IL-1β Enhances CD40 Ligand-Mediated Cytokine Secretion by Human Dendritic Cells (DC): A Mechanism for T Cell-Independent DC Activation1

CD40 ligand (CD40L) is a membrane-bound molecule expressed by activated T cells. CD40L potently induces dendritic cell (DC) maturation and IL-12p70 secretion and plays a critical role during T cell priming in the lymph nodes. IFN-γ and IL-4 are required for CD40L-mediated cytokine secretion, suggesting that T cells are required for optimal CD40L activity. Because CD40L is rapidly up-regulated by non-T cells during inflammation, CD40 stimulation may also be important at the primary infection site. However, a role for T cells at the earliest stages of infection is unclear. The present study demonstrates that the innate immune cell-derived cytokine, IL-1β, can increase CD40L-induced cytokine secretion by monocyte-derived DC, CD34+-derived DC, and peripheral blood DC independently of T cell-derived cytokines. Furthermore, IL-1β is constitutively produced by monocyte-derived DC and monocytes, and is increased in response to intact Escherichia coli or CD40L, whereas neither CD34+-derived DC nor peripheral blood DC produce IL-1β. Finally, DC activated with CD40L and IL-1β induce higher levels of IFN-γ secretion by T cells compared with DC activated with CD40L alone. Therefore, IL-1β is the first non-T cell-derived cytokine identified that enhances CD40L-mediated activation of DC. The synergy between CD40L and IL-1β highlights a potent, T cell-independent mechanism for DC activation during the earliest stages of inflammatory responses.


M aturation of dendritic cells (DC)
is a critical event for the induction of an immune response. Upon pathogen entry or tissue damage, inflammatory processes in the affected microenvironment induce DC maturation, which in turn can induce DC to migrate out of the inflamed peripheral tissues and into the lymphatic organs. DC maturation is characterized by a significant change in the cell's Ag expression pattern and functional characteristics. These mature APC can induce quiescent T cells to become activated (1).
A critical role for CD40L during the priming phase of adaptive immune responses in the lymph nodes has now been established. Here naive CD4 ϩ T cells up-regulate CD40L after interaction with CD70, the ligand for CD27. CD70 is expressed on CD8 ϩ T cells following interaction with DC (14). This suggests an important stimulatory loop, which involves DC, CTL, and Th cells. The crucial function of CD4 ϩ Th cells in priming naive CD8 ϩ CTL can be completely replaced by preactivating DC with anti-CD40 or soluble CD40L (15)(16)(17). In humans, CD40L expression is not restricted to T cells but has also been reported on activated B lymphocytes (18), eosinophils (19), platelets (20), smooth muscle cells (21), vascular endothelial cells (21), macrophages (21), and DC (22,23). Because CD40L can be expressed by non-T cells, it is possible that non-T cells may contribute to CD40L-mediated effects at sites of infection, providing an alternative, T cell-independent CD40L pathway. Although maximal CD40L-mediated IL-12p70 secretion by DC requires T cell-derived cytokines (IFN-␥ (9) or IL-4 (24 -26)), thereby suggesting a dependence on T cell assistance, a role for T cells at the earliest stages of infection is unclear. Presumably, the presence of T cells (and T cell-derived cofactors such as IFN-␥ and IL-4) at such sites would be temporally delayed and restricted to the memory effectors that eventually arrive after successful priming in the lymph nodes. A role for CD40L at the primary infection site may still be possible if other cofactors produced by non-T cells were present. A recent report indicates that IL-12p70 secretion by DC in vivo can only be amplified by CD40L in the presence of innate signals initiated by microbial stimuli. They exclude IFN-␥ as the innate signal but are unable to directly define it (27).
The present study demonstrates that IL-1␤ is a potent modulator of CD40L-induced cytokine secretion by three different human DC subsets: MoDC, CD34 ϩ -derived DC, and peripheral blood DC (PBDC). This can occur independently of IL-4 and IFN-␥ and result in the induction of IFN-␥ secretion by T cells. Because monocytes and MoDC rapidly secrete high levels of IL-1␤ following encounter with intact microbes (such as bacteria), our results identify a novel mechanism for the rapid induction of inflammatory responses, which can occur at the infection site independently of T cells and T cell-derived cytokines.

