The JI
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     
 

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Miller, G.
Right arrow Articles by DeMatteo, R. P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Miller, G.
Right arrow Articles by DeMatteo, R. P.
The Journal of Immunology, 2002, 169: 2875-2885.
Copyright © 2002 by The American Association of Immunologists

Endogenous Granulocyte-Macrophage Colony-Stimulating Factor Overexpression In Vivo Results in the Long-Term Recruitment of a Distinct Dendritic Cell Population with Enhanced Immunostimulatory Function1

George Miller*, Venu G. Pillarisetty*, Alaap B. Shah*, Svenja Lahrs*, Zhou Xing{dagger} and Ronald P. DeMatteo2,*

* Hepatobiliary Service, Memorial Sloan-Kettering Cancer Center, New York, NY 10021; and {dagger} Department of Pathology and Molecular Medicine, McMaster University, Hamilton, Ontario, Canada


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
GM-CSF is critical for dendritic cell (DC) survival and differentiation in vitro. To study its effect on DC development and function in vivo, we used a gene transfer vector to transiently overexpress GM-CSF in mice. We found that up to 24% of splenocytes became CD11c+ and the number of DC increased up to 260-fold to 3 x 108 cells. DC numbers remained substantially elevated even 75 days after treatment. The DC population was either CD8{alpha}+CD4- or CD8{alpha}-CD4- but not CD8{alpha}+CD4+ or CD8{alpha}-CD4+. This differs substantially from subsets recruited in normal or Flt3 ligand-treated mice or using GM-CSF protein injections. GM-CSF-recruited DC secreted extremely high levels of TNF-{alpha} compared with minimal amounts in DC from normal or Flt3 ligand-treated mice. Recruited DC also produced elevated levels of IL-6 but almost no IFN-{gamma}. GM-CSF DC had robust immune function compared with controls. They had an increased rate of Ag capture and caused greater allogeneic and Ag-specific T cell stimulation. Furthermore, GM-CSF-recruited DC increased NK cell lytic activity after coculture. The enhanced T cell and NK cell immunostimulation by GM-CSF DC was in part dependent on their secretion of TNF-{alpha}. Our findings show that GM-CSF can have an important role in DC development and recruitment in vivo and has potential application to immunotherapy in recruiting massive numbers of DC with enhanced ability to activate effector cells.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Dendritic cells (DC)3 are potent APC that initiate T cell-mediated immune responses. DC are ubiquitous and may be isolated from a variety of organs including the spleen, liver, kidney, heart, and lymph nodes. The diversity of DC subsets in each organ is complex and their phenotypic and functional characterization is the subject of considerable current investigation. In the mouse spleen at least three distinct DC subsets have been identified. CD11c+CD8{alpha}+ CD11bdull/-DEC205+ DC (putative lymphoid related) are located in the T cell-rich areas of the periarteriolar lymphatic sheaths (PALS), whereas CD11c+CD8{alpha}-CD11bbrightDEC205- DC (putative myeloid related) reside in the marginal zone (1, 2, 3, 4, 5, 6). The latter group may be subdivided further into a 33D1+CD4+ subset and a 33D1-CD4- subset. Regardless of subset, freshly isolated splenic DC express low to moderate levels of MHC class II, CD54, and the costimulatory molecules CD40, CD80, and CD86. After overnight culture, all DC subsets mature considerably. CD8{alpha}+ DC populations are known to express high levels of IL-12 upon exposure to a variety of bacterial pathogens or LPS (4, 7, 8, 9). Furthermore, IL-12 secretion from CD8{alpha}+ DC acts in an autocrine fashion to induce DC release of IFN-{gamma} (7). However, CD8{alpha}- DC are also capable of IL-12 secretion depending on the antigenic stimulus (10, 11). Both DC subsets take up and process soluble Ags but CD8{alpha}- DC are more efficient at capturing particulate Ag (4, 6, 12). The relative capacity of CD8{alpha}+ and CD8{alpha}- DC to stimulate T cells is controversial. Although both potently stimulate allogeneic T cells (12), the CD8{alpha}+ subset induces apoptosis in CD4+ T cells in vitro (13) and is unable to induce cytokine production from CD8+ T cells (14). In vivo, both DC subsets can prime Ag-specific CD4+ T cells; however, CD8{alpha}+ DC generally elicit a Th1 response while CD8{alpha}- DC mediate a Th2 or Th0 response (8).

Although the classical stratification of DC into myeloid (CD8{alpha}-CD11bbright) and lymphoid (CD8{alpha}+CD11bdull/-) subsets has a phenotypic, anatomic, and functional basis (15), recent findings have led investigators to question whether the subsets are developmentally distinct. Traver et al. (16) showed that both CD8{alpha}+ and CD8{alpha}- DC can arise from clonogenic common myeloid progenitors in the spleen and thymus. Martinez et al. (17) also showed that CD8{alpha}- DC can transform into CD8{alpha}+ DC in vivo. The precise factors that regulate the commitment of DC precursors to a CD8{alpha}+ or CD8{alpha}- pathway are unknown (18).

