 mM Kcl, 5 mM 2 mercaptoethanol, 1.5 mM Mg (OAC)2, 1.5 mM MnCl2, and 10 mCi [32P]ATP). The beads were then suspended in 25 ml kinase reaction buffer containing 10 mCi of [32P]ATP and incubated at 30°C for 30 min. After washing with ice-cold kinase buffer, the beads were suspended in 50 ml SDS-PAGE sample loading buffer and boiled at 100°C for 5 min. The samples were then electrophoresed on a 10% SDS-PAGE gel and visualized by radiography.
Statistical analysis The densities of the bands observed on Western blot analyses were quantitated by densitometry using NIH Image 1.61 software (National Institutes of Health, Bethesda, MD). For statistical analysis, the densities of the bands observed with treatment were normalized against the density of the IkB band immunoprecipitated from untreated cells. The phosphorylated IkB band (IkBp) density was expressed relative to the density of the hypophosphorylated band (IkBp/IkB) in each lane. The Student’s t test was used to examine the statistical significance of treatment effects using Statworks Mcintosh software.
Results
MV stimulation results in IkBa degradation and dissociation from NF-kB in glial cells We have examined IkBa phosphorylation and degradation in glial cells after MV infection to explore the hypothesis that failure to activate NF-kB by MV in neuronal cells but not in glial cells may be due to a failure of IkBa phosphorylation and degradation in neuronal cells. IkBa was immunoprecipitated from unstimulated or MV-stimulated glial cells using rabbit anti-sera to IkBa or antip65 subunit of NF-kB. The blots were probed with anti-IkBa Ab and visualized by enhanced chemiluminescence (Fig. 1A). As expected, an ;37-kDa band corresponding to IkBa was immunoprecipitated by anti-IkBa and by anti-p65 but not by normal rabbit serum in unstimulated glial cells. After MV infection, a double band corresponding to IkBa could be immunoprecipitated at 5 min. IkBa bands associated with NF-kB were no longer coprecipitated with anti-p65 Ab at 20 min (Fig. 1A). The upper band represents hyperphosphorylated IkBa, because it could be almost abolished by treatment with CIP (24) (Fig. 1B). To examine the kinetics of IkBa degradation, IkBa phosphorylation and degradation were examined at serial time points in the presence of MV. As shown in Fig. 1C, IkBa was hyperphosphorylated within 5 min and degradation occurred between 10 and 30 min after MV infection. After 2 h of viral infection, IkBa appeared to be regenerated, thus entering a new phosphorylation and degradation cycle. The mean effect of MV on IkBa phosphorylation and degradation in the glial cells obtained from three experiments are presented in Fig. 2. IkBa phosphorylation occurred as early as 5 min and significant degradation at 15 and 30 min after MV infection. Lack of IkBa phosphorylation and degradation in neuronal cells in response to MV To determine whether the inability of MV to activate NF-kB in neuronal cells is due to a failure of IkBa phosphorylation and degradation, IkBa immunoprecipitated from unstimulated or MVstimulated neuronal cells using rabbit anti-sera to IkBa or to p65 was examined. In contrast to glial cells, IkBa is not hyperphosphorylated nor degraded in neuronal cells infected with MV for up
4026
DIFFERENTIAL NF-kB ACTIVATION IN NEURONAL AND GLIAL CELLS
FIGURE 1. A, Immunoprecipitation and Western blot analysis of IkBa in unstimulated and MV-stimulated glial cell line U-251 MG. Immunoprecipitation was conducted with anti-IkBa, anti-p65, or control nonimmune rabbit serum. The blots were probed with anti-IkBa Ab and analyzed by chemiluminescence. The upper band (arrow) represents rabbit Ig heavy chain detected with the secondary anti-rabbit Ab used in the chemiluminescence assay. The lower bands represent the 37-kDa IkBa (double arrow) and the phosphorylated IkBa (arrowhead). B, Immunoprecipitation and Western blot analysis of IkBa from MV-stimulated glial cells with or without pretreatment with CIP as previously described (24). CIP pretreatment almost abolished the intensity of the upper IkBa band indicating that this band represent hyperphosphorylated IkBa. C, Time course of IkBa phosphorylation and degradation in MV-stimulated glial cells. The number at the bottom of each lane in this and in subsequent figures represents the ratio of the density of phosphorylated to hypophosphorylated IkB (IkBp/ IkB) (first row of numbers). The second row represents the ratio of the density of the IkBa band in the treated condition to that in the untreated cells. Zero indicates an undetectable band.
