Published online before print May 22, 2003
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Departments of
* Pathology and
Medicine, Section of Hematology/Oncology, University of Chicago, Illinois
Correspondence: Thomas F. Gajewski, M.D., Ph.D., University of Chicago, 5841 S. Maryland Ave., MC2115, Chicago, IL 60637. E-mail: tgajewsk{at}medicine.bsd.uchicago.edu
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: IL-12 • cancer • immunotherapy • CD40 • antigen-presenting cells • chemokines
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
However, clinical trials have used different DC preparations, and there is still much debate about the optimal DC phenotype for vaccination. Several DC subsets have been defined that appear to arise from two distinct lineages and also can exist in at least two differentiation states, making it unclear which DC population may offer superior vaccination properties in vivo. In the human, the two DC lineages are termed myeloid (CD11c+CD123-) and plasmacytoid (CD11c-CD123+) [16 , 17 ]. It has been reported that plasmacytoid DC preferentially induce Th2 T cell responses in vitro, whereas myeloid DC induce Th1 responses [18 , 19 ]. As evidence suggests that a Th1-like T cell response is desirable for optimal tumor rejection [20 , 21 ], most efforts are focused on use of myeloid-lineage DC.
Myeloid DC themselves can be derived from several sources and can exist in at least two differentiation states. Bone marrow cells (CD34+ or CD34- populations) [4 ] or peripheral blood monocytes have been used [22 ], and monocytes are the most common source. Monocyte-derived DC are usually generated by enriching peripheral blood mononuclear cells (PBMC) for adherent cells followed by culture for 5–10 days in granulocyte macrophage-colony stimulating factor (GM-CSF) + interleukin (IL)-4 [22 , 23 ]. This population has been termed "immature", as such cells display high antigen-processing capacity but relatively poor T cell activation properties in vitro and lack expression of the DC maturation marker CD83 [24 ]. Immature DC can be further cultured in the presence of inflammatory cytokines, supernatant from activated monocytes, or CD40L, which promotes acquisition of a "mature" phenotype. Mature DC express CD83 and high surface levels of multiple ligands important for T cell activation including CD80/86 [25 ] but lose the ability for large-volume antigen processing [22 ]. As such, they appear to be specialized for T cell activation rather than antigen uptake. DC of both maturation states have been used in clinical trials. The first clinical trial by Nestle and colleagues [14 ] showed a clinical response in five of 16 melanoma patients after vaccination with semimature DC loaded with peptides or tumor extracts. Subsequent clinical trials enrolling patients with kidney cancer [26 ], multiple myeloma [27 ], prostate cancer [13 ], breast/ovarian cancer [28 ], or cervical cancer [29 ] have used immature DC pulsed with peptide or lysate as a vaccine with variable clinical outcomes.
As with immature DC, successful regression of metastases in melanoma patients and/or induction of interferon-
(IFN-)-producing effector T cells has been observed using mature peptide-pulsed monocyte-derived DC [15
, 30
], although another study using mature DC as a vaccine in melanoma patients showed a poor outcome [31
]. Although a recent study has suggested that antigen-loaded, immature DC may be tolerogenic and mature DC, immunogenic [32
], successes and failures have been observed in cancer patients with vaccines using immature or mature DC [12
, 14
, 26
, 33
], making these data difficult to reconcile.
Besides differences in the maturation status of myeloid DC, the commonly used method of DC generation can lead to DC populations of varying quality. By simply culturing the plastic-adherent fraction of patient PBMC for 6–7 days in GM-CSF and IL-4, the cellular constitution of the adherent cells can vary considerably between donors as can the purity of the DC population that emerges. In studies using mature or immature DC prepared in this manner, the average purity has ranged from less than 50% [12 ] up to 95% [15 ], and other studies fall in between [26 , 30 , 34 ]. Thus, in many studies, the potential contribution of nonmonocyte-lineage cells to DC maturation in vitro or to T cell regulation following administration in vivo could be significant. For example, preparations of "immature DC" having contaminating T cells may actually be partially matured, and any DC preparation containing substantial numbers of other cells (B cells, T cells, monocytes) could potentially tolerize rather than immunize following administration.