GM-CSF is a critical growth factor for DC in vitro. GM-CSF has also been shown to enhance the ability of DC to present Ag and to stimulate T cells in culture (19, 20). However, the role of GM-CSF in DC generation and function in vivo is less certain. Initial studies using mice deficient in GM-CSF or GM-CSF-transgenic mice suggested that GM-CSF was not vital for DC development in vivo (5, 21). Similarly, Maraskovsky et al. (22) found only modest elevations in splenic DC after daily injection of GM-CSF protein. However, daily injections of GM-CSF conjugated to polyethylene glycol, which has a half-life of several hours, produced a short-lived 8-fold increase in splenic DC, suggesting that a threshold serum level of GM-CSF may be required to exert biologic effects on DC recruitment (12). Similarly, administration of tumors transduced to express GM-CSF have been shown to recruit DC locally (23, 24, 25).

Because GM-CSF plays a critical role in DC generation from bone marrow in vitro and has been shown to recruit DC in mice, we examined the effects of continuous endogenous secretion of GM-CSF on DC development in vivo. We found that transient (<2 wk) overexpression of GM-CSF in mice produced massive recruitment (up to 3 x 108 DC) of a CD11c+CD4- DC population to the spleen. Strikingly, recruited DC secreted high amounts of TNF-{alpha}. Other murine DC populations are not known to constitutively produce elevated levels of TNF-{alpha}. Recruited DC also had an enhanced ability to capture Ag and to stimulate T cells and NK cells. Their enhanced immunostimulation depended in part on their secretion of TNF-{alpha}. Our findings show that endogenous GM-CSF recruits a distinct DC population that has altered secretory properties but is highly functional. Additionally, because of its large and sustained (>75 days) effect on DC recruitment, endogenous overexpression of GM-CSF may have an important role in immunotherapy.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
DC isolation and flow cytometry

Single-cell suspensions of splenocytes were prepared by splenic injection of collagenase (Sigma-Aldrich, St. Louis, MO) and mechanical disruption. DC were then purified using anti-CD11c immunomagnetic microbeads and high-gradient LS separation columns (Miltenyi Biotec, Auburn, CA). The purity of isolated DC was >85% by CD11c staining. DC were either used immediately or after culture in complete medium (RPMI 1640 with 10% heat-inactivated FBS, 2 mM L-glutamine, 100 U/ml penicillin, 100 µg/ml streptomycin, and 0.05 mM 2-ME). Flow cytometry was performed on an EPICS-XL flow cytometer (Beckman Coulter, Fullerton, CA) after incubating 5 x 105 DC/tube with 1 µg of Fc block (anti-CD16/CD32 (2.4G2), Monoclonal Ab Core; Sloan-Kettering Institute, New York, NY) and then labeling with 1 µg of FITC- or PE-conjugated Ab. Splenocytes were stained for DC (CD11c (HL3)), B cells (B220 (RA3-6B2) and CD19 (SJ25C1)), NK cells (NK1.1 (PK136)), macrophages (Mac-3 (M3/84)), T cells (CD90.2 (Thy1.2), CD4 (Gk1.5) and CD8{alpha} (53–6.7); all from BD PharMingen, Franklin Lakes, NJ). Splenocytes or purified DC were also stained for MHC class I (H-2Kb) and class II (I-Ab), intercellular adhesion marker CD54 (ICAM-1 (3E2)), CD40 (3/23), CD80 (B7-1 (1G10)), CD86 (B7-2 (GL1)), CD11b (M1/70), Ly-6G (Gr-1 (RB6-8C5)), and CD16/CD32 (2.4G2; all from BD PharMingen).

Recombinant adenoviruses

Recombinant adenoviruses were propagated, purified, and stored as previously described (26). Adenovirus encoding the murine GM-CSF transgene (AdGM) (27) and another Ad vector encoding green fluorescent protein (AdGFP; Quantum Biotechnologies, Montreal, Quebec, Canada) each contain the CMV promoter. Ad encoding murine Flt3 ligand (AdFlt3L) was constructed as follows: a plasmid containing the Flt3L cDNA was obtained from the National Gene Vector Laboratories (Ann Arbor, MI). The SalI fragment was ligated into the SalI site of the shuttle plasmid pDC316 (Microbix, Toronto, Ontario, Canada) that contains a portion of the adenoviral genome and the murine CMV promoter (26). Cre-mediated homologous recombination at specific loxP sites was then performed on 293 human embryonic kidney cells (American Type Culture Collection (ATCC), Manassas, VA) with pBHGlox (Microbix) that contains the adenovirus type 5 genome with E1 and E3 region deletions. DNA was isolated from candidate viral plaques and screened with the restriction enzymes to confirm the presence of the Flt3L transgene. Dose for all viruses was calculated as multiplicity of infection, which equals the number of virus particles per target cell. Endotoxin was undetectable in viral stocks using the Limulus amebocyte lysate clot test (sensitivity 6 pg/ml; Associates of Cape Cod, Woods Hole, ME).

