Activation of APCs via TLRs leads to activation of NF-κB, a key transcription factor in cells of the immune system most often associated with induction of Th1-type and proinflammatory responses. The neoglycoconjugate lacto-N-fucopentaose III (12-25 molecules)-dextran (LNFPIII-Dex) activates dendritic cells (DCs) via TLR4, as does LPS. However, unlike LPS, LNFPIII-Dex-activated cells induce Th2-type CD4+ T cell responses. This observation led us to ask whether LNFPIII-activated APCs were differentially activating NF-κB, and if so, could this partly account for how DCs mature in response to these two different pathogen-associated molecular patterns (PAMPs). In this study, we show that LNFPIII-Dex stimulation of APCs induces rapid, but transient NF-κB translocation and activity in the nucleus, in comparison with the persistent activation induced by LPS. We then demonstrate that transient vs persistent NF-κB activation has important implications in the development of the APC phenotype, showing that the second wave of NF-κB translocation in response to LPS is required for production of the proinflammatory mediator NO. In contrast to LPS, LNFPIII-stimulated APCs that only transiently activate NF-κB do not induce degradation of the known IκB family members or production of NO. However, cells stimulated with LNFPIII rapidly accumulate p50, suggesting that an alternative p105 degradation-dependent mechanism is primarily responsible for NF-κB activation downstream of LNFPIII. Finally, we show that while NF-κB translocation in LNFPIII-stimulated APCs is transient, it is required for the development of the DC 2 phenotype, confirming a crucial and multifaceted role for NF-κB in innate immune responses.

Nuclear factor κB plays a pivotal role in developmental and immunological processes (1). Upon cell activation by a variety of exogenous stimuli such as TLR engagement or cytokine stimulation, a cascade of signaling proceeds from the cell surface, converging upon the activation by phosphorylation of the IκB kinase complex, which then phosphorylates IκB, allowing it to be ubiquitinated, thus targeting it for degradation by the proteasome. This releases NF-κB subunits, allowing access to the nuclear localization signals that had previously been blocked. The NF-κB complex can now be transported into the nucleus, where it can activate a number of target genes (2).

The mammalian NF-κB family includes RelA (p65), RelB, and c-Rel, as well as p50 and p52 (and their precursors p105 and p100, respectively) (3). NF-κB exists as preformed homo- or heterodimers (RelB is the only subunit that cannot form homodimers). The most common observed complex is p65 with p50. Knockouts of all of the subunits have been achieved, including double knockouts, some of which are embryonically lethal (4).

Our investigation of NF-κB function and regulation evolved from our studies on activation of APCs with the Schistosoma mansoni-related glycan lacto-N-fucopentaose III (LNFPIII).4 We have shown that LNFPIII is a potent inducer of Th2 responses both in vivo and in vitro (5) and recently reported that LNFPIII directly activates dendritic cells (DCs), inducing their maturation to a DC2-type phenotype capable of driving naive T cells to differentiate into Th2 cells (6). Interestingly, the ability of LNFPIII to activate DCs was dependent upon TLR4. This result was surprising as the TLR family in general, and TLR4 in particular, is generally associated with strong proinflammatory and type I responses, rather than type 2 and anti-inflammatory responses. The ligands for the TLR family that have been identified are mostly bacterial lipid and protein products known to be strong type I drivers. LNFPIII was one of the first purely carbohydrate moieties shown to signal through a TLR as well as one of the most finely characterized, because the nonfucosylated homologue of LNFPIII, lacto-N-neotetraoase (LNnT), does not signal through TLR4, establishing the necessity of the fucose residue for activity. LNFPIII contains the asialo and asulfo Lewis X trisaccharide, and Dissanayake et al. (7) recently reported that Lewis X activates APCs via TLR4, corroborating our findings. Interestingly, although not yet shown to be a TLR4-dependent pathway, Helicobacter pylori, which express Lewis X on their LPS, induce Th2 responses, whereas those bacteria that are Lewis negative drive proinflammatory responses (8).

NF-κB activation is generally considered one of the primary targets of TLR engagement and has been associated with proinflammatory and Th1-type responses. Because LNFPIII and LPS both activate APCs in a TLR4-dependent process, yet drive distinct APC maturational processes, we decided to investigate whether differences in patterns of NF-κB activation downstream of TLR4 activation could be detected in APCs stimulated with a neoglycoconjugate composed of multiple molecules of LNFPIII conjugated to dextran (LNFPIII-Dex) vs cells stimulated with LPS. We found that LNFPIII-Dex stimulation of APCs induced NF-κB translocation but in a transient manner vs the persistent activation observed for LPS. Furthermore, we show that while LNFPIII-Dex does induce transient translocation of NF-κB, this is not accompanied by any observable degradation of IκB family members. However, LNFPIII-Dex-stimulated cells did accumulate p50, apparently as the result of the processing of p105. We then examined how transient vs persistent activation of NF-κB could alter APC phenotype and found that persistent activation of NF-κB was required for APC production of the proinflammatory mediator NO, and, in addition, that even though LNFPIII-Dex only induces transient activation of NF-κB in APCs, this level is required for maturation of DC2s.

LNFPIII and LNnT were separately conjugated to 40-kDa molecules of dextran (LNFPIII-Dex, LNnT-Dex) and were obtained along with dextran from Neose Technologies. The glycoconjugate consisted of ∼12–25 LNFPIII or LNnT molecules/molecule of dextran, as described previously (6). We have observed that the sugar substitution on the dextran or human serum albumin (HSA) carriers is critical for the ability to activate APCs. Based on our data, it appears that a sugar to carrier ratio minimally of 10:1 is required for activation of APCs. Unfortunately, LNFPIII conjugates obtained from most commercial suppliers have a hapten to carrier ratio of 4:6, below the activation threshold that we have observed.