to 120 min, as it could be immunoprecipitated with anti-p65 as shown in Fig. 3A. Identical results were obtained when anti-IkBa Ab was used to immunoprecipitate IkBa (data not shown). Subsequent experiments showed that IkBa was not phosphorylated nor degraded at time points as late as 24 h after MV stimulation (data not shown). However, neuronal cells stimulated with TNF-a showed a hyperphosphorylated IkBa after 5 min and degradation within 30 min (Fig. 3B). The average effects of MV infection or TNF-a stimulation of the neuronal cell line IMR-32 on IkBa expression obtained from three experiments are presented in Fig. 4. Although TNF-a produced significant degradation of IkBa beginning 10 min after stimulation, MV failed to do so at time points varying from 5 to 60 min. To ascertain that these findings are not peculiar to the neuronal cell lines IMR-32, we examined the effects of MV and TNF-a stimulation on IkBa degradation in another neuronal cell line CHP-126. As shown in Fig. 5, TNF-a but not MV stimulation resulted in phosphorylation and degradation of IkBa at time points ranging from 10 to 60 min, which is consistent with the results obtained with the IMR-32 cells.
FIGURE 2. Effect of MV infection of glial cells on IkBa phosphorylation and degradation. The experimental design was identical to that of Fig. 1. IkBa density on Western blots after infection was expressed as a fraction of the IkBa band density corresponding to the uninfected condition. Phosphorylated IkB (IkBp) was expressed as a ratio to IkB in the same lane. The figure represents the mean of three experiments 6 SEs. P values corresponding to the difference between treated and untreated (time 0) conditions are shown in the figure.
Failure to induce IkBa degradation in neuronal cells infected with MPS To determine whether differential IkBa degradation in neuronal and glial cells is MV specific, another neurotropic virus, MPS, was
FIGURE 3. A, Immunoprecipitation and Western blot analysis of IkBa in unstimulated and MV-stimulated neuronal cells for different time periods. IkBa was immunoprecipitated with anti-p65 Ab and probed with antiIkBa Ab. Note the absence of a phosphorylated IkBa band and lack of IkBa degradation. B shows phosphorylation and degradation of neuronal IkBa in response to TNF-a stimulation. IkBa was immunoprecipitated with anti-p65 and probed with anti-IkBa Ab.
The Journal of Immunology
FIGURE 4. Effect of stimulation with MV or TNF-a on IkBa expression in the neuronal cell line IMR-32. The experimental design and quantitation of IkBa is as described above. The results represent mean 6 SEs from three experiments. Changes in IkBa expression after MV infection relative to the uninfected cells were not statistically significant. In contrast, IkBa degradation after TNF-a stimulation was statistically significant beginning at 10 min and at later time points (p , 0.001).
tested in both cell lines. Additional conditions included stimulation with MV, TNF-a, or a combination of MV and TNF-a for comparison. A representative of two experiments is shown in Fig. 6. In all conditions, cells were stimulated for 15 min. All stimuli resulted in IkBa degradation in the glial cell line U-251 MG (Fig. 6A). In contrast, neither MPS nor MV virus resulted in appreciable IkBa degradation in the neuronal cells IMR-32 (Fig. 6B). As observed previously, TNF-a resulted in phosphorylation and degradation of IkBa in the neuronal cell line, which was not inhibited when cells were stimulated with MV and TNF-a simultaneously (Fig. 6B). Failure of dsRNA to induce IkBa degradation in the neuronal cells
4027
FIGURE 6. Differential IkBa degradation in glial and neuronal cells induced by MPS, TNF-a, or a combination of TNF-a and MV. IkBa was immunoprecipitated with anti-p65 Ab and probed with anti-IkBa Ab in the Western blot analysis.
results were obtained with PIPC stimulation for shorter periods of time (1, 2, 4, or 8 h) (data not shown). PKR autophosphorylation in MV-stimulated neuronal and glial cells PKR activation requires autophosphorylation before phosphorylating its target protein (17). Therefore, we examined the effect of MV stimulation on PKR autophosphorylation in neuronal and glial cells. Cells were exposed to MV for 20 h before the assay to allow sufficient time for viral dsRNA formation. Cells were then harvested, and PKR was immunoprecipitated with rabbit anti-PKR Ab followed by the in vitro autophosphorylation assay. A representative of two experiments is shown in Fig. 8. MV stimulation in both cell types resulted in phosphorylation of a 68-kDa protein consistent with the molecular mass of PKR. This band was specifically immunoprecipitated with the anti-PKR, but not the control Ab.