Immunohistochemistry

Tissue was placed in OCT medium (Miles, Elkhart, IN) and frozen in an isopentane/dry ice bath. Cryostat sections (8 µm) were fixed in Formalin, rinsed with PBS, blocked with 0.3% H2O2, and then stained for DC using a biotin-conjugated anti-CD11c Ab (BD PharMingen) for 1 h at room temperature. Afterward, sections were incubated with streptavidin-HRP for 30 min at room temperature and visualized using diaminobenzidine substrate solution (BD PharMingen) for 5 min. Slides were counterstained in hematoxylin and dehydrated in ethanol and mounted. Photographs were taken with an Olympus SZH10 stereoscope (Olympus, Melville, NY) and a Zeiss Axioplan 1 microscope (Zeiss, Thornwood, NY). For cytospins, DC were spun at 500 rpm onto ProbeOn slides (Fischer Scientific, Pittsburgh, PA), fixed in Formalin and then in 100% methanol, and visualized with Giesma stain.

Cytokine measurement and Ag uptake assays

Serum or cell culture supernatant was tested by ELISA for IL-4 (BD PharMingen), IL-6, IL-10, IL-12 (p70), TNF-{alpha}, GM-CSF, Flt3L, and IFN-{gamma} (all R&D Systems, Minneapolis, MN) according to the respective manufacturer’s protocol. For in vitro cytokine assays, freshly isolated DC were cultured in 24-well dishes at a concentration of 1 x 106 cells/ml. Supernatant was harvested after 24 h of culture. LPS (Sigma-Aldrich) was used at 10 ng/ml and TNF-{alpha} (R&D Systems) was used at 100 ng/ml. For in vitro Ag uptake assays, DC (2 x 105) were incubated with FITC-dextran, FITC-albumin, or FITC-mannose albumin (all Sigma-Aldrich) for various durations at 37°C at a concentration of 1 mg/ml (28). To stop reactions, cells were harvested and placed on ice. In vivo Ag uptake assays were performed by i.v. injecting 2.5 mg of FITC-dextran into mice. After 30 min, splenic DC were harvested and analyzed by flow cytometry.

T cell proliferation and CTL assays

For MLR, DC were irradiated (3000 rad) and added at various amounts to 3 x 105 syngeneic or allogeneic T lymphocytes (purified using Thy1.2 (CD90.2) immunomagnetic microbeads (Miltenyi Biotec)) in 96-well plates and then pulsed with thymidine (1 µCi/well) on day 3 for 20 h. For some assays, TNF-{alpha} blockade was performed using the clone 2E2 (29) at 50 µg/ml. 2E2 was produced in the Monoclonal Ab Core Facility at Memorial Sloan-Kettering Cancer Center. For Ag-specific T cell stimulation assays, an H-2Kb-restricted CD8+ T cell hybridoma specific for OVA257–264 peptide was used (30). DC were incubated with either OVA (10 µg/ml; Peptide Synthesis Core, Sloan-Kettering Institute) or OVA protein (2 mg/ml; Sigma-Aldrich) for 90 min before being irradiated (3000 rad) and plated at various concentrations with 5 x 104 OVA-restricted T cells in a 96-well plate for 2 days. T cell stimulation was determined by assessing IL-2 levels in its supernatant by ELISA (BD PharMingen). CTL assays were performed as described with modifications (31). Briefly, splenocytes from treated and control animals were harvested and plated at 5 x 106 cells/well in 24-well plates with 10 µg/ml OVA for 5 days. Afterward, effectors were harvested and tested against 1 x 104 51Cr-labeled EL4.OVA (EG7) or EL-4 (both from ATCC) target cells for 4 h in 96-well plates. After the incubation, 30 µl of supernatant was transferred to Luma plates (Packard Bioscience, Meriden, CT) and radioactivity was read in a TopCount NXT gamma counter (Packard Bioscience). Percent lysis was calculated according to the following formula: percent specific lysis = ((cpm experimental - cpm spontaneous release) x 100)/(cpm maximum release - cpm spontaneous release). Spontaneous release was always <10% of maximum release. All assays were done in triplicate and repeated at least three times.

NK coculture and cytotoxicity assays

DC-NK coculture assays were performed as described previously (32) with slight modifications. Briefly, 1 x 106 splenic NK cells were isolated using DX5 immunomagnetic microbeads and high-gradient LS separation columns (Miltenyi Biotec) and plated with 2 x 106 DC in 24-well plates in a total of 700 µl of complete RPMI 1640 medium for 18 h. The purity of the isolated NK cell populations was >85% by FACS analysis using PE-conjugated NK1.1 and DX5 Abs (both BD PharMingen). After coculture, NK cells were harvested and tested against 3 x 103 51Cr-labeled Yac-1 cells (ATCC). Spontaneous release and maximum release were assayed in a manner similar to the CTL assays.