Six- to 8-wk-old female BALB/c, 129P2-Nfkb1tm 1 Bal/J, 129PF2/J, and DO.11.10αβ TCR transgenic mice were purchased from The Jackson Laboratory and maintained under specific pathogen-free conditions in the animal facility at the Harvard School of Public Health following institutional guidelines.

The RAW264.7 cell line was obtained from the American Type Culture Collection and cultured in complete medium, which consists of DMEM supplemented with 10% FCS (HyClone), 100 U/ml penicillin, 100 μg/ml streptomycin, 0.05 mM 2-ME, and 2 mM glutamine (Sigma-Aldrich). Cells were split every 2 days, maintaining a density of ∼5 × 105 cells/ml.

Bone marrow (BM)-derived DCs were produced and treated, as described, from DO.11.10αβ TCR, NF-κB1-deficient, and wild-type mice. DCs were isolated, as previously described (1). Briefly, PBS was used to flush BM from femurs and tibias, using a 0.45-mm-diameter needle. BM cells were inoculated into bacteriological petri dishes at 2 × 106 total cells in 10 ml of complete medium (see above). A total of 20 ng/ml murine rGM-CSF (5 × 106 U/mg; PeproTech) was included in BM leukocyte culture on days 0, 3, and 6. A total of 10 ng/ml murine rGM-CSF was added to cultures on day 8. On day 6, CD11c+ cells were purified via MACS from BM leukocyte culture and plated at 5 × 104 cells/well in 96-well plates. On days 6 and 8, DCs were treated with LPS, LNFPIII, LNnT, or an equal volume of PBS. On day 10,200 μl of fresh complete medium containing 2 × 105 MAC-purified naive, splenic CD4+ T cells was added to wells containing DCs. Supernatants were harvested from cocultures 72 h postaddition of CD4+ T cells. For experiments using CD4 cells from DO.11.10αβ TCR transgenic mice, DCs cells were obtained from congenic mice and added with 10 μM OVA peptide (323-ISQAVHAAHAEINEAGR-339). SN50 and SN50M (Calbiochem) were added on days 6 and 8 of DC culture, 30 min before addition of LNFPIII-Dex or LPS. Mixed lymphocyte reactions were established using DCs derived from either wild-type (B6; 129PF2/J) or p105 knockout mice (B6; 129P2-Nfkb1tm 1 Bal/J) and allogeneic CD4+ T cells from naive BALB/c mice. In Fig. 5, cycloheximide was used at 10 μg/ml.

FIGURE 5.

A NF-κB inhibitor blocks DC2 differentiation by LNFPIII-Dex. IL-4 (▪) and IFN-γ (▦) production of OVA-specific (D0.11) CD4+ T cells cocultured for 72 h with DCs matured under control conditions with or without NF-κB inhibitor (SN50, 18 nM) or LNFPIII-Dex (50 μg/ml) with or without NF-κB inhibitor. Asterisks indicate p < 0.05 by ANOVA.

FIGURE 5.

A NF-κB inhibitor blocks DC2 differentiation by LNFPIII-Dex. IL-4 (▪) and IFN-γ (▦) production of OVA-specific (D0.11) CD4+ T cells cocultured for 72 h with DCs matured under control conditions with or without NF-κB inhibitor (SN50, 18 nM) or LNFPIII-Dex (50 μg/ml) with or without NF-κB inhibitor. Asterisks indicate p < 0.05 by ANOVA.

Close modal

Nitrite was measured in culture supernatants using Greiss reagent. A total of 50 μl of supernatants was mixed with 100 μl of Greiss reagent and incubated for 5 min at room temperature. Absorbance was read at 550 nm using a Spectramax plate reader (Molecular Devices). Nitrite concentrations were determined by comparing absorbance values of the test samples with a standard curve generated by serial dilution of 62.5 μM sodium nitrite.

Nuclear extracts of RAW264.7 cells were prepared by washing cells twice with PBS and then harvesting in 400 μl of buffer A (10 mM HEPES (pH 7.9), 10 mM KCl, 0.1 mM EDTA, 0.1 mM EGTA, 0.1 mM DTT, and 0.5 mM PMSF). Cells were incubated for 15 min on ice, followed by addition of 25 μl of 10% Nonidet P-40 (Calbiochem). Extracts were spun for 1 min at maximum speed in a microcentrifuge, producing a nuclear pellet. Nuclear pellets were extracted in 50 μl of buffer B (20 mM HEPES (pH 7.9), 400 mM KCl, 1 mM EDTA, 1 mM EGTA, 0.1 mM DTT, and 1 mM PMSF) for 30 min on ice, followed by a 5-min spin at maximum speed. Nuclear extracts were aliquoted and stored at −80°C until use. For DC extracts, a similar protocol as above was followed using CellLytic NuCLEAR extraction kit (Sigma-Aldrich). For gel shifts, the DNA consensus oligonucleotide for NF-κB was purchased from Promega and 32P labeled using T4 polynucleotide kinase. Extracts were incubated for 10 min in Gel Shift Binding Buffer (Promega), followed by 20-min incubation with radiolabeled probe. Reactions were resolved on a 6% DNA retardation gel (Invitrogen Life Technologies) at 300 V and were subsequently dried and exposed to autoradiography film or captured using a model 400B phosphorimager and ImageQuant software (Amersham Biosciences).

RAW264.7 cells grown to 50% confluence were transfected with an NF-κB-luciferase reporter construct (a generous gift from Z.-M. Yuan, Harvard School of Public Health, Boston, MA) using the Lipofectamine Plus Reagent (Invitrogen Life Technologies), and incubated at 37°C for 3 h. The cells were allowed to rest overnight, and then were plated on a 48-well plate. They were stimulated with 50 μg/ml Dex, LNnT-Dex, LNFPIII-Dex, or 1 μg/ml LPS and incubated for 4 h. The cells were washed with PBS and lysed with 100 μl of passive lysis buffer (10 mM HEPES (pH 7.9), 10 mM KCl, 0.1 mM EDTA, and 0.1 mM EGTA) for 15 min. A total of 20 μl of the lysate was mixed with 100 μl of Luciferase Reagent (Promega), and activity was measured on a luminometer.