Cell sources
For CD34 ϩ progenitor cells, leukapheresis harvest samples were obtained from either normal donors or patients of the Department of Medical Oncology, Austin and Repatriation Medical Center (Heidelberg, Australia). Patients with non-Hodgkin's lymphoma or solid tumors received stem cellmobilizing chemotherapy and G-CSF as part of their treatment. PBDC were isolated from the blood of patients with stage III and IV melanoma enrolled in a Phase I clinical study (LUD97-012), receiving 14 consecutive days of Flt3 ligand (FL) (25 g/kg/day) followed by peptide vaccines. Blood for PBDC was taken at day 15. The protocol was approved by the Ludwig Institute's Investigators Review Board and the Ethics Committee at the Austin and Repatriation Medical Center, and informed consent was obtained from all patients. Monocytes and PBDC were also isolated from the PBMC fraction of healthy volunteer donors provided by the Australian Red Cross Blood Bank (Southbank, Melbourne, Australia) and used to produce MoDC.

MoDC
For MoDC generation, CD14 ϩ monocytes (5 ϫ 10 5 ) were affinity-purified using the MACS CD14 isolation kit (Miltenyi Biotech) and cultured in 1 ml of RPMI 1640, 10% FCS, GM-CSF (40 ng/ml), and IL-4 (500 IU/ml) in 24-well plates. At day 7, MoDC represented Ͼ90% of cultured cells. All wells were pooled and readjusted to a concentration of 1 ϫ 10 5 DC/ml. Maturation-inducing factors were added on day 7 and cells and supernatants were harvested on day 10 for functional assessment. MoDC-conditioned medium refers to the medium from MoDC cultures after 7 days of culture in GM-CSF and IL-4.

Enrichment of PBDC from FL-treated patients
PBDC were enriched from frozen PBMC samples obtained from a Phase I randomized study performed in HLA-A2 ϩ patients with evaluable stage III and IV malignant melanoma receiving FL (Immunex) with or without peptide vaccines (LUD97-012). Informed consent was obtained from all individuals and the protocol and consent forms were approved by an Investigators Review Board of the Ludwig Institute for Cancer Research. After thawing, CD14 ϩ monocytes were depleted using immunomagnetic beads (MACS; Miltenyi Biotech) according to the manufacturer's instructions. These CD14-depleted PBMC underwent a second round of depletion using MACS beads coupled to anti-CD3, anti-CD14, anti-CD19 (Miltenyi Biotech), and purified anti-CD16, anti-CD56, and anti-glycophorin A (BD PharMingen) in combination with rat anti-mouse IgG MACS beads (Miltenyi Biotech). This depletion procedure yielded Ͼ80% CD1b/c ϩ CD14 Ϫ HLA-DR ϩ PBDC as assessed by FACS. These immature PBDC were then cultured in 96-well plates (1 ϫ 10 5 /well) in RPMI 1640 -10% FCS for 3 days with various combinations of stimuli before examination of function. In later experiments, CD1b/c ϩ HLA-DR ϩ PBDC were sorted by FACS on a MoFlo cell sorter (Cytomation, Fort Collins, CO).

Blocking of CD40L-mediated IL-12p70 secretion with anti-IL-4 or anti-IL-1R mAb
MoDC (1 ϫ 10 5 /ml) were stimulated in their own conditioned medium (CM) with CD40L (1 g/ml) for 24 h. Neutralizing anti-IL-4 or IL-1R mAb or both (BD PharMingen) were added at the initiation of culture in a range of concentrations (0.05-2 g/ml). An isotype-matched mouse mAb was also used as control (BD PharMingen). After 24 h, supernatants (SN) were harvested and examined for IL-12p70 secretion by ELISA.

Measurement of Ag uptake
MoDC were harvested after culture in maturation-inducing conditions. Following incubation with 1 mg/ml FITC-dextran (44 and 260 kDa; Sigma-Aldrich, St. Louis, MO) for 30 -60 min at 0 or 37°C. Cells were washed three times in PBS 5% FCS and then incubated with PE-anti-CD11c. FITCdextran uptake was quantified as mean fluorescence intensity on gated CD11c ϩ cells. Nonspecific FITC signal was assessed by incubating MoDC with FITC-dextran at 0°C. Phagocytosis was assessed by incubating cells with 1 mg/ml PE-latex beads (Sigma-Aldrich) for 90 min at 37°C. In some conditions, cells were pretreated with 10 M cytochalasin D (Sigma-Aldrich) for 30 min at 37°C to depolymerize actin. To verify that the flow cytometry-based FITC signal represented internalized dextran or beads, cells were analyzed by epifluorescence and phase-contrast microscopy.