In addition to surface expression of costimulatory receptors, DC also can produce cytokines that contribute to T cell proliferation and differentiation into effector cells, such as IL-12 and IL-6 [21 , 35 , 36 ]. Cytokine production by DC has most commonly been observed in response to ligation of CD40 by CD40L [37 ], but engagement of Toll-like receptors (TLRs) by pathogen-derived products also can induce IL-12 synthesis [38 39 40 41 ]. It is also thought that DC need to migrate from peripheral tissues into secondary lymphoid organs, where activation of specific T cells likely occurs. This is largely driven by chemokines highly expressed in lymph nodes, such as secondary lymphoid-tissue chemokine/6Ckine or macrophage-inflammatory protein-3ß (MIP-3ß), which engage the chemokine receptor (CCR)7 [42 , 43 ]. CCR7-deficient mice show defective T cell priming and DC migration in vivo [42 ], arguing that CCR7-dependent signals play a dominant role in physiologic migration.
We propose that DC used for immunization be characterized by their biologic qualities rather than the empiric method of generation. In the current study, to begin to determine which human myeloid DC population may have superior biologic properties that could contribute to T cell priming in vivo, we have examined highly purified immature and mature, monocyte-derived DC for expression of costimulatory molecules, production of cytokines in response to CD40 engagement, and migration to CCR7-binding chemokines. The underlying hypothesis is that all three of these biologic properties may be critical for optimal immunization. In fact, we observed that immature DC are far superior at cytokine production, whereas mature DC express high levels of costimulatory molecules and CCR7 and show superior migration to 6Ckine + MIP-3ß. Our results suggest that the same DC population does not optimally coexpress all three hypothetically important DC functions and that immature and mature DC likely possess distinct, CD40-mediated signaling events.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Generation of immature DC
To obtain immature DC, PBMC were first prepared by centrifugation over a lymphoprep gradient. To eliminate an influence of contaminating, nonmonocytoid cells on DC maturation, CD14+ monocytes were purified by negative enrichment using magnetic anti-CD2, -CD3, -CD19, -CD56, and -CD66b and antiglycophorin A beads to eliminate T cells, B cells, natural killer (NK) cells, and granulocytes (Stem Cell Technologies, Vancouver, BC). Isolated monocytes (7x105/ml) were cultured in RPMI with 10% fetal calf serum (FCS), 2 mM L-glutamine, penicillin (100 IU/ml), streptomycin (100 µg/ml), and gentamycin (20 µg/ml) supplemented with GM-CSF and IL-4 (50 ng/ml each; Peprotech, Rocky Hill, NJ) in six-well plates. Medium supplemented with fresh cytokines was changed every 2 days. Cells were harvested at day 6 and designated immature DC. Serum-free medium conditions used AIM-V (Gibco, Grand Island, NY) without exogenous FCS with the same cytokine concentrations and throughout maturation as well.
Monocyte-conditioned medium (MCM)
Autologous MCM was prepared as described previously [25
, 44
]. Briefly, petri dishes were coated with 1% human immunoglobulin G for 1 h at room temperature in phosphate-buffered saline (PBS). Petri dishes were washed with PBS, and PBMC (2x106) were added in 5 ml complete RPMI medium. After incubation for 2 h at 37°C, nonadherent cells were removed by gentle rinsing, and adherent cells were incubated for further 24 h. Conditioned medium was collected, stored at -70°C, and used at 25% v/v for experiments.
DC maturation
To induce DC maturation, immature DC were cultured for an additional 2 days in the presence of MCM (25% v/v) or permutations of prostaglandin E2 (PGE2; 10-6 M; Sigma-Aldrich, St. Louis, MO), tumor necrosis factor 
(TNF-; 50 ng/ml), IL-1ß (10 ng/ml), and IL-6 (1000 U/ml; all from R&D Systems, Minneapolis, MN). Control, immature DC were maintained in GM-CSF and IL-4 for the same period.
Flow cytometric analysis of surface phenotype
Cells were analyzed by three-color flow cytometric analysis using anti-CD11c-fluorescein isothiocyanate (FITC; Dako, Carpinteria, CA) and anti-CD14-peridinin chlorophyll protein (PerCP; Becton Dickinson, Braintree, MA), along with anti-CD83, -CD40, -CD80, -CD86, or -human leukocyte antigen (HLA)-DR monoclonal antibody, each phycoerythrin (PE)-labeled (Becton Dickinson). Samples were acquired using a FACScan flow cytometer (Becton Dickinson) interfaced with an Apple computer. Data were analyzed on 10,000 events using CellQuest software.