IL-4 (pg/ml) and IFN-γ (ng/ml) (BD Pharmingen) sandwich ELISAs were performed using paired mAbs, as previously described (6).

Anti-ERK Abs were purchased from Cell Signaling Technology. Anti-IκBα, IκBβ, IκBε Abs, and p105 Abs were purchased from Santa Cruz Biotechnology. The p105 Ab binds to a sequence within the p50 portion of the molecule. Western blots were performed using standard methodology and as described previously (9). Cellular extracts were run on a 12% Tris-HCl-SDS polyacrylamide gel (Bio-Rad). Proteins were transferred onto polyvinylidene difluoride membranes (Millipore) and blocked in 5% milk-TBS for 1 h at room temperature. Ab binding was performed, according to the manufacturer’s recommendations.

Data were analyzed for significance using Student’s t test or ANOVA software in Microsoft Excel. Data are expressed as the mean ± SE. Significant differences (p < 0.05) are indicated with an asterisk.

To determine whether LNFPIII-Dex stimulation of APCs induced NF-κB activation, we looked for the appearance of NF-κB complexes in the nucleus by EMSA after stimulation of RAW264.7 cells or BM-derived DCs with LNFPIII-Dex or LPS. We found that at 30 min poststimulation, both LNFPIII-Dex and LPS induced an up-regulation of NF-κB-binding activity in the nucleus in both cell populations (Fig. 1, a and b). The levels of NF-κB in the nucleus after LPS stimulation are known to drop at 4 h poststimulation in RAW264.7 cells, probably due to the activity of IκBα (10). At 6 h poststimulation in RAW264.7 cells and 10 h poststimulation in DCs, we observed the second phase of LPS-induced NF-κB binding, which was not seen in samples from LNFPIII-Dex-induced cells (Fig. 1, a and b).

FIGURE 1.

LNFPIII-Dex induces the translocation of NF-κB to the nucleus. RAW264.7 cells (a) or BM-derived DCs (b) were treated with 50 μg/ml dextran (lane 1), LNnT-Dex (lane 2), LNFPIII-Dex (lane 3), or 1 μg/ml LPS (lane 4). Nuclear extracts were prepared at the indicated time points and tested for binding with a radiolabeled NF-κB consensus oligonucleotide. Data are representative of at least three independent experiments (c). RAW264.7 cells were transfected with an NF-κB luciferase reporter construct, rested overnight, and treated with 50 μg/ml dextran (Dex), LNnT-Dex (NT), LNFPIII-Dex (F3), or 1 μg/ml LPS for 3 h. Luciferase activity was detected in cellular extracts. Data are representative of at least three independent experiments.

FIGURE 1.

LNFPIII-Dex induces the translocation of NF-κB to the nucleus. RAW264.7 cells (a) or BM-derived DCs (b) were treated with 50 μg/ml dextran (lane 1), LNnT-Dex (lane 2), LNFPIII-Dex (lane 3), or 1 μg/ml LPS (lane 4). Nuclear extracts were prepared at the indicated time points and tested for binding with a radiolabeled NF-κB consensus oligonucleotide. Data are representative of at least three independent experiments (c). RAW264.7 cells were transfected with an NF-κB luciferase reporter construct, rested overnight, and treated with 50 μg/ml dextran (Dex), LNnT-Dex (NT), LNFPIII-Dex (F3), or 1 μg/ml LPS for 3 h. Luciferase activity was detected in cellular extracts. Data are representative of at least three independent experiments.

Close modal

We also tested the activity of an NF-κB luciferase reporter plasmid in response to LPS and LNFPIII-Dex. As can be seen in Fig. 1 c, LNFPIII-Dex induced a ∼4-fold increase in luciferase activity vs a 10-fold increase by LPS. The lower level of activation measured in this study with LNFPIII-Dex is consistent with other outputs of LNFPIII-Dex activity on RAW264.7 cells measured in this study, which, while lower than LPS, are considerably higher than background levels.

LPS stimulation leads to the induction of a variety of inflammatory and other immunological mediators downstream of NF-κB such as NO, via the induction of inducible NO synthase (iNOS). The RAW264.7 cell line has been heavily studied in terms of the mechanisms involved in the induction of iNOS (11, 12, 13). In contrast to LPS stimulation of APCs, we found that LNFPIII-Dex was unable to induce any measurable level of NO (Fig. 2,a) even when used at 500 μg/ml, 100 times the minimum concentration required for LNFPIII-Dex to induce MIP-1α in RAW 264.7 cells (data not shown). In addition, LNFPIII-Dex did not induce IL-10 (data not shown), another product that is readily induced in the RAW264.7 cell line by LPS. LNFPIII-Dex also did not induce any NO in DC cultures at 12, 24, and 48 h (Fig. 2 b, and data not shown), despite high levels of NO produced from these cells in response to LPS.

FIGURE 2.

LNFPIII-Dex does not induce NO production. RAW264.7 cells (a) or BM-derived DCs (b) were treated with 50 μg/ml dextran (Dex), LNnT-Dex, LNFPIII-Dex, or 1 μg/ml LPS, as indicated, for 24 (a) or 48 h (b), following which supernatants were analyzed for nitrite using Greiss reagent. c, RAW264.7 cells were treated with LPS for 30 min, washed, and rested for 3.5 h. The SN50 inhibitor was added at a concentration of 50 μg/ml, and cells were incubated for 24 h before nitrite analysis using the Greiss reagent. Asterisks indicate p < 0.05 by Student’s t test. d, The SN50 inhibitor does not inhibit AP-1 activation. Cells were incubated with SN50 for 30 min and then stimulated with 1 μg/ml LPS. Nuclear extracts were analyzed by EMSA for NF-κB or AP-1 translocation, and band intensities were quantified using Kodak 1D software. Representative of at least three independent experiments.