Mixed leukocyte reaction
Graded numbers of DC subsets were cultured in replicates in 96-well round-bottom plates (Falcon, Franklin Lakes, NJ) with 10 5 allogeneic T cells for 5 days in RPMI 1640 with 10% human serum. T cells were prepared from healthy volunteers by first rosetting with 2-aminoethylisothiouronium bromide-treated SRBC and then further fractionating by negative enrichment using anti-CD4 (CD8 ϩ T cells) or anti-CD8 (CD4 ϩ T cells) MACS beads (Miltenyi Biotech). After 5 days, 200 l of supernatants were harvested and fresh medium containing 1 Ci/well [ 3 H]thymidine (DuPont, Sydney, MA) was added for 8 h. Cells were transferred onto a glass fiber filter (Wallac, Turku, Finland), and [ 3 H]thymidine incorporation was measured using an LKB 1205 Betaplate scintillation counter (Wallac).

Cytokine secretion by CD40L activated MoDC in the presence or absence of their own CM
It has previously been shown that the induction of high levels of IL-12p70 secretion by MoDC requires a combination of either T cell signals (CD40L) (5,8) with IFN-␥ (9), or pathogen signals (LPS) (10 -14). We evaluated cytokine secretion by MoDC following stimulation with CD40L. When added directly into MoDC cultures, CD40L alone induced IL-12p70 secretion (200 -1000 pg/ ml; n Ͼ 10). These culture SN did not contain measurable levels of IFN-␥ as assessed by ELISA (sensitivity, 10 pg/ml; n Ͼ 40). This suggested that the CM of MoDC cultures after 7 days of culture in GM-CSF and IL-4 may contain factor(s) other than IFN-␥ that can enhance CD40L-induced IL-12p70 secretion. Fig. 1 shows the effect of MoDC-CM on the secretion of IL-6, IL-10, and IL-12p70 by MoDC following 3 days of stimulation with either CD40L or IFN-␥ or the combination of CD40L and IFN-␥. MoDC-CM contained high levels of IL-6 (Ͼ5000 pg/ml), in the absence of stimulation, and this was not enhanced by either IFN-␥ or CD40L stimulation (Fig. 1A). In contrast, washed MoDC cultured in fresh RPMI 1640/FCS (in the absence of exogenous GM-CSF and IL-4) produced low levels of IL-6 (Ͻ500 pg/ml) over the 3-day culture period which could somewhat be enhanced by CD40L (max 1000 pg/ml) but not by IFN-␥ (Fig. 1A). The combination of IFN-␥ and CD40L enhanced IL-6 secretion by MoDC in either the presence or absence of MoDC-CM. Similarly, IL-10 was detected at significant levels in unstimulated (day 7) MoDC-CM and was even detected in the CM of washed MoDC recultured for 3 days in fresh RPMI 1640/FCS (Fig. 1B). CD40L, but not IFN-␥, induced IL-10 secretion by MoDC in both the presence and absence of MoDC-CM. Unlike with CD40L-mediated IL-6 secretion, IFN-␥ did not augment CD40L-mediated IL-10 secretion.
Unlike with IL-6 and IL-10, IL-12p70 was not detected in either the MoDC-CM of unstimulated MoDC or the CM of washed MoDC (Fig. 1C). Furthermore, CD40L only induced IL-12p70 secretion by MoDC in the presence of MoDC-CM. The addition of IFN-␥ did not induce IL-12p70 secretion in either the presence or absence of MoDC-CM. As reported previously (9), the addition of IFN-␥ to CD40L enhanced IL-12p70 secretion even in the absence of MoDC-CM (Fig. 1C); however, the highest IL-12p70 secretion was induced in the presence of MoDC-CM (1-15 ng/ml; n Ͼ 10). Finally, the finding that CD40L-mediated IL-10 secretion was not enhanced by IFN-␥ suggests that IL-12p70 and IL-10 secretion are differentially regulated. These results indicate that a factor(s) present in MoDC-CM synergize(s) with CD40L or CD40L and IFN-␥ to induce IL-12p70 secretion and augment IL-10 secretion.
Finally, the addition of either anti-IL-1R or anti-IL-4 mAb blocked CD40L-mediated IL-12p70 secretion by MoDC, indicating that both of these cytokines are important cofactors in CD40L-mediated cytokine secretion (Fig. 2D).

