A critical component of the host’s innate immune response involves lipid Ag presentation by CD1d molecules to NK T cells. In this study we used murine CD1d1-transfected L (L-CD1) cells to study the effect of viruses on CD1d-mediated Ag presentation to NKT cells and found that an infection with vesicular stomatitis and vaccinia (but not lymphocytic choriomeningitis) virus inhibited murine CD1d1-mediated Ag presentation. This was under the reciprocal control of the MAPKs, p38 and ERK, and was due to changes in the intracellular trafficking of CD1d1. The reciprocal regulation of CD1d1-mediated Ag presentation by MAPK suggests that the targeting of these pathways is a novel means of immune evasion by viruses.

Classically, studies involving Ag processing and presentation have focused on the role of the MHC class I and II molecules in host defense (1, 2). CD1d molecules are structurally similar to MHC class I molecules in their domain organization and noncovalent association with β2-microglobulin for stable cell surface expression (3, 4). They are expressed in virtually all mammals examined and are restricted mainly to hemopoietic cells (3, 5, 6). Rather than the MHC class I or class II molecules that present peptides to mainstream T lymphocytes, the MHC class I-like CD1d molecule presents lipid Ags to an unique T cell subpopulation called NKT cells (7). NKT cells produce both Th1 and Th2 cytokines and appear to be important in regulating immune responses to infectious diseases and tumors as well as autoimmune diseases (7, 8). The ability of NKT cells to rapidly produce cytokines, activate cells of both the innate and adaptive immune responses, and recognize Ag in the context of CD1d molecules strongly suggests that NKT cells, via their interaction with CD1d, play a pivotal role in antiviral immunity.

Vaccinia virus (VV),3 the prototypic member of the Poxviridae family, replicates in the cytoplasm of infected cells. Its linear dsDNA genome encodes >190 gene products (9). Vesicular stomatitis virus (VSV), the prototypic member of the family Rhabdoviridae, is an enveloped virus with a single-stranded nonsegmented RNA genome that encodes five gene products (10). VSV was an early example of a virus shown to induce the morphological changes and DNA fragmentation associated with apoptosis (11). In contrast to VV and VSV, lymphocytic choriomeningitis virus (LCMV) is relatively noncytopathic. LCMV is a prototypic member of the family Arenaviridae and is an enveloped virus with a bisegmented negative-strand RNA genome that encodes five structural proteins (12). The cellular immune response to LCMV is probably the best characterized and thus very useful for studying viral immunopathogenesis (13).

Cells respond to changes in their environment. Part of this response involves the MAPK-signaling molecules that serve as transducers of extracellular stimuli. There are three distinct MAPK pathways regulated by individual kinases: p38, ERK, and JNK. MAPK pathways are highly evolutionarily conserved and play key roles in many diverse physiological processes, including immunity (14). Cell survival is vital for virus multiplication, and thus, not surprisingly, many viruses modulate the activity of some (or all) MAPK pathways. ERK1/2 plays a vital role in the transmission of both mitogenic and survival signals in response to a variety of extracellular stimuli (15). JNK and p38 function as key mediators of stress and immune signaling in mammals (14, 16).

The present study was undertaken to investigate the roles of acute VSV, VV, and LCMV infections on CD1d1-mediated Ag presentation to NKT cells and to understand the potential role of MAPK in this context. As many viruses use various strategies to circumvent antiviral immunity (17), our studies reveal that viruses can exploit the host cell’s own MAPK pathways as a novel means of evading the innate antiviral immune response.

L-CD1 cells (18, 19) are CD1d1-transfected L cells and were provided by Dr. W. Paul (National Institutes of Health, Bethesda, MD). This cell line was cultured in DMEM supplemented with 10% FBS, 2 mM l-glutamine, and 500 μg/ml G418. Murine LMTK fibroblasts were purchased from American Type Culture Collection and were cultured in DMEM supplemented with 10% FBS and 2 mM l-glutamine. The Vα14+ (canonical) mouse CD1d-specific NKT cell hybridomas, DN32.D3 and N38-2C12, and the Vα5+ (noncanonical) mouse CD1d-specific hybridoma, N37-1A12, have been described (20, 21, 22) and were cultured in IMDM supplemented with 5% FBS, 2 mM l-glutamine, and antibiotics. Purified and biotinylated mAb specific for mouse IL-2 and biotin-labeled rat anti-mouse CD1d mAb (1B1) were purchased from BD Pharmingen. Recombinant mouse IL-2 used as a standard in the ELISAs, was obtained from PeproTech. Texas Red-labeled donkey anti-rat Ig antiserum and FITC-conjugated rat anti-mouse Ig antiserum were purchased from Jackson ImmunoReserach Laboratories. Rat serum was purchased from Sigma-Aldrich. The rat anti-mouse lysosome-associated membrane protein-1 hybridoma, 1D4B was purchased from American Type Culture Collection. The WR strain of VV and the Indiana strain of VSV were provided by Drs. J. Yewdell and J. Bennink (Laboratory of Viral Diseases, National Institute of Arthritis and Infectious Diseases, National Institutes of Health, Bethesda, MD). Virus stocks were propagated and titrated in LMTK cells as previously described (21, 23, 24). VV stocks were propagated and titrated in human 143B osteosarcoma cells as previously described (21, 23). Mock-infected cell lysates were also prepared from 143B cells as for the viral stocks, only without virus. The Armstrong strain of LCMV was provided by Dr. R. Welsh (University of Massachusetts Medical Center, Worcester, MA). The p38-specific inhibitor SB203580 was purchased from Promega, whereas the MEK (ERK pathway)-specific inhibitors U0126 and PD098049 and Abs specific for individual phosphorylated or total MAPK were purchased from Cell Signaling Technology. α-Galactosylceramide (α-GalCer) was synthesized as previously described (25).