Animals procedures and tumor models

Male C57BL/6 (H-2Kb) and BALB/c (H-2Kd) mice (6–10 wk old) were purchased from Taconic Farms (Germantown, NY). All procedures were approved by the Institutional Animal Care and Use Committee. Animals were given a single tail vein injection of recombinant adenovirus or saline. All adenovirus injections were at a dose of 8 x 1010 particles unless otherwise indicated. For some tumor experiments, mice were administered an intrasplenic injection of 2.5 x 104 B16F10 murine melanoma cells (B16) or 5 x 104 CT26 colorectal tumor cells (both from ATCC) via a left flank incision followed by splenectomy. For other tumor experiments, mice were challenged with a s.c. flank inoculation of 1 x 105 CT26 or 3.5 x 105 EL4.OVA cells. For DC immunization experiments, mice were given two i.p. immunizations of 5 x 105 DC pulsed with OVA at 1-wk intervals and then 1 wk later sacrificed for CTL assays or challenged with a flank injection of 3.5 x 105 EL4.OVA cells.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
AdGM administration produces transiently high serum GM-CSF levels

To determine the amount of GM-CSF produced in vivo after i.v. administration of AdGM, we tested several viral doses (Fig. 1GoA). Serum levels of GM-CSF became detectable (>8 pg/ml) after injection of 5 x 109 virus particles and markedly higher GM-CSF levels were noted as the dosage was increased. Three days after injection of 2 x 1011 viral particles, GM-CSF levels reached 55 ng/ml. However, this dose produced fatal toxicity in nearly 50% of mice within 2 wk. A dose of 8 x 1010 particles was therefore chosen for further studies. This dose is known to accomplish gene transfer to ~90% of murine hepatocytes (31). Next, we determined the time course and duration of GM-CSF expression. After inoculation of 8 x 1010 particles of AdGM, serum GM-CSF levels peaked at 8 ng/ml by day 3, but dropped precipitously to 0.5 ng/ml by day 7 and were no longer detectable by 2 wk (Fig. 1GoB). Comparatively, recombinant GM-CSF protein has a half-life in vivo of <12 min and even GM-CSF stabilized by polyethylene glycol modification has a half-life of only ~5 h (12).



View larger version (14K):
[in this window]
[in a new window]
  		FIGURE 1. AdGM produces marked but transient elevations in serum GM-CSF. A, Various doses of an adenovirus encoding GM-CSF were administered to mice by i.v. injection. Serum GM-CSF concentration was determined by ELISA on days 3 and 7 after treatment. A dose of 2 x 1011 particles resulted in a serum protein concentration of 55 ng/ml on day 3. Administration of saline or AdGFP did not produce detectable serum GM-CSF. Averages of two mice per dose are shown. B, Serum GM-CSF levels were determined at various time points after an i.v. injection of 8 x 1010 particles of AdGM. GM-CSF levels peaked by day 3 after treatment and returned to baseline by day 14. Averages of five mice per time point are shown.

 
GM-CSF recruits massive numbers of DC to the spleen

To determine whether endogenous secretion of GM-CSF results in DC recruitment to the spleen, we inoculated mice with various doses of AdGM and determined the number of splenocytes that stained for CD11c by flow cytometry (Fig. 2GoA). At doses below 4 x 1010 particles, the total number of splenocytes increased 2- to 4-fold and modest increases were also noted in the percentage of CD11c+ cells (from 1–2% to 7–9%). Conversely, at doses of 8 x 1010 particles or higher, the number of splenocytes increased >10-fold and the percentage of CD11c+ cells increased to 14–24%. This resulted in a 200- to 260-fold increase (depending on the experiment) in the total number of splenic DC. In contrast, the percentage of T cells, NK cells, and macrophages were unchanged while there was a sharp decrease (66–38%) in the percentage of splenic B cells (Table IGo). However, the absolute numbers of all cell types was increased. This resulted in an 8-fold increase in splenic weight compared with saline-treated mice. Only a small fraction of the effects was due to adenovirus alone (Table IGo).



View larger version (15K):
[in this window]
[in a new window]
  		FIGURE 2. Endogenous secretion of GM-CSF results in the long-term recruitment of large numbers of DC to the spleen. A, Mice were treated with various i.v. doses of AdGM and their splenocytes were counted (after RBC lysis) 7 days later and the percentage of CD11c-positive cells was determined by flow cytometry. Up to 3 x 108 DC were recruited at the highest dose. Averages of two animals per group are shown. B, Mice were given a single injection of saline or 8 x 1010 particles of AdGFP, AdFlt3L, or AdGM and the number of splenic DC was determined at various time points. The numbers of DC were highest on day 7 but still remained elevated by >15-fold over controls by 75 days after treatment. Averages of two animals per group are shown. C, The total number of splenocytes was also elevated for an extended duration after treatment with 8 x 1010 particles of AdGM. The percentages of various splenic leukocytes at later time points were similar to the representative data from day 7 shown in Table IGo.