Stimulation of DC for cytokine production
Immature DC and DC cultured under the indicated maturation conditions were harvested, washed, and seeded onto a 12-well plate at 0.5 x 106/ml in complete RPMI medium containing GM-CSF and IL-4. After a preincubation period of 1 h in the presence of IFN- (10 ng/ml unless otherwise indicated), fibroblasts transfected with CD40L or CD70L as a control (0.25x106/ml) were added. Alternatively, lipopolysaccharide (LPS; 0.1–1 µg/ml), CpG 2006 (5'-TCG TCG TTT TGT CGT TTT GTC GTT-3'; 1–50 µM; Integrated DNA Technologies, Coralville, IA), poly I:C (10–100 µM, Sigma-Aldrich), or phorbol 12-myristate 13-acetate (PMA; 40–320 ng/ml) + ionomycin (1.6–12.8 µg/ml; Sigma-Aldrich) were used as stimuli. If intracellular flow cytometry was being performed, 2 mM monensin was added 1 h after stimulation, and cells were recovered after 16 h to perform intracellular flow cytometry. For enzyme-linked immunosorbent assay (ELISA), supernatants from cultures without monensin were collected after 16 h and analyzed for IL-12, IL-6, IL-10, IL-8, or IL-15 content.
Intracellular flow cytometric analysis
Monensin-treated, stimulated DC were harvested and stained using anti-CD11c-FITC and -CD14-PerCP antibodies. After fixation of cells in 1.5% paraformaldehyde and 0.2% EDTA in PBS, cells were incubated in permeabilizing buffer (10% FCS, 0.3% saponin in PBS) containing anti-IL-12 (p40/70)-PE or anti-IL-6-PE antibodies (PharMingen, San Diego, CA). After washing the cells with 5% FCS and 0.1% saponin in PBS, cells were resuspended in flow cytometry buffer and analyzed by flow cytometry. Percent cells positive for IL-12 or IL-6 staining among the CD11c+CD14- population was calculated.
ELISA
Production of IL-12 p70, IL-6, IL-10, IL-8, and IL-15 was examined in supernatants by ELISA. Reagents specific for the human IL-12p70 were obtained from PharMingen, and for the other cytokines, from R&D Systems, and were used according to the manufacturer’s instructions.
Migration assay
The measurement of DC migration in response to chemokines was assessed using 24 transwell plates with an 8-µm membrane (Costar, Corning, NY). Cells (1x105) were seeded in RPMI with 10% FCS onto the upper chamber of the transwell system. The lower chamber contained 500 µl RPMI supplemented with combinations of MIP-3ß and 6Ckine (100 ng/ml each, R&D Systems), as indicated in each figure. After an incubation period of 2 h at 37°C, the medium in the lower chamber was harvested, and the number of migrated DC was determined using a hemocytometer. Control cells were kept in the absence of chemokines to assess random migration activity. Results for each experiment are presented as the percentage of the maximal migration, which was always seen with mature DC.
CCR and TLR analysis by reverse transcriptase-polymerase chain reaction (RT-PCR)
Total RNA from immature and mature DC (3–5x106 cells) was isolated using Trizol (Life Technologies, Gaithersburg, MD) to lyse the cells. Following phase separation using chloroform, RNA was precipitated with isopropyl alcohol, washed with 75% ethanol, and resuspended in RNase-free water. cDNA was synthesized using an oligo-dT primer. For semiquantitative PCR of CCR mRNA expression, the following primers were used (Integrated DNA Technologies): for CCR5, sense, 5'-TTA AAA GCC AGG ACG GTA AC-3', and antisense, 5'-TGA ACT TCT CCC CGA CAA AG-3'; for CCR6, sense, 5'-AGG TGG AAG CTG ATG TT-3', and antisense, 5'-TAG TGA AGG ACG ACG CAT TG-3'; for CCR7, sense, 5'-ACA TCG GAG ACA ACA CCA CA-3', and antisense, 5'-GTA CAG GAG CTC TGG GAT GG-3'; and for CXCR4, sense, 5'-TTC CTG CCC ATC TAC TC-3', and antisense, 5'-CCA TGA TGT GCT GAA ACT GG-3'.