To measure endogenous Ag presentation by CD1d1 molecules before and after a virus infection, L-CD1 cells were mock infected or infected with VSV, VV, or LCMV at a multiplicity of infection (MOI) of 5 for the indicated time intervals. The cells were then fixed in 0.05% paraformaldehyde, washed twice in PBS, and cocultured (5 × 105 cells/well) with the NKT cell hybridomas, DN32.D3 (26), N38-2C12, and N37-1A12 (20) (all 5 × 104 cells/well) in triplicate wells in 96-well microtiter plates. After a 20- to 24-h coculture, supernatants were harvested, and IL-2 production was measured by ELISA as previously described (20, 21). To examine the roles of p38 and ERK1/2 in CD1d1-mediated Ag presentation after infection, L-CD1 cells were pretreated with the p38-specific inhibitor SB203580 or the ERK1/2 pathway-specific inhibitor U0126 for 1 h at 37°C. The cells were then washed in ice-cold PBS, infected with VSV (MOI, 1) for 30 min or with VV (MOI, 5) for 1 h in the presence or the absence of the same drug, washed in ice-cold PBS, fixed in 0.05% paraformaldehyde, washed two additional times, and then cocultured with the NKT cell hybridomas as described above, and IL-2 release was measured by ELISA. To examine the effect of a viral infection on the ability of α-GalCer to stimulate NKT cells, L-CD1 cells were uninfected or infected with the indicated virus for 1 h. The cells were then washed with ice-cold PBS and treated with vehicle (DMSO) or α-GalCer (100 ng/ml) for 1 h, extensively washed, fixed, and then cocultured with the indicated NKT cell hybridomas as described above. Production of IL-2 by NKT cells was measured by ELISA.

Aliquots of L-CD1 cells mock infected or infected with the indicated viruses used for the NKT cell coculture experiments were fixed in 1% paraformaldehyde in PBS for 10 min at room temperature, washed twice in PBS, and stained with the anti-mouse CD1d1 mAb (1H6) (21), followed by a PE-conjugated rabbit anti-mouse Ig antiserum (DakoCytomation). Analysis was performed by flow cytometry as previously described (20, 21).

L-CD1 cells were infected with VSV, VV, or LCMV (MOI, 5) for different lengths of time as indicated. The cells were washed in PBS and lysed in 2× SDS sample buffer (4% SDS, 100 mM Tris-HCl (pH 6.8), 20% glycerol, 2% (w/v) 2-ME, and 0.1% bromophenol blue). Equal amounts of protein were then resolved on a 10% SDS-PAGE gel and transferred to a polyvinylidene difluoride membrane (Millipore). The blot was processed using anti-p38 or anti-ERK1/2 Abs specific for the phosphorylated forms and developed using chemiluminescence before exposure on film. The blot was then stripped and reprobed with Abs for the detection of total p38 and ERK1/2. Quantitation of relative band intensity was determined by ChemiImager 4000 software (Alpha Innotech).

L-CD1 cells were pretreated with the p38 (SB203580) or ERK1/2 (U0126) pathway inhibitors (40 μm) for 1 h or infected with the indicated virus for 1 h at 37°C and then the cells were washed and treated with biotin-labeled rat anti-mouse CD1d mAb (1B1) for 30 min on ice. The cells were incubated at 37°C in the presence of the same drug, and at the indicated time points, an aliquot of the cells was placed in ice-cold PBS and kept on ice. After the last time point, the cells were fixed in 1% paraformaldehyde and stained with streptavidin-allophycocyanin. Analysis was performed by flow cytometry as previously described (20, 21).

L-CD1 cells were plated in sterile glass-bottom 35-mm dishes coated with collagen (MatTek) at a density of 1 × 106 cells/dish. After overnight adherence, the cells were infected with VSV, VV, or LCMV at an MOI of 5 or were treated with SB203580, U0126, or PD0980496 at 40 μm for 2 h. The cells were washed twice in PBS and fixed in 4% paraformaldehyde for 10 min at room temperature. Excess paraformaldehyde was quenched using a 10 mM PBS-glycine solution. LAMP-1 staining was performed by incubating the cells with supernatant from the rat anti-mouse LAMP-1-secreting hybridoma, 1D4B, followed by a Texas Red-conjugated donkey anti-rat Ig antiserum. After blocking the free Ab-reactive sites with rat serum (Sigma-Aldrich), immunofluorescent localization of CD1d1 molecules was performed by incubating the cells with the anti-CD1d mouse mAb, 1H6 (21), followed by an FITC-conjugated donkey anti-mouse Ig antiserum. All Abs were diluted and incubated in permeabilizing buffer (HBSS/BSA with 0.1% saponin) for 45 min at room temperature. After each step of Ab incubation, the dishes were washed three times in permeabilizing buffer. The cells were stored in HBSS/BSA in the dark at 4°C until confocal analysis. For analysis, the cells were placed in mounting medium (10 mM Tris and 2% 1-4-diazabicyclo[2.2.2]octane) and viewed with a Bio-Rad MRC-1024 confocal laser-scanning microscope equipped with a krypton-argon laser that had been modified for one-photon microscopy. The FITC and Texas Red emissions were recorded using an oil immersion lens at ×60.

Student’s t test was performed using GraphPad PRISM software (version 3.00 for Windows). A value of p < 0.05 was considered significant. The error bars in the bar graphs show the SD from the mean.

To determine whether a virus infection can alter Ag presentation by CD1d to NKT cells, murine CD1d1-transfected L cell fibroblasts (L-CD1 cells) were infected with VSV, VV, or LCMV for various lengths of time, fixed, and used as target cells in cocultures with a panel of murine NKT cell hybridomas. The production of IL-2 by the supernatant was used as a measure of CD1d1-dependent NKT cell activation (21). As expected, uninfected L-CD1 cells could be recognized by all NKT cell hybridomas (Fig. 1). However, in VSV-infected L-CD1 cells, there was a concomitant decrease in NKT cell activation as the infection progressed, suggesting a virus-induced defect in CD1d1-mediated Ag presentation. This inhibition in Ag presentation could be detected as early as 15 min after infection (data not shown), and this was apparent in both canonical (i.e., Vα14Jα18+) and noncanonical NKT cells (Fig. 1 A). NKT cells did not secrete any cytokines when cocultured with mock-transfected L cells (data not shown).

FIGURE 1.

Kinetics of VSV-, VV-, and LCMV-induced inhibition of CD1d1-mediated Ag presentation to NKT cells. L-CD1 cells were infected with VSV (A), VV (B), and LCMV (C; MOI, 5) for the indicated time periods and cocultured for 20 h with the NKT cell hybridomas, DN32.D3, N38-2C12, and N37-1A12. The culture supernatants were collected for an IL-2 ELISA. The data shown are a percentage of the control values (mock infected = 100%), and each bar is the mean of triplicate cultures ± SD. The data shown are representative of three independent experiments.

FIGURE 1.