FIGURE 2.

LNFPIII-Dex does not induce NO production. RAW264.7 cells (a) or BM-derived DCs (b) were treated with 50 μg/ml dextran (Dex), LNnT-Dex, LNFPIII-Dex, or 1 μg/ml LPS, as indicated, for 24 (a) or 48 h (b), following which supernatants were analyzed for nitrite using Greiss reagent. c, RAW264.7 cells were treated with LPS for 30 min, washed, and rested for 3.5 h. The SN50 inhibitor was added at a concentration of 50 μg/ml, and cells were incubated for 24 h before nitrite analysis using the Greiss reagent. Asterisks indicate p < 0.05 by Student’s t test. d, The SN50 inhibitor does not inhibit AP-1 activation. Cells were incubated with SN50 for 30 min and then stimulated with 1 μg/ml LPS. Nuclear extracts were analyzed by EMSA for NF-κB or AP-1 translocation, and band intensities were quantified using Kodak 1D software. Representative of at least three independent experiments.

Close modal

Because we observed two phases of NF-κB activation in response to LPS, but not in response to LNFPIII-Dex, we decided to determine whether this difference was relevant to the ability of LPS or LNFPIII-Dex to induce functional changes in APCs. To study this question, we used the cell-permeable NF-κB inhibitor, SN50, which contains the NF-κB nuclear localization sequence. This inhibitor has been shown to specifically block NF-κB translocation to the nucleus at 18 nM, the concentration used in this study (14). Our approach was to allow the first wave of NF-κB to enter the nucleus; then during the minimum of NF-κB-binding activity at 4 h, the inhibitor was added. If the NF-κB requirement for the iNOS promoter were already in the nucleus from the first phase, there should be no significant change in NO output. Alternatively, if the second phase of NF-κB is required for iNOS transcription, a decrease should be observed in the NO output of the cells. We found that addition of SN50 at 4 h significantly reduced the levels of nitrite in the supernatants of LPS-stimulated cells (Fig. 2,c), strongly suggesting that the second phase of NF-κB is required for iNOS transcription, a novel finding. In Fig. 2 d, we demonstrate the specificity of the SN50 inhibitor by comparing the inhibition of NF-κB translocation with that of AP-1 by gel shift analysis. There is no significant effect on AP-1 translocation, whereas NF-κB complexes are severely reduced in the nucleus.

Because LPS stimulation is known to induce degradation of IκBβ, whereas many other stimuli do not, and this degradation is most likely associated with the biphasic nature of the NF-κB response, we performed analyses of IκB degradation in response to LNFPIII-Dex (10, 15). As expected, LNFPIII-Dex did not induce the degradation of IκBβ, whereas LPS induced rapid degradation of IκBα and IκBβ (Fig. 3) in both RAW264.7 cells and DCs. These results lend further weight to the importance of the regulation of IκB proteins as discriminatory points in the signal transduction cascades of various stimuli. Because most stimuli that lead to the translocation of NF-κB into the nucleus induce the degradation of IκBα, we were surprised that LNFPIII-Dex stimulation did not induce any observable degradation of IκBα (Fig. 3). This finding did not change even when we loaded varying levels of extracts, as well as looking at multiple time points (from 10 min to 2 h in 10-min increments). Because LNFPIII-Dex induces measurable NF-κB translocation within 30 min by gel shift and also induces significant activity of an NF-κB luciferase reporter plasmid, we decided to investigate the possibility that an alternate IκB was being degraded in response to LNFPIII-Dex stimulation. As can be seen in Fig. 3, there was no observable degradation of IκBα, IκBβ, or IκBε isoforms of the IκB family at 10 or 20 min poststimulation with LNFPIII-Dex. One alternative explanation for these data is that LNFPIII-Dex induces IκBα degradation, but rapid resynthesis of IκBα is such that we cannot observe its degradation. To exclude this possibility, we included cycloheximide, a potent inhibitor of protein synthesis, in our DC cultures and still did not see any IκB degradation (Fig. 3 b).

FIGURE 3.

LNFPIII-Dex does not induce the degradation of IκB. RAW264.7 cells (a) or BM-derived DCs (b) were treated for the indicated times with 50 μg/ml dextran (lane 1), LNnT-Dex (lane 2), LNFPIII-Dex (lane 3), or 1 μg/ml LPS (lane 4). To rule out the possibility of IκB resynthesis, cells in (b) were pretreated for 1 h with 10 μg/ml cycloheximide. Extracts were subjected to SDS-PAGE and Western blotted with IκB-specific Abs. Blots were stripped and blotted with an anti-total MAPK Ab. Data are representative of at least three independent experiments.

FIGURE 3.

LNFPIII-Dex does not induce the degradation of IκB. RAW264.7 cells (a) or BM-derived DCs (b) were treated for the indicated times with 50 μg/ml dextran (lane 1), LNnT-Dex (lane 2), LNFPIII-Dex (lane 3), or 1 μg/ml LPS (lane 4). To rule out the possibility of IκB resynthesis, cells in (b) were pretreated for 1 h with 10 μg/ml cycloheximide. Extracts were subjected to SDS-PAGE and Western blotted with IκB-specific Abs. Blots were stripped and blotted with an anti-total MAPK Ab. Data are representative of at least three independent experiments.