 

View this table:
[in this window]
[in a new window]
  		Table I. Composition of splenocytes 1 wk after treatment with saline or virus (8 x 1010 particles)a

 
To determine the time course and duration of DC recruitment after AdGM administration, we treated a cohort of mice with 8 x 1010 viral particles and analyzed their splenocytes at weekly intervals for DC recruitment. The number of splenic DC was greatest at 1 wk after AdGM administration (2–3 x 108 DC). However, DC levels remained elevated by >15-fold over baseline for >75 days after a single dose of virus (Fig. 2GoB). This was striking especially considering that serum GM-CSF levels returned to baseline by day 14 (Fig. 1GoB). In contrast, in mice treated with AdFlt3L, which produced a peak day 3 serum level of 1300 ng/ml murine Flt3L, DC numbers returned to normal by 5 wk after injection. Notably, AdGFP injection alone resulted in only a 2-fold increase in the number of splenic DC and this lasted ~4 wk. The minimal DC recruitment effect of systemic adenovirus alone did not appear to be from adenoviral-induced secretion of GM-CSF or Flt3L because serum levels of these cytokines were undetectable after AdGFP injection. In addition to expanding the splenic DC population for an extended duration, treatment with AdGM also increased the absolute numbers of all splenic leukocytes for a comparable period (Fig. 2GoC). In contrast, mice treated with AdGM had fewer bone marrow cells 7 days later (data not shown). Subsequently, these bone marrow progenitors generated only 18% as many bone marrow-derived DC as saline controls by 8 days of culture.

AdGM recruits distinct DC subsets

We next compared the phenotype of saline, AdGFP, AdFlt3L, and AdGM recruited splenic DC. There were major differences in the DC subsets that were recruited among the groups (Fig. 3GoA). GM-CSF recruited DC were mostly CD8{alpha}-CD11bbright consistent with a myeloid origin while Flt3L DC were predominantly CD8{alpha}+ CD11bdim/-with a smaller population that was CD8{alpha}-CD11bbright. Cross-staining of CD4 and CD8{alpha} revealed additional differences between the groups (Table IIGo). GM-CSF DC were 72% CD8{alpha}-CD4- and the remainder were essentially CD8{alpha}+CD4-. In contrast, 20–27% of DC from saline- and AdGFP-treated mice expressed CD4. B220 expression (but not CD19 which is more restricted to B cells) was present in CD11c+ DC from control mice but GM-CSF-recruited DC had minimal staining (Fig. 3GoB). In contrast, GM-CSF recruited DC had high staining for CD16/CD32, the Fc{gamma}III/II receptor (33), compared with controls necessitating blockade of this receptor to reduce nonspecific Ab binding during flow cytometric analysis.



View larger version (33K):
[in this window]
[in a new window]
  		FIGURE 3. Endogenous GM-CSF recruits distinct DC subsets. A, Animals were treated with saline or 8 x 1010 particles of AdGFP, AdFlt3L, or AdGM. Splenic DC were isolated 7 days later with immunomagnetic beads and analyzed by flow cytometry in comparison to isotype controls which were set within the first decade (data not shown). A, DC were dual stained for CD8{alpha} and CD11b. DC recruited by GM-CSF were mostly CD8{alpha}-CD11bbright but CD8{alpha}+CD11bdull/- DC were also expanded. In contrast, a much higher percentage of DC recruited by Flt3L were CD8{alpha}+CD11bdull/-. B, Unfractionated GM-CSF-recruited DC expressed lower levels of B220 than controls but considerably higher levels of the Fc{gamma}III/II receptor CD16/CD32. CD11c+B220+ cells had minimal cross-staining for Gr-1 (<2%, data not shown), suggesting that the majority were not plasmacytoid DC. Data are representative of experiments repeated more than four times.

 

View this table:
[in this window]
[in a new window]
  		Table II. Percentage of CD8{alpha} and CD4 expression by splenic DC

 
We next examined the maturational level of recruited DC (Fig. 4Go). Freshly isolated GM-CSF-recruited DC were similar to DC from saline-treated animals in terms of expression of MHC class I, class II, CD40, and CD54 but they had slightly higher CD80 (93 vs 76%) and CD86 (86 vs 50%) expression. However, treatment with AdGFP alone resulted in small elevations in DC MHC and costimulatory molecule expression. Compared with AdFlt3L-recruited DC, GM-CSF DC had weaker MHC class I, CD40, CD54, and CD86 expression but expressed slightly stronger CD80. The flow cytometry results were similar when the DC were isolated without using collagenase. Furthermore, the DC phenotype was similar on days 7, 14, 21, 42, and 75 after treatment with AdGM.



View larger version (36K):
[in this window]
[in a new window]
  		FIGURE 4. GM-CSF-recruited DC are slightly less mature than Flt3L-recruited DC. Freshly isolated splenic DC were analyzed by flow cytometry on day 7 after treatment with AdGM or controls. GM-CSF-recruited DC expressed slightly lower levels of MHC class I, CD40, CD54, and CD86 but higher CD80 than Flt3L-recruited DC. Isotype controls were set within the first of four decades. Data are representative of experiments repeated more than four times.