Effect of IL-1␤ on DC maturation
Given that IL-1␤ was identified as a potent cofactor for CD40Lmediated cytokine secretion, we next examined whether IL-1␤ could also act as a maturation factor when added to immature MoDC. Fig. 3A shows that immature MoDC express negligible levels of the maturation markers CD25, CD80, CD83, and CD86 and low to intermediate levels of surface HLA-I and HLA-II. Up-regulation of all of these molecules was achieved when immature MoDC were stimulated with CD40L alone but not by IL-1␤ alone (Fig. 3A). Furthermore, the combination of IL-1␤ with CD40L was no more potent than CD40L alone at inducing phenotypic maturation of immature MoDC (Fig. 3A). This suggests that IL-1␤ is not a potent factor for the induction of DC phenotypic maturation.

Effect of IL-1␤ on DC Ag uptake capacity and migration to chemokines
Immature MoDC capture a variety of Ags using several different mechanisms (1). As immature MoDC mature, they undergo a stepwise coordinated process of reducing their Ag uptake capacity, up-regulating the expression of chemokine receptors (such as CCR7), and acquiring migratory capacity toward chemokines such as CCL21 (MIP-3␤), which can direct them to draining lymphoid tissues (1). Therefore, we examined endocytic and phagocytic capacity of immature MoDC as compared with those matured with IL-1␤ or CD40L or combinations of CD40L and IL-1␤. Analysis of MoDC by FACS revealed that, as expected, immature MoDC were maximally capable of internalizing soluble dextran (260 kDa) and phagocytosing 1-m latex particles (Fig. 3, B and C). Maturation with IL-1␤ did not affect the capacity of MoDC to ingest FITC-dextran (260 kDa) or PE-latex (1 m). Only CD40L or CD40L plus IL-1␤ maximally reduced the ability of MoDC to ingest these particulates. This suggests that IL-1␤ is not a major regulator of MoDC Ag uptake capacity. Finally, immature MoDC matured with IL-1␤ did not migrate toward the CCR7 ligands, CCL21 (MIP-3␤) (Fig. 3D) or CCL19 (6Ckine) (data not shown), whereas those matured with IFN-␣2a, TNF-␣ and PGE 2 did (Fig.   FIGURE 2. IL-1␤ and IL-4 independently enhance CD40L-induced cytokine secretion. MoDC were washed on day 7 and recultured (1 ϫ 10 5 /ml) for a further 3 days in either fresh culture medium alone (RPMI 1640/FCS) or medium containing GM-CSF and IL-4, with or without the indicated cytokine stimuli. Culture SN were examined for cytokine production by ELISA. A, Secretion of IL-6. B, Secretion of IL-10. C, Secretion of IL-12p70. Addition of IL-1␤ alone or in combination with GM-CSF and IL-4 was tested separately and did not induce substantial cytokine secretion in these cultures. D, Blocking of CD40L-mediated IL-12p70 secretion by anti-IL-1R and anti-IL-4 mAb. MoDC were stimulated with CD40L for 24 h in the presence of either anti-IL-1R (0.5 g/ml) or anti-IL-4 (1 g/ml), or both, or with isotype-matched IgG1 control mAb. Culture SN were then harvested and examined for IL-12p70 production by ELISA. Data represent the mean Ϯ SD of three replicate wells. The figure is a representative experiment from three to four separate donors. 3D). This indicates that although IL-1␤ can synergize with CD40L to enhance IL-10 and IL-12p70 production, IL-1␤ as a single agent is inefficient at altering MoDC phenotypic maturation, migration, or Ag uptake capacity.