For PCR of TLRs 1–10, we used the following primers: TLR1, 5'-TAG TGT GCT GCC AAT TGC TC-3', 5'-TCC AGC TGA CCC TGT AGC TT-3'; TLR2, 5'-GGC CAG CAA ATT ACC TGT GT-3', 5'-TTC AAA AAG ACG GAA ATG GG-3'; TLR3, 5'-CAT GGG TTC CCA GTG AGA CT-3', 5'-GCC AGT TCA AGA TGC AGT GA-3'; TLR4, 5'-CAG AGT TGC TTT CAA TGG CA-3', 5'-TGT TGC ACA TTC CAT TCG TT-3'; TLR5, 5'-CGC TTC TCC TCC TGT AGT GG-3', 5'-AAG AGG GAA ACC CCA GAG AA-3'; TLR6, 5'-TGG ATC TGC CCT GGT ATC TC-3', 5'-GCT GTT CTG TGG AAT GGG TT-3'; TLR7, 5'-ACT CCT TGG GGC TAG ATG GT-3', 5'-GTA GGG ACG GCT GTG ACA TT-3'; TLR8, 5'-TTG GAG ATT TCC GAA GAT GG-3', 5'-TTG CTT TGG TTG ATG CTC TG-3'; TLR9, 5'-CCT CCT CTA CAA ATG CAT CAC T-3', 5'-GTG ACA GAT CCA AGG TGA AGT-3'; TLR10, 5'-GGC ACA GGG TTA GGA AAA CA-3', 5'-CCC AGA AAA GCC CAC ATT TA-3'.
For semiquantification, serial dilutions of cDNA samples were made before PCR. ß-Actin PCR was performed as a control, and cDNA concentration was adjusted in some cases to yield normalized ß-actin results.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Immature and mature DC were analyzed for surface phenotype by flow cytometry. Cells grown in GM-CSF + IL-4 alone were immature, as defined by poor expression of CD83 (9% positive) and relatively lower expression of CD40, CD80, CD86, and HLA-DR. The majority of cells matured in autologous MCM was CD83+ (61% positive) and showed up-regulation of CD40, CD80, CD86, and HLA-DR (data not shown). These results are in keeping with those reported by others [25 , 44 ] and indicate that relatively pure, immature and mature DC populations can be obtained without the influence of significant numbers of contaminating, nonmonocytoid cells.
Immature DC are superior to mature DC at cytokine production in response to CD40L-transfected fibroblasts
CD40 ligation has been shown to be a potent inducer of cytokine production by DC [37
, 45
]. As IL-12 is a major promoter of Th1/Tc1-type T cell responses and as IL-12 appears to be an important contributor to the efficacy of tumor antigen vaccines [46
47
48
49
50
51
], we analyzed the synthesis of IL-12 by immature and mature human monocyte-derived DC in response to CD40L-transfected fibroblasts. CD70L transfectants were used as a control stimulus. ELISA measured IL-12 to be certain that mature p70 protein was in fact secreted. As shown in Figure 1A
, immature DC produced substantially greater IL-12 than did DC matured with MCM. CD70L-transfected fibroblasts did not induce cytokine synthesis in any of the two DC subsets (data not shown). To be sure that the DC and not the CD40L-transfected fibroblasts produced the IL-12, intracellular cytokine staining was performed. As shown in the inset of Figure 1A
, the IL-12 generated in immature DC cultures was produced by the CD11c+ DC, and poor production of IL-12 by mature DC was confirmed by intracellular flow cytometry. Meaningful IL-12 production by mature DC was not induced by the addition of IFN- (50 ng/ml) along with DC40L transfectants (data not shown).
 View larger version (27K): [in a new window] |
Figure 1. Production of cytokines is reduced following DC maturation. Immature DC and DC (0.5x106/ml) matured using MCM were stimulated overnight with CD40L-transfected human fibroblasts (0.25x106/ml) and IFN- (10 ng/ml) and were analyzed for cytokine production. Culture supernatants were collected after 18 h, and IL-12 (p70; A), IL-6 (B), IL-10 (C), and IL-8 (D) were quantified by ELISA. Intracellular cytokine staining for IL-12 and IL-6 in immature (A and B, upper inset) and mature (A and B, lower inset) DC is shown. For FACS, cells were stained with FITC anti-CD11c and PE anti-IL-12 (A, inset) or PE anti-IL-6 (B, inset). The numbers on the dot plots represent the percent of CD11c+ cells expressing the particular cytokine. Similar results were seen in five independent experiments. IL-15 was not detected in any of these supernatants (data not shown). O.D., Optical density.
|
In contrast to cytokine synthesis, mature DC show a higher migration activity to CCR7-binding chemokines
As a third property, we assessed the migration activity of immature and mature DC in response to chemokines MIP-3ß and 6Ckine. The transition of immature to mature DC has been reported to be associated with a switch in CCR expression, such that CCR7 is up-regulated following maturation [42
, 52
]. CCR7 mediates signaling by chemokines preferentially highly expressed in lymph nodes, such as 6Ckine and MIP-3ß, which are thought to be responsible for migration of mature DC to T cell areas.