Kinetics of VSV-, VV-, and LCMV-induced inhibition of CD1d1-mediated Ag presentation to NKT cells. L-CD1 cells were infected with VSV (A), VV (B), and LCMV (C; MOI, 5) for the indicated time periods and cocultured for 20 h with the NKT cell hybridomas, DN32.D3, N38-2C12, and N37-1A12. The culture supernatants were collected for an IL-2 ELISA. The data shown are a percentage of the control values (mock infected = 100%), and each bar is the mean of triplicate cultures ± SD. The data shown are representative of three independent experiments.

Close modal

When L-CD1 cells were similarly infected with VV, a comparable decrease in Ag presentation by CD1d1 over time was detected (Fig. 1,B). In contrast, NKT cell recognition of LCMV-infected L-CD1 cells was no different from that of uninfected cells (Fig. 1,C). A simple explanation for these results would be that both VSV and VV reduced the cell surface level of CD1d1, because we and others have shown that either of these viruses can cause the down-regulation of MHC class I molecules depending on the time point analyzed (27, 28). Thus, the surface expression of CD1d1 on uninfected or VSV-, VV-, and LCMV-infected L-CD1 cells 2 h after infection was analyzed by flow cytometry. None of these viruses caused a reduction in the cell surface level of CD1d1 (Fig. 2) or MHC class I molecules (data not shown) during the time frames analyzed. It should be noted that the amount of time required to detect a reduction in cell surface MHC class I molecules postinfection was substantially longer than that necessary to alter Ag presentation by CD1d1 (data not shown). Therefore, these data suggest that an infection of L-CD1 cells with VSV and VV (but not LCMV) results in qualitative (rather than quantitative) changes in the functional cell surface expression of CD1d1.

FIGURE 2.

Cell surface CD1d1 expression after viral infection or treatment with MAPK inhibitors. L-CD1 cells were mock infected; infected with VSV, VV, or LCMV; or treated with a p38 (SB203580) or ERK1/2 pathway (U0126) inhibitor and stained with the anti-mouse CD1d1 mAb 1H6 and a PE-conjugated rabbit anti-mouse Ig antiserum. Cell surface CD1d1 expression was analyzed by flow cytometry. □, Isotype control; ▪, anti-CD1d. The data shown are representative of three independent experiments.

FIGURE 2.

Cell surface CD1d1 expression after viral infection or treatment with MAPK inhibitors. L-CD1 cells were mock infected; infected with VSV, VV, or LCMV; or treated with a p38 (SB203580) or ERK1/2 pathway (U0126) inhibitor and stained with the anti-mouse CD1d1 mAb 1H6 and a PE-conjugated rabbit anti-mouse Ig antiserum. Cell surface CD1d1 expression was analyzed by flow cytometry. □, Isotype control; ▪, anti-CD1d. The data shown are representative of three independent experiments.

Close modal

Because we found no change in the cell surface level of CD1d1 after VSV or VV infection, it was unclear why these two viruses inhibited Ag presentation to NKT cells, whereas an LCMV infection did not. It is known that a number of viruses can cause the activation of MAPK (29, 30, 31). It has been shown that p38 and ERK1/2 can play opposing roles in various immune functions (19, 32, 33, 34), but these MAPK have not been investigated in the context of CD1d-mediated Ag presentation. If changes in the activation of specific MAPK by VSV and VV were responsible for the observed inhibition of Ag presentation by CD1d1, then it would be expected that MAPK plays a role in CD1d1-mediated Ag presentation under normal conditions. Thus, to address this question, L-CD1 cells were treated with or without the p38 inhibitor SB203580 or the ERK1/2 pathway (MEK-specific) inhibitor U0126 for 1 h, extensively washed, fixed, and used as targets for a 20-h coculture with NKT cells. As shown in Fig. 3, treatment of L-CD1 cells with the p38 inhibitor actually resulted in an increase in Ag presentation to NKT cells. In contrast, blocking ERK1/2 activation caused a decrease in NKT recognition. It should be noted that the ERK1/2 pathway inhibitor did not alter the cell surface expression of CD1d1, but the p38 inhibitor did reduce CD1d expression slightly (Fig. 2). Thus, p38 and ERK1/2 reciprocally regulate normal CD1d1-mediated Ag presentation, with p38 inhibiting (and ERK1/2 promoting) this activity. To determine whether the inhibition of Ag presentation by CD1d1 to NKT cells after infection with VSV or VV was related to the activation of MAPK, L-CD1 cells were pretreated for 1 h with either p38 or ERK1/2 pathway inhibitors. The cells were then infected with VV or VSV in the presence or the absence of the respective MAPK inhibitors, washed, fixed, and used as targets with NKT cells as described above. Although the p38 inhibitor rescued the virus-induced decrease in CD1d1-mediated Ag presentation, blocking the ERK1/2 pathway actually enhanced the inhibition (Fig. 3). The VV-induced decrease in CD1d-dependent NKT cell activation was significantly rescued by the p38 inhibitor (DN32.D3 (p = 0.0228) and N38-2C12 (p = 0.0017)), but not by the ERK pathway inhibitor. Although not statistically significant, the VSV-induced decrease in CD1d/NKT cell interactions was reproducibly rescued by the p38 inhibitor, whereas the ERK1/2 pathway inhibitor further reduced Ag presentation below that of VSV-infected cells in the presence of vehicle only (Fig. 3; N38-2C12 (p = 0.0021) and N37-1A12 (p = 0.0195)).

FIGURE 3.

Effect of MAPK inhibition on CD1d1-mediated Ag presentation to NKT cells. L-CD1 cells were pretreated with a p38 or ERK1/2 pathway inhibitor, or vehicle. The cells were then infected with VV or VSV in the presence or the absence of the indicated inhibitors and cocultured with NKT cell hybridomas. NKT cell activation was measured by an IL-2 ELISA. The data shown are the mean of triplicate cultures ± SD, given as a percentage of the control (vehicle-treated). ∗, p < 0.05, between the indicated group and virus only. V, vehicle (DMSO); p38i, p38 inhibitor, SB203580; ERKi, MEK (ERK1/2 pathway) inhibitor, U0126. The results are representative of three independent experiments.

FIGURE 3.