The inability of MV to induce IkBa degradation in neuronal cells may involve a defective kinase activity critical for IkBa phosphorylation such as PKR. PKR is a kinase activated by dsRNA and also by viral RNA and is implicated in the signal cascade leading to IkBa phosphorylation and NF-kB activation (13, 17, 25). Thus, neuronal and glial cells were treated with the dsRNA PIPC for 20 h. As PIPC may not readily penetrate the cell membrane, cells were pretreated with dextran before PIPC stimulation. IkBa phosphorylation and partial degradation were observed in the glial but not in the neuronal cells after exposure to PIPC (Fig. 7). Similar
FIGURE 5. Effect of MV and TNF-a stimulation on IkBa phosphorylation and degradation in the neuronal cell line CHP-126.
FIGURE 7. Effect of the dsRNA PIPC on IkBa phosphorylation in glial and neuronal cells as determined by immunoprecipitation with anti-p65 Ab and immunoblotting with anti-IkBa Ab. The lanes correspond to unstimulated cells, cells treated with MV, PIPC, dextran (Dx), or pretreatment with Dx followed by PIPC.
4028
DIFFERENTIAL NF-kB ACTIVATION IN NEURONAL AND GLIAL CELLS
FIGURE 8. MV-induced PKR autophosphorylation in glial (A) and neuronal (B) cells. Lane 1, Immunoprecipitation with anti-PKR Ab in unstimulated cells; lane 2, immunoprecipitation with anti-PKR Ab in cells stimulated with MV for 20 h; and lane 3, immunoprecipitation with nonimmune rabbit serum in MV-stimulated cells. The numbers at the bottom of each lane represent the ratio of the density of the PKR band relative to that in the untreated condition (lane 1).
Induction of IkBa degradation by UV-inactivated MV The observation that MV stimulation of glial cells resulted in IkBa phosphorylation within 5 min suggested that early signaling events generated by MV binding to its cellular receptor may be involved in NF-kB activation. Thus, UV-inactivated MV was examined for its ability to induce IkBa degradation. Cross-linking viral RNA by UV inhibits viral transcription and replication without affecting receptor binding (our unpublished data). U-251 MG glial cells stimulated with 5 moi of MV or the equivalent UV-MV for 1 h were examined for IkBa degradation. A representative of duplicate experiments is shown in Fig. 9. UV-MV was able to induce IkBa degradation (Fig. 9A) and NF-kB DNA-binding activity (Fig. 9B) with similar efficiency as that of live virus. UV inactivation of MV was ascertained by the absence of live virus in a standard plaque assay. These findings suggested that NF-kB activation by MV may occur as a result of an early signal generated from the binding of the MV to its receptor.
Discussion Neurons, unlike glial cells, are prone to viral persistence, a prototype of which is SSPE caused by a persistent MV infection. In this disease, MV persists predominantly in neurons despite the presence of an intense cellular infiltrate in the central nervous system. Our earlier studies of this disease as well as in the animal model, subacute measles encephalitis (5), indicated a paucity of HLA class I expression on infected neurons but not on glial cells. These findings are supported by the demonstration that neither virus nor dsRNA induce MHC class I in mouse neonatal neuronal cultures (26). Collectively, these observations are consistent with the hypothesis that lack of class I expression on infected neurons may allow escape from immune surveillance by virus-specific CTL (1). Consequently, we have addressed the molecular mechanisms involved in the differential regulation of class I in virus-infected neuronal and glial cells. We have found that HLA class I induction by MV in glial cells was primarily mediated by IFN-b (3). Unlike the noncoordinated expression of IFN-b and MHC class I reported in Sendai virus-infected mouse neonatal neuronal cells (26), the
FIGURE 9. Induction of IkBa degradation (A) and NF-kB/DNA-binding activity (B) by UV-MV. A shows immunoprecipitation of IkBa using anti-p65 Ab from untreated (UnRx), MV-stimulated, and UV-MV-stimulated glial cells U251-MG. B represents a gel shift assay using a 32Plabeled NF-kB probe and nuclear extracts from untreated, MV-stimulated, or UV-MV-stimulated glial cells as described in our previous publication (4).