 
AdGM DC are located in the marginal zone

The predominantly CD8{alpha}- DC population recruited by endogenously produced GM-CSF was located in the marginal zone of the spleen consistent with previous reports of myeloid-related DC (Fig. 5Go, A and B) (4). On cytospin analysis, AdGM-recruited DC were slightly larger and had more pronounced dendritic processes than DC from untreated or AdGFP-treated mice (Fig. 5Go, C–E). The larger size of the GM-CSF-recruited DC was further evidenced on flow cytometry. In a representative experiment (of >10) in which 20,000 cells were counted per group, AdGM DC had a mean forward scatter of 360 compared with 250 for DC from untreated mice and 270 for DC from AdGFP-treated animals.



View larger version (121K):
[in this window]
[in a new window]
  		FIGURE 5. Endogenous GM-CSF recruits DC to the marginal zone of the spleen. A, The spleen (magnification, x4) from an AdGM-treated mouse was stained with biotin-conjugated anti-CD11c Ab, incubated with streptavidin-HRP, and counterstained with hematoxylin. B, At x40 magnification, the majority of DC were located in the marginal zone (M) directly adjacent to the follicles (F). DC were exceedingly rare in the PALS. Cytospins of DC from mice treated with injections of saline (C), AdGFP (D), or AdGM (E) revealed that GM-CSF-recruited DC were slightly larger and had more pronounced dendritic processes.

 
Endogenous GM-CSF recruited DC secrete high TNF-{alpha}

To determine whether there were other differences between GM-CSF-recruited DC and controls, we tested their cytokine secretory profiles at baseline and after stimulation. We cultured freshly isolated DC (106 cells/ml for 24 h) either alone or with LPS or TNF-{alpha}. We used ELISA to quantify supernatant cytokine levels. Strikingly, GM-CSF-recruited DC secreted 2000 pg/ml TNF-{alpha} by 24 h compared with only 80 pg/ml or lower for saline or AdGFP DC (Fig. 6GoB). In contrast, Flt3L-recruited DC did not secrete any detectable TNF-{alpha} even after stimulation by LPS or CD40 ligand (data not shown). The production of TNF-{alpha} by GM-CSF-recruited DC occurred regardless of whether they were cultured in GM-CSF. In addition, GM-CSF-recruited DC secreted up to 500% higher levels of IL-6 compared with controls when stimulated with either LPS (10 ng/ml) or TNF-{alpha} (100 ng/ml; Fig. 6GoB). In contrast to their elevated production of TNF-{alpha} and IL-6, DC from mice treated with AdGM produced dramatically lower IFN-{gamma} compared with DC from saline-treated mice. IL-10 expression was similar for all DC groups. However, DC secretion of IL-10 decreased after stimulation with LPS or TNF-{alpha} (Fig. 6GoC). IL-4, IL-12, or GM-CSF production was not detected from any of the DC groups even after stimulation.



View larger version (14K):
[in this window]
[in a new window]
  		FIGURE 6. DC recruited by endogenously produced GM-CSF secrete high levels of TNF-{alpha} and IL-6 but low IFN-{gamma}. Mice were treated with saline or 8 x 1010 particles of AdGFP or AdGM and then 7 days later their splenic DC were purified and cultured at a concentration of 1 x 106 cells/ml in complete medium only or with LPS (10 ng/ml) or TNF-{alpha} (100 ng/ml) for 24 h. Supernatants were then tested for TNF-{alpha} (A), IL-6 (B), IFN-{gamma} (C), and IL-10 (D). Averages of triplicates are shown. Data are representative of experiments repeated three separate times.

 
Recruited DC have enhanced Ag capture and T cell stimulatory capacity

We postulated that GM-CSF-recruited DC would have an enhanced ability to capture Ag since GM-CSF has been shown to increase Ag uptake in splenic DC and in peritoneal cells (12, 19, 34). DC macropinocytosis was assessed by uptake of both albumin and dextran while specialized endocytosis via the mannose receptor was tested by uptake of mannosylated albumin. In all cases, DC recruited by GM-CSF had both a faster rate of Ag capture (based on the percentage of fluorescent cells) and a higher total amount of Ag taken up per cell (represented by median fluorescence) (Fig. 7Go). For example, after 60 min of incubation with FITC-albumin, the median fluorescence for GM-CSF-recruited DC was 10 compared with 3–5 for DC from saline- and AdGFP-treated controls. However, splenic DC from mice treated with a similar injection of AdFlt3L had an even higher rate of uptake of albumin (Fig. 7GoB) and mannosylated albumin (Fig. 7GoC) than DC from AdGM-treated mice. To determine whether GM-CSF-recruited DC had increased Ag capture in vivo, mice were inoculated i.v. with 2.5 mg of FITC-dextran and then their splenic DC were harvested and analyzed for Ag uptake by flow cytometry. Again, GM-CSF-recruited DC had a far greater rate of Ag capture than DC from saline- or AdGFP-treated mice (Fig. 7GoD). However, in consort with our in vitro data, DC from mice treated with AdFlt3L demonstrated the highest uptake (data not shown).