Effect of MAPK inhibition on CD1d1-mediated Ag presentation to NKT cells. L-CD1 cells were pretreated with a p38 or ERK1/2 pathway inhibitor, or vehicle. The cells were then infected with VV or VSV in the presence or the absence of the indicated inhibitors and cocultured with NKT cell hybridomas. NKT cell activation was measured by an IL-2 ELISA. The data shown are the mean of triplicate cultures ± SD, given as a percentage of the control (vehicle-treated). ∗, p < 0.05, between the indicated group and virus only. V, vehicle (DMSO); p38i, p38 inhibitor, SB203580; ERKi, MEK (ERK1/2 pathway) inhibitor, U0126. The results are representative of three independent experiments.

Close modal

To better understand how virus-induced alterations in MAPK might affect CD1d-mediated Ag presentation, the levels of activated p38 and ERK1/2 after a virus infection were assessed. L-CD1 cells were infected with VSV, VV, or LCMV for various lengths of time. The cells were then lysed, and the amounts of activated and total p38 and ERK1/2 were determined by Western blot analysis. Infection of L-CD1 cells with either VSV or VV resulted in the rapid activation of p38 (Fig. 4,A). Although phosphorylated p38 could be detected in L-CD1 cells infected with LCMV, no appreciable changes were noticed even up to 4 h after infection (Fig. 4,A). When ERK1/2 levels were analyzed, the opposite of that observed with p38 levels was found with VSV, but a VV infection resulted in the sustained activation of ERK1/2 (Fig. 4,B), with the relative levels of activated p38 higher than ERK for the first hour after infection. This sustained ERK1/2 kinase activity after a VV infection is in agreement with a previous study (35). As in the case of p38, infection with LCMV did not result in the rapid phosphorylation of ERK1/2 (Fig. 4). These results suggest that an infection with VSV stimulates p38 activation and reduced that of ERK1/2. Although VV also quickly activated p38, due to the sustained levels of activated ERK1/2, the kinetics of the inhibition of CD1d1-mediated Ag presentation were slower than VSV (compare Fig. 1, A and B). Thus, altering the balance of activated p38 and ERK1/2 by a virus infection results in a concomitant change in Ag presentation by CD1d. Similarly, the inability of LCMV to affect the activation of either p38 or ERK1/2 is consistent with the lack of a change in CD1d1-mediated Ag presentation after infection with this virus. Taken together, these data suggest that a VSV or VV infection causes the rapid activation of p38, which results in the inhibition of CD1d1-mediated Ag presentation.

FIGURE 4.

Altered activation of p38 and ERK1/2 after virus infection. L-CD1 cells were infected with VSV, VV, or LCMV for the indicated lengths of time and lysed, and then equal amounts of protein were loaded per well for the detection of phosphorylated and total p38 (A) and ERK1/2 (B) expression by Western blot analysis. The relative amounts of phosphorylated p38 and ERK1/2 compared with the total respective proteins were quantitated by densitometry. The results are representative of two independent experiments.

FIGURE 4.

Altered activation of p38 and ERK1/2 after virus infection. L-CD1 cells were infected with VSV, VV, or LCMV for the indicated lengths of time and lysed, and then equal amounts of protein were loaded per well for the detection of phosphorylated and total p38 (A) and ERK1/2 (B) expression by Western blot analysis. The relative amounts of phosphorylated p38 and ERK1/2 compared with the total respective proteins were quantitated by densitometry. The results are representative of two independent experiments.

Close modal

We found that a VSV and VV infection caused an impairment of CD1d1-mediated Ag presentation that was significantly rescued by inhibiting p38 (but not ERK1/2) activity. It was possible that an infection with these viruses altered CD1d1 trafficking, thus interfering with ligand loading. We and others have shown that CD1d traffics to late endocytic compartments for Ag loading before being able to stimulate NKT cells (21). To understand the molecular mechanism(s) used by these viruses in the reduction in CD1d1-mediated Ag presentation, L-CD1 cells were pretreated with p38 (SB203580) or ERK1/2 (U0126) pathway inhibitors, and then the cells were washed and treated with a biotin-labeled rat anti-mouse CD1d mAb. The cells were then incubated at 37°C in the presence of the same drug, and at the indicated time points, an aliquot of the cells was harvested. After the last time point, the cells were fixed and stained with streptavidin-allophycocyanin. The p38 inhibitor reduced the cell surface CD1d1 levels by up to 20% over vehicle-treated cells (Fig. 5,A). This suggested that the enhanced Ag presentation with reduced p38 levels (but with fewer CD1d1 molecules on the cell surface) meant that more CD1d1 molecules were probably trafficking to the appropriate endosomal compartments. This would then result in an elevation in the loading of the CD1d1 molecule with the cognate ligand for NKT cells (7). After treatment with an ERK1/2 pathway (MEK-specific) inhibitor, there was not a decrease in CD1d surface levels (Fig. 5,A), suggesting that CD1d either remained on the cell surface longer or simply did not reach late endocytic compartments. To understand the effect of a virus infection on functional CD1d trafficking, L-CD1 cells were infected with VSV, VV, or LCMV for 1 h at 37°C, then the cell surface expression of CD1d1 was studied as described above. Interestingly, the cell surface level of CD1d1 on VSV-infected cells at various time points was comparable to that on ERK1/2 pathway inhibitor-treated cells, whereas an LCMV infection resulted in a slight (<10%) reduction in surface CD1d levels compared with controls (Fig. 5,B). To our surprise, the level of CD1d1 on the cell surface after VV infection mirrored that after LCMV. This suggested that VSV and VV inhibited Ag presentation by CD1d1 by a different mechanism and/or the effect on this activity was more apparent intracellularly than by quantitative changes at the cell surface. To address this question, L-CD1 cells were treated with p38 (SB203580) or ERK pathway (U0126 and PD098049) inhibitors or were infected with VSV, VV, or LCMV. The intracellular distribution of CD1d1 molecules was then analyzed by confocal microscopy. Compared with vehicle-treated cells, L-CD1 cells treated with the p38 inhibitor showed substantially enhanced colocalization of CD1d1 with LAMP-1, with a high degree of punctate staining in this compartment (Fig. 6). The intracellular staining of CD1d1 in LCMV-infected cells was comparable to that in control cells. Interestingly, however, the CD1d1 molecules in VSV- and VV-infected cells were sequestered to one particular site, almost identical with that observed in cells treated with either of the ERK pathway inhibitors, with very limited colocalization with LAMP-1. These results strongly support our hypothesis that there is impaired intracellular trafficking of CD1d1 molecules upon infection with VSV or VV or when ERK activation is inhibited.

FIGURE 5.