Secretion of IFN-␥ by allogeneic T lymphocytes stimulated with MoDC
IL-12p70 is an important regulator of IFN-␥ secretion by T cells (2)(3)(4)(5). We investigated whether the conditions which induced the highest levels of IL-12p70 production by MoDC translated into increased IFN-␥ secretion by allogeneic T cells following DC stimulation. MoDC were washed and activated with either CD40L alone or in combination with GM-CSF and IL-4 and/or IL-1␤ for 24 h, washed again, and then cultured with allogeneic T cells. Fig.  4A shows that IL-1␤ could potently enhance the ability of CD40Lactivated MoDC to stimulate IFN-␥ secretion by T cells. Furthermore, the cytokine combination that induced MoDC to secrete the highest levels of IL-12p70 (i.e., GM-CSF, IL-4, IL-1␤, and CD40L) also induced the highest IFN-␥ production by T cells (Fig.  4A). Interestingly, DC activated with the combination of CD40L and the T cell-independent factor (IL-1␤) were as potent as DC exposed to CD40L plus IFN-␥ at inducing IFN-␥ secretion in alloreactive T cells (Fig. 4B).

Examination of IL-1␤ secretion by MoDC, monocytes, CD34 ϩderived DC, and PBDC
IL-1␤ is predominantly produced by monocytes and macrophages in response to bacterial-derived signals (28). To investigate whether different DC subsets were comparable in their capacity to produce this cytokine, we measured IL-1␤ secretion in the culture SN of monocytes, MoDC, CD34 ϩ -derived DC, and PBDC following stimulation. All cultures were stimulated in the presence of GM-CSF and IL-4 with either CD40L plus IFN-␥ or intact Escherichia coli. Table I clearly shows differences with respect to basal and inducible IL-1␤ production by the four types of APC examined. Only monocytes and MoDC secreted significant basal levels of IL-1␤ (170 -210 pg/ml), and this was increased 17-and 30-fold, respectively, following stimulation with intact E. coli. CD40L plus IFN-␥ enhanced the level of IL-1␤ produced by monocytes and to a lesser degree by MoDC (Table I). In contrast, neither CD34 ϩderived DC nor PBDC secreted IL-1␤ in response to these stimuli. These results suggest that MoDC are closely related to their monocyte precursors with respect to IL-1␤ production, whereas CD34 ϩderived DC and PBDC, which are poor producers of IL-1␤ following stimulation, may represent functionally distinct APC populations.

Cytokine secretion by CD34 ϩ -derived DC and PBDC in response to CD40L and IL-1␤ stimulation
Because CD34 ϩ -derived DC and PBDC differed from MoDC in their capacity to secrete IL-1␤ following stimulation, we investigated whether CD40L-mediated cytokine secretion by CD34 ϩ -derived DC and PBDC could be enhanced by the addition of IL-1␤. CD34 ϩ -derived DC were generated under serum-free conditions as previously described using GM-CSF, TNF-␣, and IL-4, and yielded CD1a ϩ DC in the range of 35-65% of cultured cells (Fig.   FIGURE 3. Effect of IL-1␤ on immature MoDC phenotype and function. MoDC were prepared by culturing purified CD14 ϩ monocytes for 7 days in GM-CSF and IL-4. On day 7, CD40L (1 g/ml) and/or IL-1␤ were added for 3 days. 5A). CD34 ϩ -derived DC used in these experiments were phenotypically immature, as they expressed negligible levels of CD80, CD83, and CD86 and represented Langerhans-like cells expressing Lag and E-cadherin rather than interstitial DC, which express CD14 (29). Fig. 5B shows that compared with MoDC, CD34 ϩderived DC did not produce detectable IL-12p70 in response to CD40L in the presence of their own CM containing IL-4. In contrast, the addition of either IL-1␤ or IFN-␥ to CD40L induced IL-12p70 secretion by CD34 ϩ -derived DC, and this was further augmented when IL-1␤ and IFN-␥ were combined (Fig. 5B). Although IL-1␤ and/or IFN-␥ significantly enhanced the levels of IL-12p70 production by CD34 ϩ -derived DC, these were still 10fold lower than the levels produced by MoDC under the same stimulation conditions (Figs. 1C and 2C), highlighting a striking difference in the cytokine-secreting potential of these two in vitrogenerated DC populations. Furthermore, unlike MoDC, which secrete high levels of IL-10 in response to these stimuli, IL-10 was not substantially produced by CD34 ϩ -derived DC under these conditions. Finally, as with MoDC, IL-1␤ could potently enhance the ability of CD40L-activated CD34 ϩ -derived DC to stimulate IFN-␥ secretion by T cells. However, consistent with their lower IL-12p70-secreting potential, CD34 ϩ -derived DC induced lower levels of IFN-␥ in T cells than did MoDC (Fig. 5C).