To assess the responsiveness of immature and mature DC to chemokine-driven migration, we used a transwell system and assessed the number of DC migrating through a membrane into the lower chamber that contained medium supplemented with chemokines. As shown in Figure 2A , using a combination of MIP-3ß and 6Ckine, mature DC showed a substantially higher migration activity than did immature DC. Similar results were observed with chemokine alone and across chemokine doses ranging from 1 ng/ml to 250 ng/ml (data not shown). These results indicate that highly purified, mature DC are superior to immature DC at migration to CCR7-binding chemokines.
 View larger version (17K): [in a new window] |
Figure 2. Migration activity of immature DC and DC matured using MCM in response to the chemokines MIP-3ß and 6Ckine. (A) Migration. Each DC population (0.1x106 cells) was seeded in complete medium without GM-CSF and IL-4 onto a 0.8-µ membrane of a transwell system. Medium of the lower chamber was supplemented with 100 ng/ml MIP-3ß and 6Ckine. As control for random migration activity, no chemokines were added to the lower chamber. After 2 h at 37°C, the contents of the lower chamber were harvested, and the number of cells was determined. Data are given as absolute number of migrated cells and are representative of two independent experiments. (B) CCR expression. Immature DC and DC matured using the cocktail of cytokines were lysed, and cDNA was prepared. PCR was performed with primer pairs specific for CCR5, CCR6, and CCR7. For semiquantitation, PCR was done using fivefold dilutions of the cDNA template from left to right.
|
The combination of TNF-, PGE2, IL-1ß, and IL-6 for DC maturation also induces a defect in IL-12 production in response to CD40L
Although monocyte-conditioned medium is commonly used for in vitro maturation of DC [24
, 25
, 30
], it is a crude mixture and contains a variety of factors that could potentially directly inhibit the capacity of DC to produce IL-12. To be able to dissect the content of MCM and the influence of the different biologically active substances on DC function, we replaced MCM by a defined cocktail of TNF-, PGE2, IL-1ß, and IL-6, as described previously [44
]. DC matured in the cytokine cocktail were even more uniformly mature by surface phenotype compared with DC exposed to MCM, being >98% CD83+ by flow cytometry (Fig. 3A
). DC matured using the cocktail also showed the same biological properties as DC matured using MCM, displaying an even greater diminution in IL-12 production in response to CD40L (Fig. 3B)
and increased migration to the chemokines MIP-3ß + 6Ckine (Fig. 3C)
.
 View larger version (43K): [in a new window] |
Figure 3. DC matured with the cocktail of cytokines TNF-, PGE2, IL-1ß, and IL-6 show the same biological properties as DC matured using MCM. Immature DC were cultured in the presence of MCM or a combination of TNF-, PGE2, IL-1ß, and IL-6 for 2 days. (A) Surface phenotype: Cells were harvested, and a fraction of cells was stained for CD11c, CD14, and PE-labeled antibodies against CD40, CD80, CD83, or CD86. Numbers depicted indicate the mean fluorescence intensity (MFI) for CD40 and CD80 and the percentage of positive cells for CD83 and CD86. (B) Cytokine production: Each DC population (0.5x106 cells) was stimulated with CD40L-transfected fibroblasts and IFN- overnight. IL-12 synthesis is depicted as percentage of 100% IL-12 synthesis in immature DC. (C) Migration to chemokines: Cells (0.1x106) were seeded onto the transwell system for determination of migration activity in response to MIP-3ß and 6Ckine. Migration activity is shown as the percentage of maximal migration activity seen in DC matured using the four-cytokine combination. As a control, cells were seeded onto the transwell membrane without adding chemokines to the lower chamber. These data represent the mean of two independent experiments using two different healthy donors.
|
 View larger version (16K): [in a new window] |
Figure 4. Incubation of mature DC without maturation stimuli for 48 h fails to recover capability of IL-12 synthesis. After generation of immature DC and mature DC using the cocktail of cytokines, mature DC were cultured in complete medium without maturation stimuli or as a control in complete medium in the presence of maturation stimuli for 2 additional days. All three subsets (immature DC, DC matured for 4 days, and DC matured for 2 days and rested for 2 days) were harvested and stimulated for cytokine synthesis as described above. Culture supernatants were harvested after 18 h. IL-12 p(70) was determined by ELISA. Numbers given show the mean ± SEM of two separate cytokine stimulations of one experiment. Similar results were observed by intracellular cytokine staining (data not shown).
|
, PGE2, IL-1ß, and IL-6 is responsible for the loss of IL-12-producing capacity by mature DC, PGE2, IL-1ß, and IL-6 and measured surface phenotype, IL-12 production in response to CD40L, and migration to chemokines.