View larger version (13K):
[in this window]
[in a new window]
  		FIGURE 7. DC recruited by endogenous GM-CSF have enhanced Ag uptake in vitro and in vivo. Mice were treated with saline or 8 x 1010 particles of AdGFP, AdGM, or AdFlt3L and then 7 days later their splenic DC were purified. DC were incubated with FITC-dextran for 15 min (A), FITC-albumin for up to 60 min (B), or FITC-mannose albumin for up to 60 min (C), and the percentage of cells that were fluorescent (A and C) or the median fluorescence (B) was determined by flow cytometry at various time points. D, To assess Ag uptake in vivo, FITC-dextran was injected i.v. into mice 7 days after treatment. Splenic DC were harvested 30 min later and tested for fluorescence by flow cytometry. AdGM DC had higher Ag uptake than saline or AdGFP controls in both in vitro and in vivo models. Averages of triplicates are shown.

 
Since GM-CSF-recruited DC produce increased levels of activating cytokines, we postulated that they would have an enhanced ability to stimulate effector cells. We first tested their T cell allostimulatory capacity in an MLR. GM-CSF-recruited DC induced considerably higher allogeneic T cell proliferation than DC from saline- or adenovirus-treated mice (Fig. 8GoA). Since TNF-{alpha} is a potent T cell activator, we postulated that the enhanced alloproliferation induced by GM-CSF-recruited DC was a result of their TNF-{alpha} secretion. In consort with this hypothesis, TNF-{alpha} blockade partially abrogated their increased T cell stimulation (Fig. 8GoB). We next assessed the ability of the DC to induce Ag-specific T cell stimulation. Consistent with our allogeneic data, freshly isolated GM-CSF-recruited DC loaded with either OVA257–264 peptide (Fig. 8GoC) or OVA (Fig. 8GoD) induced markedly higher stimulation of OVA-restricted T cells than DC from saline- or AdGFP-treated controls. GM-CSF-recruited DC also induced higher allogeneic and Ag-restricted T cell stimulation than Flt3L-recruited DC (data not shown).



View larger version (11K):
[in this window]
[in a new window]
  		FIGURE 8. GM-CSF-recruited DC induce higher stimulation of allogeneic and Ag-specific T cells. Mice were treated with saline or 8 x 1010 particles of AdGFP or AdGM and then 7 days later their splenic DC were purified. A, An MLR was performed by mixing irradiated DC with allogeneic (BALB/c) splenic T cells that had been purified with immunomagnetic beads. Triplicate wells were plated using 3 x 105 T cells/well. Syngeneic T cells had minimal proliferation (data not shown). B, Blockade of TNF-{alpha} partially abrogated the higher alloproliferation caused by GM-CSF-recruited DC. Ag-specific T cell proliferation was measured by incubating DC with either OVA peptide (C) or OVA (D) and then mixing them with 5 x 104 OVA-specific T cells and then measuring supernatant IL-2 48 h later. DC plated alone and T cells plated alone did not produce detectable IL-2. All T cell assays were done in triplicate and repeated more than three times with similar results.

 
Immunization with Ag-pulsed GM-CSF-recruited DC induces CTL

To determine whether the distinct DC population recruited by endogenously produced GM-CSF could stimulate cytotoxic T cells in vivo, we immunized naive mice with GM-CSF-recruited DC or controls that had been loaded in vitro with OVA257–264 peptide (DC.OVA). After two immunizations, we isolated splenocytes from immunized mice, restimulated them in vitro with OVA, and then tested them against EL4.OVA cells in a CTL assay. Splenocytes from mice immunized with DC.OVA, regardless of DC origin, induced similarly potent CTL-mediated lysis of EL4.OVA target cells (Fig. 9GoA). Lysis against EL4 targets was 75% lower, confirming the specificity of the assay (data not shown). However, in contrast to our MLR and Ag-restricted T cell stimulation data, immunization with Ag-pulsed GM-CSF-recruited DC did not induce higher CTL than Ag-pulsed DC from control mice. Nonetheless, splenic T cells from mice immunized with GM-CSF-recruited DC.OVA secreted nearly 50% higher levels of IFN-{gamma} than controls (Fig. 9GoB).



View larger version (16K):
[in this window]
[in a new window]
  		FIGURE 9. Immunization of mice with Ag-pulsed DC recruited by endogenous GM-CSF-induced CTL activity and elevated T cell IFN-{gamma} secretion. DC were harvested from mice 7 days after treatment with saline or 8 x 1010 particles of AdGFP or AdGM. Immediately after harvest, DC were loaded in vitro with OVA for 60 min and used to immunize naive mice. After two immunizations, splenocytes were harvested from immunized mice, restimulated in vitro with OVA, and tested in a CTL assay (A) and for cytokine production (B). Splenocytes from mice immunized with GM-CSF-recruited DC. OVA induced similar CTL lytic activity in controls but secreted higher levels of IFN-{gamma}. CTL and cytokine assays were repeated twice and done in triplicate wells.