Close modal

Because we did not observe any IκB degradation in response to LNFPIII-Dex, we hypothesized that active p50 complexes were being generated via an alternative pathway that did not require IκB degradation, but instead relied upon the degradation of p105 to generate p50 and release other Rel subunits. One prediction of this hypothesis would be the accumulation of p50 very early after stimulation, before new transcription and translation were occurring. Using Western blot analysis, we found that 10 min after stimulation with either LNFPIII-Dex or LPS there was a measurable increase in the amount of total cellular p50 with respect to a total ERK control Ab. The relative level of induction was ∼2-fold for LNFPIII-Dex and ∼2.5-fold for LPS vs the control of p50 (Fig. 4). Although in some experiments we did observe a measurable degradation of p105, it was not consistent in all experiments. However, depending on the initial ratio of p105/p50, a 2-fold increase in p50 may not cause an easily measurable decrease in the levels of p105 even if the p50 is produced by p105 degradation. This hypothesis seems likely due to the short time of exposure.

FIGURE 4.

LNFPIII-Dex and LPS induce the accumulation of p50 (a). RAW264.7 cells were treated for 10 min with either 50 μg/ml LNFPIII-Dex or 1 μg/ml LPS. Levels of p50 were measured by Western blot, and quantitation was performed on a Kodak ISF440 imaging unit with Kodak 1D analysis software. Levels of p50 were normalized to an ERK control. Asterisks indicate p < 0.05 by t test (b). Representative Western blot used in analysis for a.

FIGURE 4.

LNFPIII-Dex and LPS induce the accumulation of p50 (a). RAW264.7 cells were treated for 10 min with either 50 μg/ml LNFPIII-Dex or 1 μg/ml LPS. Levels of p50 were measured by Western blot, and quantitation was performed on a Kodak ISF440 imaging unit with Kodak 1D analysis software. Levels of p50 were normalized to an ERK control. Asterisks indicate p < 0.05 by t test (b). Representative Western blot used in analysis for a.

Close modal

LNFPIII-Dex was previously shown to activate immature DCs and then promote DC2 development in an in vitro culture system (6). To determine whether NF-κB was necessary for naive DCs to functionally mature to the DC2 phenotype, we added the inhibitor SN50 to DCs during activation with LNFPIII-Dex. As can be seen in Fig. 5, addition of the NF-κB inhibitor during DC maturation with LNFPIII-Dex abrogated the Th biasing previously observed with LNFPIII-Dex-stimulated cells with the levels of IL-4 decreased, and the levels of IFN-γ increased significantly vs LNFPIII-Dex plus a control inhibitor. These experiments were performed with T cells derived from transgenic mice that are specific for a class II-restricted OVA peptide, and so reflect the ability of LNFPII-Dex to modulate Ag-specific responses. Although it has been suggested that the SN50 inhibitor has some specificity problems, we used it at 18 mM, a concentration well below the level at which most groups have claimed to see nonspecific interactions (14, 16).

To confirm the results obtained with the SN50 inhibitor, we generated immature DCs from wild-type and NF-κB1-deficient mice and matured them either with LPS or LNFPIII-Dex to generate DC1 and DC2 cells, respectively. We then cocultured these mature DCs with allogeneic CD4+ T cells in an MLR. As expected, wild-type DCs matured with LPS or LNFPIII-Dex resulted in a DC1 or DC2 phenotype, respectively, confirming our previously reported results (6). Conversely, DCs generated from NF-κB1 knockout mice were unable to promote Th1 or Th2 differentiation, demonstrating the essential role for NF-κB in promoting DC1 and DC2 responses (Fig. 6). The T cells in our DC:T cell coculture were wild type for NF-κB; therefore, the observed defect in differentiation had to occur in the p105−/− DCs themselves. The difference in IFN-γ levels observed between LNFPIII-Dex-stimulated DC:T cell cocultures in Figs. 5 and 6 is most likely a result of the different experimental conditions used; Fig. 5 uses an Ag-specific T cell priming system, whereas Fig. 6 is an MLR. In the Ag-specific system, we see a more obvious suppression/prevention of IFN-γ production by LNFPIII-Dex-stimulated DCs vs the suppression observed in the MLR.

FIGURE 6.

LNFPIII-Dex cannot stimulate DC2 development in p105 knockout mice. DCs from wild-type (a) or p105 knockout mice (b) were generated, as described in Materials and Methods. Cells were matured with the indicated stimuli and then cocultured with allogeneic T cells in an MLR. After 96 h, cultures were examined by ELISA for IL-4 and IFN-γ. Results are representative of three independent experiments.

FIGURE 6.

LNFPIII-Dex cannot stimulate DC2 development in p105 knockout mice. DCs from wild-type (a) or p105 knockout mice (b) were generated, as described in Materials and Methods. Cells were matured with the indicated stimuli and then cocultured with allogeneic T cells in an MLR. After 96 h, cultures were examined by ELISA for IL-4 and IFN-γ. Results are representative of three independent experiments.

Close modal

We report that the helminth-related neoglycoconjugate LNFPIII-Dex induces a transient activation of NF-κB in APCs and that this activity is required for LNFPIII-Dex-driven maturation of DCs to DC2s that promote Th2 responses. We additionally show that the nature of NF-κB activation, transient vs prolonged, is in fact extremely important in determining the functional development of APCs, with transient NF-κB activation related to maturation of DC2s and prolonged (or the second wave) activation of NF-κB being required for the production of the proinflammatory mediator NO in LPS-activated cells. Although LNFPIII-Dex stimulation only drove transient activation of NF-κB, we found using both the SN50 inhibitor peptide and NF-κB1 knockout mice that NF-κB is essential for LNFPIII-Dex-induced development of a DC2 phenotype. These results indicate that NF-κB activation cannot be considered as a single output but consists of multiple parameters, including kinetics, that determine the functional role of NF-κB in response to a particular stimulus.