Regarding single cytokines, IL-1ß showed the highest impact on maturation, resulting in a mean of 47% CD83+ cells (Table 1A
). Cells matured in IL-1ß also showed the greatest diminution of IL-12 production and augmentation of chemokine-dependent migration. The individual effects of TNF-, PGE2, or IL-6 were less marked, but as with IL-1ß, all three biologic properties of DC changed in a correlative manner (Table 1A)
.
|
View this table: [in a new window] |
Table 1. CD83 Expression, IL-12 Synthesis, and Migration Activity of DC Matured with Different Combinations of Cytokines
|
, and IL-1ß was the most active of the three-cytokine mixtures, generating 91% CD83+ cells (Table 1C)
. However, IL-12-producing activity decreased proportionally. The combination of all four factors gave the most reproducible maturation, as determined by CD83 expression, and also gave the greatest gain of chemokine migratory capability and diminution of IL-12-producing function. In total, all three biologic properties of mature DC appeared to occur in a correlative manner, and no combinations of maturation factors preserved IL-12 production in the face of improved chemokine-induced migration and augmented expression of maturation markers.
As a recent report has suggested that the inclusion of IFN- during DC maturation might yield an IL-12-producing DC subset [53
], we also included IFN- as an additional variable. However, in our system using purified monocytes as starting material, we did not observe an increase in IL-12 production by mature DC in response to any combinations of factors along with IFN- (data not shown).
IL-12 synthesis cannot be induced in mature DC using TLR ligands
To investigate whether mature DC, which seem to lose sensitivity to CD40 ligation during maturation, might gain sensitivity to other cytokine-inducing molecules, we examined responsiveness to alternative stimuli. LPS from gram-negative bacteria appear to induce IL-12 synthesis via TLR4 engagement [54
]. Another microbial molecule, bacterial DNA containing unmethylated CpG, has been identified to engage TLR9 [39
]. Therefore, we stimulated immature and mature DC with LPS or CpG DNA and analyzed IL-12 production by ELISA. As shown in Figure 5
, LPS induced IL-12 production by immature but not by mature DC, and CpG DNA failed to induce IL-12 production by either DC subset. Other stimuli (poly I:C [40
] and PMA±ionomycin) also failed to stimulate IL-12 synthesis in either DC subset (data not shown).
 View larger version (17K): [in a new window] |
Figure 5. LPS and CpG DNA fail to induce IL-12 production by mature DC. Immature and mature DC were stimulated as above overnight with CD40L-transfected fibroblasts and IFN- or with LPS (1 µg/ml) or CpG DNA (50 µM). Culture supernatants were collected, and ELISA quantified IL-12 (p70) concentration. All three stimuli induced only minimal IL-12 synthesis in mature DC (CD40L+IFN-: 0.005 ng/ml; LPS: 0.007 ng/ml; CpG: 0.005 ng/ml). CpG DNA failed to induce IL-12 production by immature DC as well. Data show the concentration of IL-12 in the supernatants measured by ELISA of one representative experiment out of three.
|
 View larger version (50K): [in a new window] |
Figure 6. mRNA expression of TLRs in PBMC, immature DC, and mature DC. DC (3x106) were immature or matured using the cytokine cocktail. cDNA was prepared, and RT-PCR was performed using the primer sets described in Materials and Methods. The amount of template used was equalized by the amplification of ß-actin. cDNA from PBMC was used as a comparison. For semiquantitation, PCR was performed using fivefold dilutions of cDNA from left to right. All reactions were run at the same time under the same conditions. Similar results were obtained from a second donor.
|
, and PGE2, showed expression of surface marker CD83 (Fig. 7A
). Expression of CD40 and the costimulatory molecules CD80 and CD86 was up-regulated as was seen with DC from healthy donors. IL-12 (p70) production after stimulation with CD40L-transfected fibroblasts was superior in immature DC compared with mature DC (Fig. 7B)
. Finally, migration in response to MIP-3ß and 6Ckine was assessed and revealed that mature DC were superior in migration activity compared with immature DC (Fig. 7C)
. These results are all in accordance with our data obtained from healthy donors and suggest that the presence of advanced melanoma or kidney cancer does not interfere with the ability to generate pure populations of immature and mature DC.