Alterations in CD1d1 trafficking after virus infection or treatment with MAPK inhibitors. L-CD1 cells were pretreated with vehicle (DMSO), p38 (SB203580), or ERK1/2 pathway (U0126) inhibitor (A) or were infected with the indicated virus (B). The cells were then treated with a biotin-labeled rat anti-mouse CD1d mAb, and the cells were incubated at 37°C in the presence of the same drug. At the indicated time points, an aliquot of the cells was taken and stained using streptavidin-allophycocyanin. Cell surface CD1d1 was analyzed by flow cytometry and is shown as a percentage of the vehicle-treated (or mock-infected) L-CD1 cell value.

FIGURE 5.

Alterations in CD1d1 trafficking after virus infection or treatment with MAPK inhibitors. L-CD1 cells were pretreated with vehicle (DMSO), p38 (SB203580), or ERK1/2 pathway (U0126) inhibitor (A) or were infected with the indicated virus (B). The cells were then treated with a biotin-labeled rat anti-mouse CD1d mAb, and the cells were incubated at 37°C in the presence of the same drug. At the indicated time points, an aliquot of the cells was taken and stained using streptavidin-allophycocyanin. Cell surface CD1d1 was analyzed by flow cytometry and is shown as a percentage of the vehicle-treated (or mock-infected) L-CD1 cell value.

Close modal
FIGURE 6.

Intracellular CD1d trafficking is altered by both viruses and MAPK inhibitors. The indicated MAPK inhibitor-treated or virus (VSV, VV, or LCMV)-infected L-CD1 cells were stained with a rat anti-mouse LAMP-1 mAb (1D4B), followed by Texas Red-labeled donkey anti-rat Ig antiserum, and a mouse anti-mouse CD1d mAb (1H6), followed by a FITC-conjugated rat anti-mouse Ig antiserum. Analysis was performed by confocal microscopy. The results are representative of two independent experiments.

FIGURE 6.

Intracellular CD1d trafficking is altered by both viruses and MAPK inhibitors. The indicated MAPK inhibitor-treated or virus (VSV, VV, or LCMV)-infected L-CD1 cells were stained with a rat anti-mouse LAMP-1 mAb (1D4B), followed by Texas Red-labeled donkey anti-rat Ig antiserum, and a mouse anti-mouse CD1d mAb (1H6), followed by a FITC-conjugated rat anti-mouse Ig antiserum. Analysis was performed by confocal microscopy. The results are representative of two independent experiments.

Close modal

To understand whether a viral infection can alter the ability of the synthetic CD1d ligand, α-GalCer to stimulate NKT cells, L-CD1 cells were uninfected or infected with VSV, VV, or LCMV for 1 h; washed; treated with vehicle or α-GalCer for 1 h; extensively washed; fixed; and then used as targets in a 2- h coculture with NKT cells. As shown in Fig. 7 and as expected (36), α-GalCer treatment of uninfected L-CD1 cells substantially increased canonical (e.g., DN32.D3), but not noncanonical (e.g., N37-1A12), NKT cell stimulation. NKT cells cocultured with mock-transfected L cells did not secrete any IL-2 in the presence or the absence of α-GalCer (data not shown). After either VSV or VV infection, α-GalCer partially rescued the virus-induced decrease in CD1d-mediated Ag presentation to DN32.D3. Although not inducing the same level of IL-2 production as that induced by uninfected L-CD1 cells, the enhanced stimulation of DN32.D3 by α-GalCer in VSV- and VV-infected L-CD1 cells increased by a similar factor. Thus, on a relative basis, the α-GalCer-receptive CD1d1 molecules would appear to be comparable in both uninfected and VSV- (or VV)-infected L-CD1 cells, although these viruses altered the qualitative expression of CD1d molecules upon infection. IL-2 production by DN32.D3 or N37-1A12 when cocultured with α-GalCer-treated, LCMV-infected L-CD1 cells was no different from that observed in uninfected cells (Fig. 7). Therefore, these results suggest that in addition to affecting the intracellular trafficking of CD1d molecules in a MAPK-dependent manner (Figs. 5 and 6), VSV and VV (but not LCMV) induce qualitative changes in the cell surface expression of CD1d molecules, as reflected by diminished Ag presentation.

FIGURE 7.

α-GalCer partially rescues the inhibition of CD1d-dependent NKT cell stimulation by VSV or VV infection. L-CD1 cells were infected with VSV, VV, or LCMV, treated with α-GalCer, then cocultured for 20 h with the NKT cell hybridomas, DN32.D3 and N37-1A12. The culture supernatants were collected for an IL-2 ELISA. The data shown are a percentage of the control values (uninfected and vehicle-treated L-CD1 cells = 100%), and each bar is the mean of triplicate cultures ± SD. The left bar in each pair corresponds to vehicle treatment, whereas the right bar corresponds to α-GalCer-treated uninfected or infected L-CD1 cells. The data shown are representative of two independent experiments.

FIGURE 7.

α-GalCer partially rescues the inhibition of CD1d-dependent NKT cell stimulation by VSV or VV infection. L-CD1 cells were infected with VSV, VV, or LCMV, treated with α-GalCer, then cocultured for 20 h with the NKT cell hybridomas, DN32.D3 and N37-1A12. The culture supernatants were collected for an IL-2 ELISA. The data shown are a percentage of the control values (uninfected and vehicle-treated L-CD1 cells = 100%), and each bar is the mean of triplicate cultures ± SD. The left bar in each pair corresponds to vehicle treatment, whereas the right bar corresponds to α-GalCer-treated uninfected or infected L-CD1 cells. The data shown are representative of two independent experiments.

Close modal

CD1d molecules present glycolipid and phospholipid Ags to NKT cells (7), but information about the regulation of Ag presentation by these molecules is limited. CD1d1 molecules traffic to endocytic compartments, where endogenous Ags are probably processed and loaded before presentation by CD1d1 molecules to NKT cells on the cell surface (21). We have found that virus infections result in the loss of a large proportion of canonical (i.e., Vα14Jα18+) NKT cells by apoptosis in an IFN-αβ-dependent process within 2–3 days after infection (37, 38), and this lasts over a long term (39). Recent evidence from our laboratory has demonstrated that within 1 wk after infection with VV or VSV, CD1d1 levels on splenic DCs and macrophages are reduced (40), suggesting that this quantitative change in cell surface CD1d1 could also contribute to the loss of NKT cells in vivo. Alterations in the quantitative and/or qualitative expression of MHC class I molecules is a common mechanism by which certain viruses can evade the host’s antiviral immune response at the level of Ag presentation (17). In the current study, we investigated the effect of a virus infection on Ag presentation by CD1d1 molecules to NKT cells. Within 1 h after infection, a dramatic decrease in NKT recognition with VSV and VV (but not with LCMV) was observed. We also found that C57BL/6 thymocytes became more resistant to recognition by NKT cells after VV (but not LCMV) infection (data not shown). Therefore, the inhibitory effects observed after virus infection are probably due to a qualitative change in CD1d1-mediated Ag presentation.