Cytokine secretion by PBDC in response to CD40L and IL-1␤ stimulation
We next examined the effect of IL-1␤ upon CD40L-mediated cytokine secretion by PBDC isolated from healthy donors or expanded in vivo by the administration of FL to cancer patients. FL-generated PBDC were isolated from patients with malignant melanoma. FL-generated PBDC preparations were enriched by depletion of cells expressing CD3, CD14, CD16, CD19, CD56, and glycophorin A using Abs and magnetic beads. PBDC were also FACS sorted on the basis of CD1b/c and HLA-DR expression (Fig. 6, C and D). As previously reported, FL-generated PBDC were phenotypically immature when examined immediately ex vivo, expressing negligible levels of CD80 and CD83 and relatively low levels of CD86 and HLA-DR ( Fig. 6A and Refs. 30 and 31). However, maturation was rapidly induced upon in vitro culture, resulting in up-regulation in the expression of CD80, CD83, CD86, and HLA-DR molecules (Fig. 6A). Although isolated from cancer patients, FL-generated PBDC were not functionally defective as assessed by in vitro functional studies, but rather were potent stimulators of allogeneic T cell proliferation and cytokine secretion comparable to PBDC isolated from the blood of un- . In contrast to either MoDC or CD34 ϩ -derived DC, freshly isolated PBDC from FL-treated cancer patients (Fig. 6B) or from healthy donors (Fig. 6C) did not produce detectable levels of IL-12p70 in response to a 72-h culture with the indicated stimuli. However, CD40L plus IL-1␤ did increase IL-6 and IL-10 secretion by PBDC, indicating that PBDC could be stimulated to secrete cytokines in vitro and that this was modulated by IL-1␤ (Fig. 6, B and C). Finally, induction of IL-12p70 by PBDC in response to stimuli previously shown to induce IL-12p70 in MoDC (such as the combination of GM-CSF plus IL-4 plus CD40L plus IFN-␥ and intact E. coli bacteria as a source of pathogen signals) (24 -26) induced less than 100 pg/ml of IL-12p70 (Fig. 6D), this being the highest IL-12p70 level detected from three separate healthy donors. These results demonstrate that IL-1␤ enhances CD40L-mediated cytokine secretion in all three DC populations examined, but that the type (i.e., IL-6, IL-10, and IL-12p70) and levels of cytokines induced varies among the DC subsets.