 View larger version (22K): [in a new window] |
Figure 7. Surface phenotype, IL-12 synthesis, and migration activity of monocyte-derived, immature and mature DC from cancer patients. Immature and mature DC were generated using peripheral blood of cancer patients with metastatic disease. DC were matured for 48 h in the presence of the cytokine cocktail. Surface phenotype (A) was determined by flow cytometry; IL-12 synthesis (B) was determined following stimulation with CD40L-transfected fibroblasts and IFN- overnight, followed by ELISA. Immature DC show higher IL-12 synthesis (2.3 ng/ml±0.6) than mature DC (0.09 ng/ml±0.06). Migration activity (C) was measured in response to MIP-3ß and 6Ckine and is shown as the percentage of migration activity compared with that seen with mature DC. All data represent the mean of five independent experiments with DC generated from five different donors.
|
 View larger version (37K): [in a new window] |
Figure 8. Characteristics of immature and mature DC generated in AIM-V medium. Immature and mature DC were generated using peripheral blood of normal donors in AIM-V medium instead of RPMI/FCS. DC were matured for 48 h in the presence of the cytokine cocktail. Surface phenotype (A) was determined by flow cytometry; IL-12 synthesis (B) was determined following stimulation with CD40L-transfected fibroblasts and IFN- overnight, followed by ELISA. Migration activity (C) was measured in response to MIP-3ß and 6Ckine and is shown as the number of migrated cells. Similar results were observed in an additional donor.
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
We reasoned that understanding which DC preparation offers superior immunization potential in humans in vivo may require generating DC populations of greater uniformity and also may benefit from describing the functional characteristics of each DC preparation so that a correlation between DC properties and clinical outcome can be established. Our previous work has supported an important role for IL-12 in promoting optimal antitumor immunity in murine models and in a recent vaccine trial in patients with melanoma [47 48 49 ]. We therefore began by seeking the conditions that resulted in the highest IL-12-producing DC subset but also examined expression of costimulatory molecules and CD83 as well as migration to CCR7-binding chemokines. Although highly purified, mature (CD83+) DC did up-regulate the costimulatory molecules CD80 and CD86 and also up-regulated CD40, they surprisingly showed approximately tenfold lower production of bioactive IL-12 (p70) following CD40 ligation. This was not confined to IL-12, as IL-6 and IL-10 production was also superior by immature DC. It also was not restricted to CD40L as a stimulus, as LPS also stimulated greater IL-12 production by immature DC. In contrast to cytokine production, mature DC were far superior at migrating in response to 6Ckine and MIP-3ß, two CCR7-binding chemokines known to be critical in DC trafficking to secondary lymphoid organs.
It recently has been reported that DC from IL-12-deficient mice were fully capable of priming for antitumor T cell responses when used to vaccinate mice in vivo [55 ]. Although at first glance, this may imply that IL-12 is dispensible for antitumor immunity, a more recent study has demonstrated that IL-12 from host cells becomes required for type-1 T cell induction when IL-12-deficient DC are used [56 ]. It thus continues to seem prudent to ensure that IL-12 is provided during vaccination if feasible.
Sallusto and Lanzavecchia [57] recently suggested that DC matured with CD40L were "exhausted" and thus failed to produce IL-12, perhaps as a result of refractoriness to CD40 signaling. Our current results question this model and rather suggest that DC matured with any combination of factors universally lose the ability to produce IL-12. Thus, lack of IL-12 production appears to be a consistent and stable property of mature DC. We observed that mature DC rested for 48 h did not recover IL-12-producing capability. In preliminary experiments, we tested the notion that mature DC may need to migrate in response to chemokines to reacquire the ability to make IL-12, as they might in a lymph node. However, exposure of mature DC for 48 h to MIP-3ß and 6Ckine also failed to restore IL-12 production (data not shown). We also examined whether IL-12 was produced during cytokine-induced maturation, as had been reported to occur upon CD40L-induced maturation, but meaningful IL-12 was not detected during the maturation culture (data not shown). Thus, the cytokine-producing potential of fully mature DC is distinct from that of immature DC or during CD40L-mediated DC maturation. It might be predicted that this poor cytokine-producing capability would be preserved following administration of fully mature DC as a vaccine in vivo.