lack of class I induction in MV-infected neuronal cells correlates with the failure to induce IFN-b (4). As IFN-b has a critical antiviral effect, the absence of both class I and IFN-b in infected neurons is likely to predispose these cells to foster virus persistence. Therefore, understanding the molecular mechanisms responsible for the differential regulation of IFN-b in neuronal and glial cells could be important for the understanding of the mechanisms underlying virus persistence in the central nervous system. NF-kB is a transcription factor involved in the induction of a number of cytokine genes including IFN-b and possibly the upregulation of the HLA class I gene (7– 8). We have previously shown that in contrast to glial cells lack of IFN-b induction in neuronal cells by MV correlates with lack of NF-kB binding to PRDII (5). This does not appear to be due to lack of NF-kB expression in neuronal cells as TNF-a induces NF-kB binding to PRDII in these cells and is not due to lack of MV receptor expression or infectivity of these cells (5). In this study, we demonstrated that the lack of activation of NF-kB in neuronal cells after viral infection correlates with the failure of phosphorylation and degradation of the inhibitor IkBa. This finding is not specific to the IMR-32 neuronal cells, because similar findings were obtained with another neuronal cell line, CHP-126. Information on the cellular defects responsible for the inability of virus to induce IkBa phosphorylation and degradation is critical to understand neuronal responses to viral infection and persistency. Our findings may exclude a number of possibilities. It is unlikely that MV activates a phosphatase that can dephosphorylate IkBa, because MV does not block TNF-a-induced IkBa phosphorylation (Fig. 6). It is also unlikely that neuronal cells contain mutant IkBa isoforms, because TNF-a is able to phosphorylate and degrade IkBa in these cells. Currently, we favor the hypothesis that virus infection of neuronal cells fails to generate a signal pathway required for IkBa phosphorylation and degradation. In signaling, we may also exclude a failure of MV to activate the IkB kinase complex (IKK) or NK-kB inducing kinase (NIK) MAP
The Journal of Immunology kinase. Activation of IKK, a recently described kinase complex that phosphorylates IkBa at the critical serine residues S32 and S36, has been implicated for activation of NF-kB (27–30). IKK, which exists in two identical forms, IKK1 or a (85 kDa) and IKK2 or b (87 kDa), are part of a larger protein complex (;700 kDa) that associates with IkBa. In TNF-a-stimulated cells, this complex is believed to be recruited to the intracellular domain of the TNF-a receptor and phosphorylated by a MAP kinase known as NIK, which is required for TNF-a-dependent NF-kB activation (31–33). Thus, the ability of TNF-a to induce IkBa phosphorylation in the neuronal cells may exclude defects in NIK or IKK in these cells. PKR is a kinase activated by viral RNA and has been implicated in the signaling cascade leading to IkBa phosphorylation (13, 17, 26). As IkBa is not phosphorylated in neuronal cells even after 24 h of exposure to MV, we assayed PKR autophosphorylation at the 20-h time point that corresponds to the duration of the MV replicative cycle. Therefore, the 20-h time point was chosen to ensure the availability of sufficient levels of dsRNA in the infected cells. Our findings that MV was unable to phosphorylate IkBa in neuronal cells despite PKR activation suggests either a failure to activate a signal distal to PKR in the signaling cascade that leads to IkBa phosphorylation or the existence of a PKR-independent pathway used by virus to phosphorylate IkBa. In fact, recent studies in PKR knockout mice support the latter possibility (20). In addition, IkBa phosphorylation in response to MV as early as 5 min poststimulation and the ability of UV-inactivated MV to induce IkBa phosphorylation and NF-kB activation in glial cells indicate that viral transcription, replication and dsRNA formation may not be necessary to induce IkBa phosphorylation by MV. In addition, it is unlikely that PKR is activated in MV-infected cells within 5 min of exposure to virus, because MV nucleocapsid is detectable 4 h postinfection at the earliest (M. Shin, unpublished observation). This finding would suggest the existence of a signaling pathway for MV-induced IkBa phosphorylation that is independent of dsRNA and PKR activation and that this pathway may not be activated by MV in neuronal cells. Our long-term goal is to identify this pathway and to determine whether it is impaired in virus-infected neuronal cells. Our findings have implications regarding the understanding of mechanisms that favor virus persistence in neurons. IFN-b induction in an infected cell is an important host defense mechanism and is dependent on NF-kB activation. Therefore, our finding, implicating an inability of MV to activate the signaling pathway that would normally lead to IkBa phosphorylation and subsequently NF-kB activation, represents a step toward understanding cell-specific events that favor viral persistence in neurons.