Certain viruses can induce stress pathways upon infection (41). These major signal transduction pathways modulated by stress involve the MAPK p38, ERK, and JNK (14, 16). A number of viruses have been known to induce p38 MAPK activity (35). Furthermore, VV infection has been shown to activate ERK1/2 and its downstream target, the ATF1 transcription factor (31), but the effect of VV on p38 is unknown. In the current study, we found that in the presence of a p38-specific inhibitor, there was a substantial (albeit not complete) reversal in the reduction of CD1d1-mediated Ag presentation to NKT cells after VSV and VV infection, whereas the opposite was observed using a ERK1/2 pathway inhibitor. In these studies, the JNK pathway did not appear to play an important role in Ag presentation by CD1d (data not shown).

In various systems, p38 is required for cell proliferation and cell differentiation (42). ERK1/2 is known to be activated through the Ras/Raf/MEK pathway in response to various growth factors and to play a central role in cell proliferation (43). The replication of varicella zoster virus is positively regulated by activated p38, and the activation of this stress pathway activates cellular transcription factors, but prevents the activation of cellular defense mechanisms (44). MAPKs (including p38) have been shown to play important roles in the immune response from Th1/Th2 differentiation (14, 19, 45) to the host’s defense against various pathogens, including viruses (19, 46, 47). However, these studies restricted their focus to the effector cells (CD4+ and CD8+ T cells, respectively). A recent report suggested that upon infection of macrophages with Leishmania major, p38 and ERK1/2 had opposite effects on IL-10 and IL-12 production (19), and that both p38 and ERK played a role in actin remodeling in DC (34). However, neither study analyzed the regulation of Ag presentation by MAPK.

In our studies, the levels of activated p38 were increased by ≥2-fold over control values within 15 min and remained high for up to 2 h after VSV and VV infection. This observation also supported our findings that the regulation of p38 activation is critical in CD1d1-mediated Ag presentation to NKT cells. Activated ERK1/2 levels remained lower after VSV infection, suggesting that p38 and ERK1/2 are playing strong reciprocal roles in regulating Ag presentation by CD1d. After VV infection, along with activated p38 in the infected cells, its downstream targets, activation transcription factor-2 and 27-kDa heat shock protein, were found to be phosphorylated (data not shown), with elevated ERK1/2 kinase activity detectable for several hours. When we compared the kinetics of VV- and VSV-induced inhibition of CD1d1-mediated Ag presentation to NKT cells, it was observed that an infection with VV takes longer than that with VSV to reduce CD1d1-dependent NKT cell activation to similar levels, and this could be due to elevated ERK1/2 activation in cells infected with the former virus. LY294002, a specific inhibitor of PI3Ks, including Akt (48), had no effect on CD1d1-mediated Ag presentation under conditions in which Akt activity was blocked (data not shown), suggesting no role for this pathway in our observations.

We and others have shown that newly synthesized CD1d1 molecules travel to the cell surface, re-enter the cell, and traffic to intracellular vesicular compartments, where they acquire endogenous ligands necessary for NKT cell activation (7, 21, 49). The endocytosis of cell surface CD1d1 molecules occurs relatively rapidly, with ∼50% turnover within 30 min (49) (R. Litavecz and R. R. Brutkiewicz, unpublished observations). The mechanisms adapted by VSV and VV to inhibit CD1d-mediated Ag presentation to NKT cells probably act by modulating CD1d trafficking. This was supported by the unusual intracellular distribution pattern coupled with poor colocalization of CD1d1 molecules in LAMP-1+ compartments in both VSV- and VV-infected (and ERK pathway inhibitor-treated) cells, resulting in the reduction of CD1d1-mediated activation of NKT cells. Similar reciprocal regulation of the balance between tumor dormancy and metastatic growth by p38 and ERK has been recently reported to occur in breast, prostate, melanoma, and fibrosarcoma cell lines (50). The exogenous addition of α-GalCer partially rescued the reduced Ag presentation by CD1d upon a VSV or VV infection, but by the same factor as occurred in uninfected L-CD1 cells. This suggests that in addition to altering the intracellular trafficking of CD1d molecules, an infection with these viruses causes qualitative changes in CD1d molecules, and these are not mutually exclusive. We have found that long-term treatment of L-CD1 cells with the glycolipid biosynthesis inhibitor d-1-phenyl-2-hexadecanoylamino-3-morpholino-1-propanol reduced canonical and noncanonical CD1d-dependent stimulation (data not shown). Thus, the ligand recognized by N37-1A12 (if not for other noncanonical NKT cells as well) is glycolipid in nature. This could help explain why stimulation of the noncanonical NKT cell hybridoma N37-1A12 was also reduced after infection of L-CD1 cells with VSV and VV. Additional in-depth studies are essential to dissect in greater detail the molecular mechanisms adapted by these (and probably other) viruses in their evasion of CD1d-dependent immune responses.

We thank W. Paul, J. Yewdell, J. Bennink, D. Lyles, and D. Donner for reagents. E. Wang and K. Dunn provided very helpful advice about confocal microscopy. D. Jay, C. Willard, and K. Gillett provided expert technical assistance.

The authors have no financial conflict of interest.

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 Grants RO1AI46455, CA89026, and POI AI056097 from the National Institutes of Health (to R.R.B.) and National Science Foundation Grant CHE-0194682 (to J.G.-H.). T.J.R.W. was the recipient of a Minority Predoctoral Fellowship from the National Institutes of Health. R.R.B. is a Scholar of the Leukemia and Lymphoma Society.

3

Abbreviations used in this paper: VV, vaccinia virus; α-GalCer, α-galactosylceramide; LCMV, lymphocytic choriomeningitis virus; MOI, multiplicity of infection; ts, temperature sensitive; VSV, vesicular stomatitis virus; WT, wild type.