Regarding differential activation of NF-κB, it has been argued that the regulation of IκBα and IκBβ can lead to a biphasic pattern of NF-κB translocation to the nucleus. IκBα is rapidly degraded upon stimulation with a variety of mitogens, including LPS. The IκBα gene is a target of NF-κB, and, as a result, IκBα down-regulation is only transient, leading to an autofeedback inhibition of NF-κB. The IκBβ gene is not a target of NF-κB and is not resynthesized after degradation, which appears to allow a persistence of NF-κB in the nucleus (10). The ability of these two different modes of regulation to lead to the biphasic response observed downstream of LPS has been mathematically represented recently (15). Recent studies have also demonstrated that there are at least two distinct subsets of genes that are controlled by NF-κB and that this control is in part temporal. Saccani et al. (13) have shown that chromatin modifications were necessary for NF-κB binding associated with the promoters of genes in DCs in response to LPS. A subset of these chromatin modifications was shown to be restricted to proinflammatory genes dependent upon p38 kinase, whereas other modifications occurred in a p38-independent manner. Genes in the p38-independent subset included MIP-1α, which LNFPIII-Dex induces in RAW264.7 macrophages (data not shown). Because LNFPIII-Dex stimulation of APCs preferentially leads to phosphorylation of ERK and not p38, a putative mechanism to explain some of the differences that we observe between LPS- and LNFPIII-Dex-stimulated APCs may be based upon their disparate activation of p38 kinase (6). Thus, whether stimulation by a pathogen or pathogen-associated molecular pattern does or does not lead to activation of p38 may be part of the mechanism by which cells control whether or not NF-κB acts as an inflammatory or anti-inflammatory signal (17). This kind of differential MAPK activation as a way of modulating cell responses to reflect different signal qualities or strengths may not be restricted to APCs because a similar phenomenon has been observed in T cells (18).

The data reported in this work also explain why persistent NF-κB activation is necessary to activate certain downstream targets. By preventing new import of NF-κB complexes at 4 h (when NF-κB activity is minimal post-LPS stimulation), we were able to isolate and measure the effect of only the second phase of NF-κB activation, which we demonstrated was required for LPS-activated macrophages to produce NO. To our knowledge, this is the first demonstration of a functional role of the second phase of NF-κB activation showing that the persistent NF-κB response to certain stimuli (such as LPS) plays an important role in the phenotypic development of the responding cell.

A surprising finding in this study was that LNFPIII-Dex did not induce observable degradation of any member of the IκB family, despite inducing significant NF-κB activity in the nucleus shortly after stimulation. This observation may be because LNFPIII-Dex-induced degradation of one or more IκB family members was at levels below the detection of the Western blot technique used in this study, although that seems unlikely. IκB degradation in other systems has been described in terms of apparent degradation, and a finding that the degradation of an IκB isoform could be restricted to a small subset of the molecules of that isoform would be a new hypothesis. It seems more likely that another mechanism of NF-κB activation and translocation is occurring, either via a new isoform of IκB, or potentially by a non-IκB degradation-dependent function of one of the existing family members. Our further observation that LNFPIII-Dex induced the accumulation of p50 indicates that LNFPIII-Dex is able to activate an alternative pathway of NF-κB translocation that most likely depends upon the degradation of the p105 precursor protein.

Recent studies have demonstrated the existence of a non-IκB degradation-based pathway for NF-κB activation. Much of this work has been done with regard to the activation of NF-κB complexes containing p52, the degradation product of p100. Claudio et al. (19) showed that downstream of B cell activating factor (BAFF) signaling, the rate of p100 degradation is increased, releasing p52 as well as any Rel subunits that are bound to the ankyrin repeats of the full p100 molecule. This pathway has also been confirmed downstream of other stimuli (20, 21). There has been debate as to whether a similar, non-IκB-dependent pathway might exist for complexes containing p50, although there is significant evidence that one does exist (22, 23, 24, 25). Part of the controversy has arisen, as p50 can be generated by cotranslational processing event involving the 26S proteasome that allows direct p50 synthesis rather than being produced as the breakdown product of p105 (26). This cotranslational pathway appears to occur on a constitutive basis for p50, whereas it remains controversial as to whether it occurs for p100/p52. However, while p50 and p105 appeared to be produced at relatively equal rates, much less p52 is produced relative to p100 (27). This led some investigators to draw the conclusion that this cotranslational process was the only method of p50 generation. Work by several groups, however, has found that p50 can be generated by an inducible pathway in which p105 is degraded, releasing both p50 and any bound Rel subunits, as well as the MEK kinase, Tumor progression locus 2 (TPL-2) (28, 29). Conversely, several studies have suggested that this inducible p105 degradation results in the complete proteolysis of the molecule, rather than in the release of a p50 subunit (reviewed in Ref. 30). There has been evidence though of p50 accumulation by induced p105 degradation, and our data would suggest that an inducible pathway exists that can result in p50 accumulation with functional consequences, even if the observed accumulation does not appear dramatic. This would not only release a p50 subunit but would potentially release any Rel proteins associating with the ankyrin repeats in p105. This hypothesis is supported by our previous observation that LNFPIII-Dex only induces strong ERK activation vs weak p38 and JNK activation in APCs downstream of TLR4. In addition, studies have shown that TPL-2, a MEK kinase, is associated with p105 and becomes activated upon its degradation and that this activation event leads to ERK but not p38 or JNK activation. In p105 knockouts, ERK activation downstream of TLR4 is completely deficient due to TPL-2 instability resulting from the absence of p105 (28, 29).

Thus, two of the major signaling pathways downstream of LNFPIII-Dex-TLR4 stimulation of APCs, ERK activation and NF-κB translocation, intersect at the point of p105 degradation. This alternate pathway of NF-κB activation is clearly important and may have been missed for many years due to the strong and persistent activation often observed via IκBα degradation-induced NF-κB translocation. LNFPIII-Dex, then, is an important reagent to study this pathway of NF-κB activation without the noise of the IκB-mediated events. Furthermore, this pathway may be useful in explaining the role that we found for NF-κB in promoting the Th2 response. The ability to drive DC2 maturation or Th2 responses downstream of TLR4 was first shown in MyD88-deficient mice by Kaisho et al. (31) and more recently by Amsen et al. (32).