4029
6.
7.
8.
9. 10. 11.
12. 13.
14.
15.
16.
17. 18.
19.
20.
21.
22.
23.
24.
25.
Acknowledgments
26.
We thank Dr. Moon Shin for critical review of the manuscript, Ms. Xin Wang and Richard Milanich for technical assistance, and Mrs. Nayereh Dehghan for typing.
27.
References 1. Oldstone, M. B. A. 1989. Viral persistence. Cell 56:517. 2. Joly, E., L. Mucke, and M. B. A. Oldstone. 1991. Viral persistence in neurons explained by lack of major histocompatibility class I expression. Science 253: 1283. 3. Dhib-Jalbut, S., and E. P. Cowan. 1993. Direct evidence that interferon-b mediates enhanced HLA-class 1 expression in measles virus-infected cells. J. Immunol. 51:6248. 4. Dhib-Jalbut, S. S., Q. Xia, P. D. Drew, and P. T. Swoveland. 1995. Differential up-regulation of HLA class I molecules on neuronal and glial cell lines by virus infection correlates with differential induction of IFN-b. J. Immunol. 155:2096. 5. Gogate, N., P. Swoveland, T. Yamabe, L. Verma, J. Woyciechowska, E. Tarnowska-Dziduszko, J. Dymecki, and S. Dhib-Jalbut. 1996. Major histocompatibility complex (MHC)-class I expression on neurons in subacute scle-
28.
29.
30.
31.
32. 33.
rosing panencephalitis (SSPE) and experimental subacute measles encephalitis (SME). J. Neuropathol. Exp. Neurol. 55:435. Maniatis, T., L. A. Whittemore, W. Du, C. M. Fan, A. D. Keller, V. J. Palombella, and D. N. Thanos. 1992. Positive and negative control of human interferon-b gene expression. In Transcriptional Regulation, Part 2. S. L. McKnight and K. Yamamoto, eds. Cold Spring Harbor Laboratory Press, Plainview, NY, p. 1193. Massa, P. T., and H. Wu. 1995. Interferon regulatory factor element and interferon regulatory factor 1 in the induction of major histocompatibility complex class I genes in neural cells. J. Interferon Cytokine Res. 15:799. Drew, P. D, M. Lonergan, M. E. Goldstein, L. A. Lampson, K. Ozato, and D. E. McFarlin. 1993. Regulation of MHC class-I and b2-microglobulin gene expression in human neuronal cells. J. Immunol. 150:3300. Baldwin, A. S., Jr. 1996. The NF-kB and IkB proteins: new discoveries and insights. Annu. Rev. Immunol. 14:649. Finco, T. S., and A. S. Baldwin. 1995. Mechanistic aspects of NF-kB regulation: the emerging role of phosphorylation and proteolysis. Immunity 3:263. Brown, K., S. Gerstberger, L. Carlson, G. Franzoso, and U. Siebenlist. 1995. Control of IkB-a proteolysis by site-specific, signal-induced phosphorylation. Science 267:1485. Rice, N. R., and M. K. Ernst. 1993. In vivo control of NF-kB activation by IkBa. EMBO J. 12:4685. Kumar, A., J. Haque, J. Lacoste, J. Hiscott, amd B. R. G. Williams. 1994. Double-stranded RNA-dependent protein kinase activates transcription factor NF-kB by phosphorylating IkB. Proc. Natl. Acad. Sci. USA 91:6288. Lin, R., P. Beauparlant, C. Makris, S. Meloche, and J. Hiscott. 1996. Phosphorylation of Ikba in the C-terminal PEST domain by casein kinase II affects intrinsic protein stability. Mol. Cell. Biol. 14:1401. Shirakawa, F., and S. B. Mizel. 1989. In vitro activation and nuclear translocation of NF-kB catalyzed by cyclic AMP-dependent protein kinase and protein kinase C. Mol. Cell. Biol. 9:2424. Li, S., and J. M. Sedivy. 