Discussion
This study identifies IL-1␤ as a T cell-independent modulator of CD40L-induced DC activation. IL-1␤ alone was a poor stimulator of DC phenotypic maturation and function (e.g., Ag uptake capacity, migration toward chemokines, and cytokine secretion). However, when combined with CD40L, IL-1␤ enhanced the secretion of cytokines from three different DC populations: MoDC, CD34 ϩderived DC, and PBDC. The cytokine profile and quantities of cytokines secreted differed for these three DC types. For MoDC, IL-1␤ enhanced CD40L-mediated IL-6, IL-10, and IL-12p70 secretion. For CD34 ϩ -derived DC, IL-1␤ enhanced CD40L-mediated IL-6 (data not shown) and IL-12p70 but not IL-10, while for CD1b/c ϩ HLA-DR ϩ PBDC, IL-1␤ enhanced CD40L-mediated IL-6 and IL-10 but not IL-12p70 secretion. The lack of detectable IL-12p70 production by PBDC was not due to blunted function as a result of 1) cryopreservation and thaw before use in experiments, 2) expansion in vivo with FL, or 3) isolation from cancer patients, because sorted PBDC isolated from fresh blood from healthy donors also demonstrated low/negligible production of IL-12p70 following stimulation with CD40L (Fig. 6C). Finally, even when stimuli known to potently induce IL-12p70 secretion in MoDC were combined (e.g., GM-CSF plus IL-4 plus CD40L plus IFN-␥ plus intact E. coli) (24,26), they induced low levels of IL-12p70 (Ͻ100 pg/ml) by sorted PBDC in three of three healthy donors (Fig. 6D). IL-1␤ potently enhanced the ability of CD40L-activated MoDC (Fig. 4) and CD34 ϩ -derived DC (Fig. 5C)  . This finding suggests that PBDC are functionally different from MoDC and CD34 ϩ -derived DC and/or may represent a different stage of DC maturation. Finally, PBDC were similar to CD34 ϩ DC in that they both secreted 10-fold lower levels of cytokines following stimulation as compared with MoDC, suggesting that MoDC were functionally distinct from either CD34 ϩ -derived DC and PBDC even when IL-1␤ was present in the stimulation conditions. IL-1␤ not only enhanced CD40L-mediated cytokine secretion but could also synergize with IL-4 (and to a lesser extent with IL-13; n ϭ 3; data not shown) to further enhance CD40L-mediated IL-12p70 secretion by MoDC. The increased IL-12p70 production resulted in increased IFN-␥ secretion by alloreactive T cells following DC stimulation. Furthermore, neutralizing Abs to either IL-1R or IL-4 blocked the ability of CD40L to induce IL-12p70 secretion by MoDC, suggesting that both IL-1 and IL-4 are important cofactors for CD40L-mediated cytokine secretion. The neutralizing effects of anti-IL-4 mAb confirm the findings of Kalinski and colleagues (25).
The effect of IL-1␤ as a cofactor for CD40L-mediated bioactivity for MoDC was recently demonstrated by Wesa and Galy (32). In that study, they demonstrate the role of IL-1␤ on enhancing CD40L-mediated IL-12p70 secretion (32). The present study also demonstrates that monocytes and MoDC were potent producers of IL-1␤ in response to intact E. coli, whereas CD34 ϩ -derived DC and PBDC were not. This suggests that MoDC are functionally related to their monocyte precursors with respect to IL-1␤ secretion, whereas CD34 ϩ -derived DC and PBDC, which are poor producers of IL-1␤, represent functionally distinct APC populations. Furthermore, this finding also identifies a novel mechanism by which monocytes and their progeny (i.e., macrophage or MoDC) can potentiate CD40L-mediated DC activation in vivo by secreting IL-1␤ at sites of infection and/or inflammation. Finally, because certain DC subsets and their precursors (e.g., monocytes) produce IL-1␤ in response to specific stimuli while others do not (CD34 ϩ FIGURE 5. Cytokine secretion by CD34 ϩ -derived DC. CD34 ϩ -derived DC were generated from purified CD34 ϩ progenitor cells in a serum-free culture system containing GM-CSF, TNF-␣, and IL-4 for 14 days. CD34 ϩderived DC (1 ϫ 10 5 /ml) were cultured either in their own CM or in CM with the indicated cytokine stimuli for 3 days. Culture SN were examined for cytokine production by ELISA. A, At day 14, the cultures contained between 35 and 65% CD1a ϩ DC. B, Secretion of IL-10 and IL-12p70. C, Secretion of IFN-␥ by allogeneic T cells stimulated by differentially activated CD34 ϩ -derived DC. CD34 ϩ -derived DC were activated with the indicated cytokine stimuli for 48 h, washed, and then cultured (1 ϫ 10 4 / well) with CD2 ϩ allogeneic T cells (1 ϫ 10 5 /well). At day 5, supernatants were harvested and IFN-␥ secretion was detected by ELISA. Data represent the means Ϯ SEM of experiments from three separate donors. DC and CD1c ϩ PBDC), our findings suggest the potential for immunologic cross talk among different DC subsets located in or recruited to inflammatory sites.