1
Pamer, E., P. Cresswell.
1998
. Mechanisms of MHC class I-restricted antigen processing.
Annu. Rev. Immunol.
16
:
323
.-358.
2
Watts, C..
1997
. Capture and processing of exogenous antigens for presentation on MHC molecules.
Annu. Rev. Immunol.
15
:
821
.-850.
3
Calabi, F., A. Bradbury.
1991
. The CD1 system.
Tissue Antigens
37
:
1
.-9.
4
Brutkiewicz, R. R., J. R. Bennink, J. W. Yewdell, A. Bendelac.
1995
. TAP-independent, β2-microglobulin-dependent surface expression of functional mouse CD1.1.
J. Exp. Med.
182
:
1913
.-1919.
5
Porcelli, S. A..
1995
. The CD1 family: a third lineage of antigen-presenting molecules.
Adv. Immunol.
59
:
1
.-98.
6
Brossay, L., D. Jullien, S. Cardell, B. C. Sydora, N. Burdin, R. L. Modlin, M. Kronenberg.
1997
. Mouse CD1 is mainly expressed on hemopoietic-derived cells.
J. Immunol.
159
:
1216
.-1224.
7
Brutkiewicz, R. R., Y. Lin, S. Cho, Y. K. Hwang, V. Sriram, T. J. Roberts.
2003
. CD1d-mediated antigen presentation to natural killer T (NKT) cells.
Crit. Rev. Immunol.
23
:
403
.-419.
8
Porcelli, S. A., R. L. Modlin.
1999
. The CD1 system: antigen-presenting molecules for T cell recognition of lipids and glycolipids.
Annu. Rev. Immunol.
17
:
297
.-329.
9
Symons, J. A., A. Alcami, G. L. Smith.
1995
. Vaccina virus encodes a soluble type I interferon receptor of novel structure and broad species specificity.
Cell
81
:
551
.-560.
10
Balachandran, S., G. N. Barber.
2004
. Defective translational control facilitates vesicular stomatitis virus oncolysis.
Cancer Cell
5
:
51
.-65.
11
Koyama, A. H..
1995
. Induction of apoptotic DNA fragmentation by the infection of vesiclar stomatitis virus.
Virus Res.
37
:
285
.-290.
12
Lee, K. J., M. Perez, D. D. Pinschewer, J. C. de la Torre.
2002
. Identification of the lymphocytic choriomeningitis virus (LCMV) proteins required to rescue LCMV RNA analogs into LCMV-like particles.
J. Virol.
76
:
6393
.-6397.
13
Selin, L. K., R. M. Welsh.
2004
. Plasticity of T cell memory responses to viruses.
Immunity
20
:
5
.-16.
14
Dong, C., R. J. Davis, R. A. Flavell.
2002
. MAP kinases in the immune response.
Annu. Rev. Immunol.
20
:
55
.-72.
15
Sears, R. C., J. R. Nevins.
2002
. Signaling networks that link cell proliferation and cell fate.
J. Biol. Chem.
277
:
11617
.-11620.
16
Kyriakis, J. M., J. Avruch.
2001
. Mammalian mitogen-activated protein kinase signal transduction pathways activated by stress and inflammation.
Physiol. Rev.
81
:
807
.-869.
17
Tortorella, D., B. E. Gewurz, M. H. Furman, D. J. Schust, H. L. Ploegh.
2000
. Viral subversion of the immune system.
Annu. Rev. Immunol.
18
:
861
.-926.
18
Chen, H., W. E. Paul.
1997
. Cultured NK1.1+ CD4+ T cells produce large amounts of IL-4 and IFN-γ upon activation by anti-CD3 or CD1.
J. Immunol.
159
:
2240
.-2249.
19
Mathur, R. K., A. Awasthi, P. Wadhone, B. Ramanamurthy, B. Saha.
2004
. Reciprocal CD40 signals through p38MAPK and ERK-1/2 induce counteracting immune responses.
Nat. Med.
10
:
540
.-544.
20
Burdin, N., L. Brossay, Y. Koezuka, S. T. Smiley, M. J. Grusby, M. Gui, M. Taniguchi, K. Hayakawa, M. Kronenberg.
1998
. Selective ability of mouse CD1 to present glycolipids: α-galactosylceramide specifically stimulates Va14+ NK T lymphocytes.
J. Immunol.
161
:
3271
.-3281.
21
Roberts, T. J., V. Sriram, P. M. Spence, M. Gui, K. Hayakawa, I. Bacik, J. R. Bennink, J. W. Yewdell, R. R. Brutkiewicz.
2002
. Recycling CD1d1 molecules present endogenous antigens processed in an endocytic compartment to NKT cells.
J. Immunol.
168
:
5409
.-5414.
22
Lantz, O., A. Bendelac.
1994
. An invariant T cell receptor a chain is used by a unique subset of major histocompatibility complex class I-specific CD4+ and CD48 T cells in mice and humans.
J. Exp. Med.
180
:
1097
.-1106.
23
Bacik, I., J. H. Cox, R. Anderson, J. W. Yewdell, J. R. Bennink.
1994
. TAP (transporter associated with antigen processing)-independent presentation of endogenously synthesized peptides is enhanced by endoplasmic reticulum insertion sequences located at the amino- but not carboxyl-terminus of the peptide.
J. Immunol.
152
:
381
.-387.
24
Lyles, D. S., M. O. McKenzie, M. Ahmed, S. C. Woolwine.
1996
. Potency of wild-type and temperature-sensitive vesicular stomatitis virus matrix protein in the inhibition of host-directed gene expression.
Virology
225
:
172
.-178.
25
Du, W., J. Gervay-Hague.
2005
. Efficient synthesis of a-galactosyl ceramide analogues using glycosyl iodide donors.
Org. Lett.
7
:
2063
.-2065.
26
Bendelac, A., O. Lantz, M. E. Quimby, J. W. Yewdell, J. R. Bennink, R. R. Brutkiewicz.
1995
. CD1 recognition by mouse NK1+ T lymphocytes.
Science
268
:
863
.-865.
27
Hecht, T. T., D. F. Summers.
1972
. Effect of vesicular stomatitis virus infection on the histocompatibility antigen of L cells.
J. Virol.
10
:
578
.-585.
28
Brutkiewicz, R. R., S. J. Klaus, R. M. Welsh.
1992