Our data demonstrating a role for NF-κB in the differentiation of DC2 cells were initially surprising, given the strong association between NF-κB and proinflammatory responses. Indeed, a recent study by Kane et al. (33) used a soluble extract of schistosome eggs that demonstrated a potent ability to suppress LPS-induced NF-κB activation, consistent with the ability of soluble extract of schistosome eggs to drive anti-inflammatory responses. However, similar to what we show in this work, Artis et al. (34) demonstrated the importance of the p50 subunit in the development of the Th2 response during infection with Trichuris muris. It was not clear from this study in which cell type(s) the p50 subunit was essential. Furthermore, a report by Lawrence et al. (35) described the up-regulation of NF-κB activity in the nucleus during the resolution of inflammation. This is consistent with our own findings of NF-κB activity being up-regulated in response to an anti-inflammatory stimulus. In keeping with previous observations in NF-κB activation and similar to what we report in this work, the authors did not observe IκBα degradation in the anti-inflammatory wave of NF-κB activity, whereas the inflammatory wave of NF-κB translocation was dependent on IκBα degradation. These studies, taken together with our results, suggest an important role for NF-κB in both inflammatory and anti-inflammatory responses. The concurrence of this apparently novel pathway of NF-κB activation in three anti-inflammatory systems may indicate that this is a pathway common to anti-inflammatory or Th2-driving stimuli and may be characteristic of Th2 pathogen-associated molecular patterns.

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 Neose Technologies, National Institutes of Health Grant 5R01AI056484, and a Howard Hughes Medical Institute Predoctoral Fellowship (to P.G.T.).

4

Abbreviations used in this paper: LNFPIII, lacto-N-fucopentaose III; BM, bone marrow; DC, dendritic cell; iNOS, inducible NO synthase; LNFPIII-Dex, LNFPIII (12-25 molecules)-dextran; LNnT, lacto-N-neotetraoase; LNnT-Dex, LNnT (12-25)-dextran; TPL-2, Tumor progression locus 2.