In vivo, the potent combination of CD40L and T cell-derived cytokines (9, 24 -26) required for maximal IL-12p70 production likely requires T cell-Ag interactions either during priming of rare, naive T cells in the lymph nodes or during the restimulation of memory T cells at effector sites. Therefore, an alternative mechanism of rapid CD40L access is required at the ear-liest stages of primary infection where cognate, effector T cells will either be rare or temporally delayed by the requirement for priming in the lymph nodes. Non-T cells such as inflamed somatic cells (smooth muscle cells (21) and vascular endothelial cells (21)) and activated innate immune response effectors (eosinophils (19), activated platelets (20), macrophages (21), and DC (22,23)) can fulfill this role by rapidly providing both CD40L and IL-1␤ at the site of infection/inflammation. Therefore, in this context, the interaction of CD40L with IL-1␤ in the periphery would provide a mechanism for linking innate and adaptive immunity at the earliest stages of the inflammatory response.
Further support that innate immune response-derived factors are important for IL-12p70 secretion in vivo was provided by Schulz and colleagues (27), who demonstrated that CD40L is insufficient at inducing IL-12p70 secretion by DC in vivo but requires an innate signal induced by microbial stimuli. This signal was not IFN-␥, because the effect was not diminished in IFN-␥ Ϫ/Ϫ mice (27). They concluded that although CD40 signaling is important for amplifying IL-12p70 secretion by DC, it is initiated by a microbe-mediated innate signal (27). One possibility is that the microbe-mediated signal inducing IL-12 is direct interaction with the pathogen (40 -43). However, certain pathogens can inhibit IL-12 production (44 -46), and thus alternative IL-12-inducing signals will likely exist to ensure optimal anti-pathogen immune responses. IL-1␤, which is secreted by monocytes, macrophages, and MoDC following microbe stimulation, may represent such an innate signal which can synergize with non-T cell-derived CD40L during the earliest stages of infection and inflammation.
Synergy between CD40L and IL-1␤ has been previously reported in vivo in other settings. For instance, CD40L and IL-1 synergistically enhance the production of IL-6, IL-8, and RANTES, and additively stimulated monocyte chemoattractant protein-1 production during the activation of human tubular renal epithelial cells (47). This suggests a novel role for CD40L and IL-1 in the recruitment of cells into inflammatory sites. Furthermore, IL-1␤ potently enhances CD40 expression on tubular renal epithelial cells (47) and human blood DC (48) and CD40L expression on smooth muscle cells, vascular endothelial cells, and macrophages (21). Conversely, CD40 signaling up-regulates the secretion of bioactive IL-1␤ by smooth muscle cells and endothelial cells (21,49), as well as by monocytes (50) and DC (51,52). These reports all emphasize that CD40 signaling and IL-1␤ are physiologically coupled during various stages of inflammatory responses.
Finally, T cell-independent activation of DC via CD40L and IL-1␤ may lack the Ag-specific dimension of memory T cell responses, thus provoking considerable pathology, including autoimmune damage. This may underlie the inflammatory and immunological responses seen in pathogen-independent diseases such as atherosclerosis. In this regard, CD40L is expressed in atheromas (21,53), while elevated IL-1␤ levels correlate with the severity of atherosclerosis (54 -56). Indeed, IL-12 and IFN-␥ have been found abundantly in atherosclerotic plaques (57,58). In this way, the pathogenesis of atherosclerosis may represent the result of cytokine loops involving IL-1␤, CD40L, and IL-12 without requiring Ag-specific (autoreactive) T cells as primary etiological factors.
The findings reported in this work have several clinical implications. The capacity of nonimmune cells to express CD40L following stress or inflammation, and potentially to induce DC activation in the presence of IL-1␤, suggests the involvement of complex cellular networks at inflammatory sites that cannot easily be studied using in vitro systems. However, fully understanding these networks is critical to identifying how specific immune responses can be optimally generated or blunted, as in the case of immune evasion by pathogens or developing tumors. This study contributes to the understanding of how DC activation may be initially modulated at localized sites of infection in the absence of T cells and presence of innate signals. Our findings suggest a potential role for immunological cross-talk among a variety of different cell types located in or recruited to these inflammatory sites. This study may also highlight dual roles for CD40L in the lymph nodes during priming and in the periphery during inflammatory processes. Finally, the conspicuous absence of inflammation within the tumor microenvironment may point to specific mechanisms used by tumors to evade effective immune targeting. The complete understanding of how innate and adaptive immune responses are linked will be a crucial step in understanding mechanisms underlying autoimmunity, chronic inflammatory diseases, and immune escape by tumors.