. Window of vulnerability of vaccinia virus-infected cells to natural killer (NK) cell-mediated cytolysis correlates with enhanced NK cell triggering and is concomitant with a decrease in H-2 class I antigen expression.
Nat. Immun.
11
:
203
.-214.
29
Lee, S., C. Tarn, W. H. Wang, S. Chen, R. L. Hullinger, O. M. Andrisani.
2002
. Hepatitis B virus X protein differentially regulates cell cycle progression in X-transforming versus nontransforming hepatocyte (AML12) cell lines.
J. Biol. Chem.
277
:
8730
.-8740.
30
Yang, X., D. Gabuzda.
1998
. Mitogen-activated protein kinase phosphorylates and regulates the HIV-1 Vif protein.
J. Biol. Chem.
273
:
29879
.-29887.
31
de Magalhaes, J. C., A. A. Andrade, P. N. Silva, L. P. Sousa, C. Ropert, P. C. Ferreira, E. G. Kroon, R. T. Gazzinelli, C. A. Bonjardim.
2001
. A mitogenic signal triggered at an early stage of vaccinia virus infection: implication of MEK/ERK and protein kinase A in virus multiplication.
J. Biol. Chem.
276
:
38353
.-38360.
32
Zhu, J., G. Krishnegowda, D. C. Gowda.
2005
. Induction of proinflammatory responses in macrophages by the glycosylphosphatidylinositols of Plasmodium falciparum: the requirement of extracellular signal-regulated kinase, p38, c-Jun N-terminal kinase and NF-κB pathways for the expression of proinflammatory cytokines and nitric oxide.
J. Biol. Chem.
280
:
8617
.-8627.
33
Choudhry, M. A., X. Ren, A. Romero, E. J. Kovacs, R. L. Gamelli, M. M. Sayeed.
2004
. Combined alcohol and burn injury differentially regulate P-38 and ERK activation in mesenteric lymph node T cell.
J. Surg. Res.
121
:
62
.-68.
34
West, M. A., R. P. Wallin, S. P. Matthews, H. G. Svensson, R. Zaru, H. G. Ljunggren, A. R. Prescott, C. Watts.
2004
. Enhanced dendritic cell antigen capture via toll-like receptor-induced actin remodeling.
Science
305
:
1153
.-1157.
35
Andrade, A. A., P. N. Silva, A. C. Pereira, L. P. De Sousa, P. C. Ferreira, R. T. Gazzinelli, E. G. Kroon, C. Ropert, C. A. Bonjardim.
2004
. The vaccinia virus-stimulated mitogen-activated protein kinase (MAPK) pathway is required for virus multiplication.
Biochem. J.
381
:
437
.-446.
36
Sriram, V., W. Du, J. Gervay-Hague, R. R. Brutkiewicz.
2005
. Cell wall glycosphingolipids of Sphingomonas paucimobilis are CD1d-specific ligands for NKT cells.
Eur. J. Immunol.
35
:
1692
.-1701.
37
Hobbs, J. A., S. Cho, T. J. Roberts, V. Sriram, J. Zhang, M. Xu, R. R. Brutkiewicz.
2001
. Selective loss of natural killer T cells by apoptosis following infection with lymphocytic choriomeningitis virus.
J. Virol.
75
:
10746
.-10754.
38
Kronenberg, M., L. Gapin.
2002
. The unconventional lifestyle of NKT cells.
Nat. Rev. Immunol.
2
:
557
.-568.
39
Lin, Y., T. J. Roberts, C. R. Wang, S. Cho, R. R. Brutkiewicz.
2005
. Long-term loss of canonical NKT cells following an acute virus infection.
Eur. J. Immunol.
35
:
879
.-889.
40
Lin, Y., T. J. Roberts, P. M. Spence, R. R. Brutkiewicz.
2005
. Reduction in CD1d expression on dendritic cells and macrophages by an acute virus infection.
J. Leukocyte Biol.
77
:
151
.-158.
41
Koga, T., A. Wand-Wurttenberger, J. DeBruyn, M. E. Munk, B. Schoel, S. H. Kaufmann.
1989
. T cells against a bacterial heat shock protein recognize stressed macrophages.
Science
245
:
1112
.-1125.
42
Nagata, Y., N. Takahashi, R. J. Davis, K. Todokoro.
1998
. Activation of p38 MAP kinase and JNK but not ERK is required for erythropoietin-induced erythroid differentiation.
Blood
92
:
1859
.-1869.
43
O’Gorman, D. M., T. G. Cotter.
2001
. Molecular signals in anti-apoptotic survival pathways.
Leukemia
15
:
21
.-34.
44
Rahaus, M., N. Desloges, M. H. Wolff.
2004
. Replication of varicella-zoster virus is influenced by the levels of JNK/SAPK and p38/MAPK activation.
J. Gen. Virol.
85
:
3529
.-3540.
45
Zhang, S., M. H. Kaplan.
2000
. The p38 mitogen-activated protein kinase is required for IL-12-induced IFN-γ expression.
J. Immunol.
165
:
1374
.-1380.
46
Conze, D., J. Lumsden, H. Enslen, R. J. Davis, G. Le Gros, M. Rincon.
2000
. Activation of p38 MAP kinase in T cells facilitates the immune response to the influenza virus.
Mol. Immunol.
37
:
503
.-513.
47
Conze, D., T. Krahl, N. Kennedy, L. Weiss, J. Lumsden, P. Hess, R. A. Flavell, G. Le Gros, R. J. Davis, M. Rincon.
2002
. c-Jun NH2-terminal kinase (JNK)1 and JNK2 have distinct roles in CD8+ T cell activation.
J. Exp. Med.
195
:
811
.-823.
48
King, W. G., M. D. Mattaliano, T. O. Chan, P. N. Tsichlis, J. S. Brugge.
1997
. Phosphatidylinositol 3-kinase is required for integrin-stimulated AKT and Raf-1/mitogen-activated protein kinase pathway activation.
Mol. Cell. Biol.
17
:
4406
.-4418.
49
Jayawardena-Wolf, J., K. Benlagha, Y. H. Chiu, R. Mehr, A. Bendelac.
2001
. CD1d endosomal trafficking is independently regulated by an intrinsic CD1d-encoded tyrosine motif and by the invariant chain.
Immunity
15
:
897
.-908.
50
Aguirre-Ghiso, J. A., Y. Estrada, D. Liu, L. Ossowski.
2003
. ERK(MAPK) activity as a determinant of tumor growth and dormancy; regulation by p38 (SAPK).
Cancer Res.
63
:
1684
.-1695.