1993. Raf-1 protein kinase activates the NF-kB transcription factor by dissociating the cytoplasmic NF-kB-IkB complex. Proc. Natl. Acad. Sci. USA 90:9247. Clemens, M. J., and A. Elia. 1997. The double-stranded RNA-dependent protein kinase PKR: structure and function. J. Interferon Cytokine Res. 17:503. Polyak, S. J., N. Tang, M. Wambach, G. N. Barber, and M. G. Katze. 1996. The P58 cellular inhibitor complexes with the interferon-induced, double-stranded RNA-dependent protein kinase, PKR, to regulate its autophosphorylation and activity. J. Biol. Chem. 271:1702. Gale, M., Jr., S. L. Tan, M. Wambach, and M. G. Katze. 1996. Interaction of the interferon-induced PKR protein kinase with inhibitory proteins P58IPK and vaccinia virus K3L is mediated by unique domains: implications for kinase regulation. Mol. Cell. Biol. 16:4172. Yang, Y.-L., L. F. L. Reis, J. Pavlovic, A. Aguzzi, R. Schafer, A. Kumar, B. R. G. Williams, M. Aguet, and C. Weissmann. 1995. Deficient signaling in mice devoid of double-stranded RNA-dependent protein kinase. EMBO J. 14: 6095. Tumilowicz, J. J., W. W. Nichols, J. J. Cholon, and A. E. Greene. 1970. Definition of a continuous human cell line derived from neuroblastoma. Cancer Res. 30:2110. Lampson, L. A., and C. A. Fisher. 1984. Weak HLA and B2 microglobulin expression in neuronal cell lines can be modulated by interferon. Proc. Natl. Acad. Sci. USA 81:6476. Bigner, D. D., S. H. Bigner, J. Ponten, B. Westermark, M. S. Mahaley, E. Rugslahti, H. Herschman, L. F. Eny, and C. F. Wikstrand. 1981. Heterogeneity of genotypic and phenotypic characteristics of fifteen permanent cell lines derived from human gliomas. J. Neuropathol. Exp. Neurol. 40:210. Miyamoto,S., M. Maki, M. J. Schmitt, M. Hatanaka, and I. M. Verma. 1994. Tumor necrosis factor a-induced phosphorylation of IkBa is a signal for its degradation but not dissociation from NF-kB. Proc. Natl. Acad. Sci. USA 91: 12740. Der, S. D., and A. S. Lau. 1995. Involvement of the double-stranded-RNA-dependent kinase PKR in interferon expression and interferon-mediated antiviral activity. Proc. Natl. Acad. Sci. USA 92:8841. Ward, L. A., and P. T. Massa. 1995. Neuron-specific regulation of major histocompatibility complex class I, interferon-b, and anti-viral state genes. J. Neuroimmunol. 58:145. Regnier, C. H., H. Y. Song, X. Gao, D. V. Goeddel, Z. Cao, and M. Rothe. 1997. Identification and characterization of an IkB kinase. Cell 90:373. DiDonato, J. A., M. Hayakawa, D. M. Rothwarf, E. Zandi, and M. Karin. 1997. A cytokine responsive IkB kinase that activates the transcription factor NF-kB. Nature 388:548. Mercurio, F., H. Zhu, B. W. Murray, A. Shevchenko, B. L. Bennett, J. Li, D. B. Young, M. Barbosa, M. Mann, A. Manning, and A. Rao. 1997. IKK-1 and IKK-2: cytokine-activated IkB kinases essential for NFkB activation. Science 278:860. Woronicz, J. D., X. Gao, Z. Cao, M. Rothe, and D. V. Goeddel. 1997. IkB kinase-b: NF-kB activation and complex formation with IkB kinase-a and NIK. Science 278:866. Malinin, N. L., M. P. Boldin, A. V. Kovalenko, and D. Wallach. 1997. MAP3Krelated kinase involved in NF-kB induction by TNF, CD95 and IL-1. Nature 385:540. Verma, I. M., and J. Stevenson. 1997. IkB kinase: beginning, not the end. Proc. Natl. Acad. Sci. USA 94:11758. Maniatis, T. 1997. Catalysis by a multiprotein IkB kinase complex. Science 278: 818.