We also examined IL-12 production in response to TLR ligands. It has recently been shown that TLRs activate DC to up-regulate costimulatory molecules and produce inflammatory cytokines such as IL-12 [58 59 60 ]. In our current study, the TLR4 ligand LPS and the TLR9 ligand CpG DNA failed to induce IL-12 synthesis in mature DC. We further assessed the expression of TLR1-10 [54 ] in immature and mature DC by RT-PCR. Except for TLR8, which seemed to be ubiquituosly expressed, and TLR7, which was not expressed in either of the two DC populations, mRNA expression of all other TLR receptors was down-regulated in mature DC compared with immature DC. This confirms the failure of IL-12 synthesis after LPS or CpG stimulation in mature, monocyte-derived DC in concordance with recently published data [38 , 54 , 60 ]. The maintained expression of TLR8 in mature DC makes identification of a ligand for this receptor an attractive goal, as it may modulate the function of DC after maturation.
A current model for IL-12 production in response to CD40 engagement suggests that this should physiologically occur in a lymph node upon interaction between mature antigen-expressing DC and naive T cells, both of which would have migrated in response to CCR7-binding chemokines. However, if immature and mature DC in vitro correspond to tissue-resident and post-migratory lymph node DC in vivo, then our current results suggest that the major timing of IL-12 production would be within the tissue by immature DC undergoing activation by pathogen exposure and inflammation before migration to lymph nodes. If so, then a major function of IL-12 may in fact be to regulate the innate-immune response. Consistent with this notion, IL-12 has been shown to induce murine IFN--producing NKT cells, which are potent, cytotoxic effectors against tumoral targets [61
]. The production of IL-12 by neutrophils is also consistent with a role for this cytokine early during inflammation [62
, 63
]. Although IL-15 has also been reported to play a role during innate immunity [64
], immature or mature DC did not detectibly produce IL-15 in our present study.
It was surprising that mature DC up-regulated expression of CD40 yet showed poor cytokine production in response to stimulation with CD40L. These results suggest that immature and mature DC display distinct CD40-mediated signal-transduction pathways or different transcription factor activities, thus disallowing expression of the IL-12 p35/p40 or IL-6 genes. Although IFN- was also included in our DC stimulations, cytokine production in response to CD40L transfectants alone showed the same pattern (data not shown). It is conceivable that some other property of fibroblasts also could have contributed to DC cytokine production in response to CD40L transfectants. However, preliminary results have suggested that although purified CD40L trimer failed to induce IL-12 production by DC, CD40L-containing membranes purified from transduced insect cells did induce IL-12 production and only from immature DC (data not shown). Further work will be necessary to dissect the molecular basis for deficient CD40 signaling in mature DC.
The hypothesis that mature DC may be more efficient in inducing an antitumor-immune reponse is supported by two reports about the induction of regulatory T cells by immature DC in vitro [65 ] or in vivo [66 ]. Dhodapkar and colleagues [66 ] described the inhibition of influenza matrix peptide (MP)-specific CD8+ T cell effector function and the appearance of MP-specific IL-10-producing cells after immunization of two individuals with immature DC pulsed with MP and keyhole limpet hemocyanin. In contrast, Biragyn and colleagues [67 ] demonstrated in a murine model that nonantigenic peptides, fused with ligands that target receptors preferentially expressed on immature DC, are immunogenic, whereas antigens fused with ligands for receptors on mature DC were not. These observations, along with reports of clinical activity in patients immunized with antigen-loaded, immature DC, continue to question which DC population is optimal. Our results demonstrate that highly purified, immature and mature monocyte-derived DC can be obtained from patients with advanced melanoma and kidney cancer and that these DC subsets show characteristics that are similar to those obtained from normal donors. These observations pave the way for vaccine clinical trials using highly purified and biologically characterized, immature DC, mature DC, or a mixture of both to determine which biologic properties of DC correlate with superior T cell priming and antitumor activity in patients.
Received July 9, 2002; revised December 2, 2002; accepted January 6, 2003.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  S. Takenaka, E. Safroneeva, Z. Xing, and J. Gauldie Dendritic Cells Derived from Murine Colonic Mucosa Have Unique Functional and Phenotypic Characteristics J. Immunol., June 15, 2007; 178(12): 7984 - 7993. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. P. Moran, M. Collier, K. P. McKinnon, N. L. Davis, R. E. Johnston, and J. S. Serody A Novel Viral System for Generating Antigen-Specific T Cells J. Immunol., September 1, 2005; 175(5): 3431 - 3438. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