1
Ghosh, S., M. Karin.
2002
. Missing pieces in the NF-κB puzzle.
Cell
109
:
S81
-S96.
2
Li, Q., I. M. Verma.
2002
. NF-κB regulation in the immune system.
Nat. Rev. Immunol.
2
:
725
-734.
3
Verma, I. M., J. K. Stevenson, E. M. Schwarz, D. Van Antwerp, S. Miyamoto.
1995
. Rel/NF-κB/IκB family: intimate tales of association and dissociation.
Genes Dev.
9
:
2723
-2735.
4
Gerondakis, S., M. Grossmann, Y. Nakamura, T. Pohl, R. Grumont.
1999
. Genetic approaches in mice to understand Rel/NF-κB and IκB function: transgenics and knockouts.
Oncogene
18
:
6888
-6895.
5
Okano, M., A. R. Satoskar, K. Nishizaki, D. A. Harn, Jr.
2001
. Lacto-N-fucopentaose III found on Schistosoma mansoni egg antigens functions as adjuvant for proteins by inducing Th2-type response.
J. Immunol.
167
:
442
-450.
6
Thomas, P. G., M. R. Carter, O. Atochina, A. A. Da’Dara, D. Piskorska, E. McGuire, D. A. Harn.
2003
. Maturation of dendritic cell 2 phenotype by a helminth glycan uses a Toll-like receptor 4-dependent mechanism.
J. Immunol.
171
:
5837
-5841.
7
Dissanayake, S., R. S. Amith, A. Shahin.
2004
. Taenia crassiceps carbohydrates stimulate IL-6 expression in naive murine macrophages via Toll-like receptors (TLRs).
Mol. Immunol.
41
:
391
-398.
8
Bergman, M. P., A. Engering, H. H. Smits, S. J. van Vliet, A. A. van Bodegraven, H. P. Wirth, M. L. Kapsenberg, C. M. Vandenbroucke-Grauls, Y. van Kooyk, B. J. Appelmelk.
2004
. Helicobacter pylori modulates the T helper cell 1/T helper cell 2 balance through phase-variable interaction between lipopolysaccharide and DC-SIGN.
J. Exp. Med.
200
:
979
-990.
9
Towbin, H., T. Staehelin, J. Gordon.
1979
. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications.
Proc. Natl. Acad. Sci. USA
76
:
4350
-4354.
10
Thompson, J. E., R. J. Phillips, H. Erdjument-Bromage, P. Tempst, S. Ghosh.
1995
. IκB-β regulates the persistent response in a biphasic activation of NF-κB.
Cell
80
:
573
-582.
11
Denlinger, L. C., K. A. Garis, J. A. Sommer, A. G. Guadarrama, R. A. Proctor, P. J. Bertics.
1998
. Nuclear translocation of NF-κB in lipopolysaccharide-treated macrophages fails to correspond to endotoxicity: evidence suggesting a requirement for a γ interferon-like signal.
Infect. Immun.
66
:
1638
-1647.
12
Chen, C. C., J. K. Wang.
1999
. p38 but not p44/42 mitogen-activated protein kinase is required for nitric oxide synthase induction mediated by lipopolysaccharide in RAW 264.7 macrophages.
Mol. Pharmacol.
55
:
481
-488.
13
Saccani, S., S. Pantano, G. Natoli.
2001
. Two waves of nuclear factor κB recruitment to target promoters.
J. Exp. Med.
193
:
1351
-1359.
14
Lin, Y. Z., S. Y. Yao, R. A. Veach, T. R. Torgerson, J. Hawiger.
1995
. Inhibition of nuclear translocation of transcription factor NF-κB by a synthetic peptide containing a cell membrane-permeable motif and nuclear localization sequence.
J. Biol. Chem.
270
:
14255
-14258.
15
Hoffmann, A., A. Levchenko, M. L. Scott, D. Baltimore.
2002
. The IκB-NF-κB signaling module: temporal control and selective gene activation.
Science
298
:
1241
-1245.
16
Kolenko, V., T. Bloom, P. Rayman, R. Bukowski, E. Hsi, J. Finke.
1999
. Inhibition of NF-κB activity in human T lymphocytes induces caspase-dependent apoptosis without detectable activation of caspase-1 and -3.
J. Immunol.
163
:
590
-598.
17
Agrawal, S., A. Agrawal, B. Doughty, A. Gerwitz, J. Blenis, T. Van Dyke, B. Pulendran.
2003
. Cutting edge: different Toll-like receptor agonists instruct dendritic cells to induce distinct Th responses via differential modulation of extracellular signal-regulated kinase-mitogen-activated protein kinase and c-Fos.
J. Immunol.
171
:
4984
-4989.
18
Singh, R. A., J. Z. Zhang.
2004
. Differential activation of ERK, p38, and JNK required for Th1 and Th2 deviation in myelin-reactive T cells induced by altered peptide ligand.
J. Immunol.
173
:
7299
-7307.
19
Claudio, E., K. Brown, S. Park, H. Wang, U. Siebenlist.
2002
. BAFF-induced NEMO-independent processing of NF-κB2 in maturing B cells.
Nat. Immunol.
3
:
958
-965.
20
Senftleben, U., Y. Cao, G. Xiao, F. R. Greten, G. Krahn, G. Bonizzi, Y. Chen, Y. Hu, A. Fong, S. C. Sun, M. Karin.
2001
. Activation by IKKα of a second, evolutionary conserved, NF-κB signaling pathway.
Science
293
:
1495
-1499.
21
Dejardin, E., N. M. Droin, M. Delhase, E. Haas, Y. Cao, C. Makris, Z. W. Li, M. Karin, C. F. Ware, D. R. Green.
2002
. The lymphotoxin-β receptor induces different patterns of gene expression via two NF-κB pathways.
Immunity
17
:
525
-535.
22
Saitoh, T., H. Nakano, N. Yamamoto, S. Yamaoka.
2002
. Lymphotoxin-β receptor mediates NEMO-independent NF-κB activation.
FEBS Lett.
532
:
45
-51.
23
Ciechanover, A., H. Gonen, B. Bercovich, S. Cohen, I. Fajerman, A. Israel, F. Mercurio, C. Kahana, A. L. Schwartz, K. Iwai, A. Orian.
2001
. Mechanisms of ubiquitin-mediated, limited processing of the NF-κB1 precursor protein p105.
Biochimie
83
:
341
-349.
24
Heissmeyer, V., D. Krappmann, E. N. Hatada, C. Scheidereit.
2001
. Shared pathways of IκB kinase-induced SCF(βTrCP)-mediated ubiquitination and degradation for the NF-κB precursor p105 and IκBα.
Mol. Cell. Biol.
21
:
1024
-1035.
25
Heissmeyer, V., D. Krappmann, F. G. Wulczyn, C. Scheidereit.
1999
. NF-κB p105 is a target of IκB kinases and controls signal induction of Bcl-3-p50 complexes.
EMBO J.
18
:
4766
-4778.
26
Lin, L., G. N. DeMartino, W. C. Greene.
1998
. Cotranslational biogenesis of NF-κB p50 by the 26S proteasome.
Cell
92
:
819
-828.
27
Heusch, M., L. Lin, R. Geleziunas, W. C. Greene.
1999
. The generation of NFκb2 p52: mechanism and efficiency.
Oncogene
18
:
6201
-6208.
28
Beinke, S., J. Deka, V. Lang, M. P. Belich, P. A. Walker, S. Howell, S. J. Smerdon, S. J. Gamblin, S. C. Ley.
2003
. NF-κB1 p105 negatively regulates TPL-2 MEK kinase activity.
Mol. Cell. Biol.
23
:
4739
-4752.
29
Waterfield, M. R., M. Zhang, L. P. Norman, S. C. Sun.
2003
. NF-κB1/p105 regulates lipopolysaccharide-stimulated MAP kinase signaling by governing the stability and function of the Tpl2 kinase.
Mol. Cell
11
:
685
-694.
30
Beinke, S., S. C. Ley.
2004
. Functions of NF-κB1 and NF-κB2 in immune cell biology.
Biochem. J.
382
:
393
-409.
31
Kaisho, T., K. Hoshino, T. Iwabe, O. Takeuchi, T. Yasui, S. Akira.
2002
. Endotoxin can induce MyD88-deficient dendritic cells to support T(h)2 cell differentiation.
Int. Immunol.
14
:
695
-700.
32
Amsen, D., J. M. Blander, G. R. Lee, K. Tanigaki, T. Honjo, R. A. Flavell.
2004
. Instruction of distinct CD4 T helper cell fates by different notch ligands on antigen-presenting cells.
Cell
117
:
515
-526.
33
Kane, C. M., L. Cervi, J. Sun, A. S. McKee, K. S. Masek, S. Shapira, C. A. Hunter, E. J. Pearce.
2004
. Helminth antigens modulate TLR-initiated dendritic cell activation.
J. Immunol.
173
:
7454
-7461.
34
Artis, D., S. Shapira, N. Mason, K. M. Speirs, M. Goldschmidt, J. Caamano, H. C. Liou, C. A. Hunter, P. Scott.
2002
. Differential requirement for NF-κB family members in control of helminth infection and intestinal inflammation.
J. Immunol.
169
:
4481
-4487.
35
Lawrence, T., D. W. Gilroy, P. R. Colville-Nash, D. A. Willoughby.
2001
. Possible new role for NF-κB in the resolution of inflammation.
Nat. Med.
7
:
1291
-1297.