 
GM-CSF-recruited DC activate NK cells via secretion of TNF-{alpha}

In addition to their primary role in stimulating T cells, DC have recently been shown to directly enhance NK cell lytic activity by cellular contact (32). Since TNF-{alpha} can have pleiotropic effects on a variety of effector cells, we postulated that GM-CSF-recruited DC may have an enhanced ability to activate NK cells. To test this, we cocultured GM-CSF DC and controls with freshly isolated NK cells and then harvested the NK cells and plated them against 51Cr-labeled Yac-1 target cells. NK cells that had been cultured with GM-CSF DC induced considerably higher lysis of targets compared with controls at all dilutions tested (Fig. 10Go). Moreover, consistent with our hypothesis, TNF-{alpha} blockade completely abrogated the effects.



View larger version (19K):
[in this window]
[in a new window]
  		FIGURE 10. GM-CSF-recruited DC increase NK cell lytic activity via secretion of TNF-{alpha}. Freshly isolated splenic DC were plated with NK cells from naive mice for 18 h in a 2:1 DC:NK ratio. NK cells were then harvested and plated against 51Cr-labeled Yac-1 target cells. NK cells that had been cultured with GM-CSF-recruited DC caused far higher lysis of targets than controls. However, the increased lysis was completely abrogated by TNF-{alpha} blockade in the coculture wells.

 
Endogenous secretion of GM-CSF does not confer tumor protection

Since endogenous secretion of GM-CSF recruits massive numbers of DC with enhanced immunostimulatory function, we postulated that treatment with AdGM would confer tumor protection in vivo. However, it failed to protect against tumor in a variety of models including s.c. CT26 colorectal cancer, CT26 liver metastases, B16 melanoma liver metastases, and s.c. EL4.OVA lymphoma (Table IIIGo). Neither administering AdGM simultaneous to tumor challenge nor extending the interval between viral inoculation and tumor challenge to 14 days changed the outcome. Similarly, and in consort with the CTL data, immunization of naive mice with OVA-pulsed GM-CSF-recruited DC failed to provide additional tumor protection against EL4.OVA tumor growth compared with immunization with peptide-pulsed DC from saline- or AdGFP-treated mice (Fig. 11Go).


View this table:
[in this window]
[in a new window]
  		Table III. Outcome after intrasplenic or s.c. challenge with CT26, B16, or EL4.OVA

 


View larger version (20K):
[in this window]
[in a new window]
  		FIGURE 11. Immunization with peptide-pulsed GM-CSF-recruited DC protects against tumor growth. C57BL/6 mice were treated with either saline or 8 x 1010 particles of AdGFP or AdGM. Their splenic DC were harvested 7 days later, loaded in vitro with OVA, and used to immunize naive mice. After two weekly immunizations, mice were challenged 7 days later with a s.c. flank injection of 3.5 x 105 EL4.OVA cells. Immunization with DC.OVA, regardless of DC origin, slowed tumor growth. Consistent with the CTL lysis data, but unlike the MLR and Ag-specific T cell stimulation data, GM-CSF-recruited DC did not confer additional protection over peptide-pulsed control DC. Averages of five mice per group are shown.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
We have shown the most dramatic DC recruitment to date. A single treatment with a gene transfer vector encoding GM-CSF increased the number of splenic DC by 200- to 260-fold. In contrast, daily injections of 10 µg of rFlt3L have been reported to increase DC populations by ~15- to 20-fold (35). Daily administration of GM-CSF conjugated to polyethylene glycol resulted in a nearly 8-fold increase in splenic DC (12). Minimal effects have been achieved with unmodified GM-CSF. Even ProGP-1, which is an agonist of both the Flt3 and GM-CSF receptors, only increased the total number of splenic DC to 5 x 107 cells (36). In contrast, we recruited up to 3 x 108 DC to the spleen. Furthermore, DC expansion after daily administration of any of the above proteins was only sustained for a few days after the cessation of treatments (36). In contrast, we have shown a greater than 15-fold increase in the number of splenic DC over normal even 75 days later. Considering that serum GM-CSF levels returned to baseline by 2 wk after AdGM injection (Fig. 1GoB), it is surprising that the number of splenic DC remained elevated for so long. It is uncertain whether a large number of cells are recruited shortly after treatment and persist in the spleen or whether there is continued recruitment of DC for an extended period. Using bromodeoxyuridine-labeling experiments, Kamath et al. (35) showed that all DC subtypes have a rapid turnover in the spleen with a half-life of just 2–3 days. This would suggest that there is continuous ongoing DC recruitment despite normalized serum GM-CSF levels. The precise derivation of the recruited DC is uncertain. The difficulty in tracking DC in vivo makes such an analysis difficult. However, our finding that bone marrow cellularity and the number of bone marrow-derived DC were decreased in AdGM-treated animals suggest that the recruited DC originated from the bone marrow.

Distinct DC subsets were recruited by endogenous overexpression of GM-CSF when compared with splenic DC subsets found in normal or Flt3L-treated mice. The vast majority of DC had classical myeloid staining (CD8{alpha}-CD11bbright). In contrast, Flt3L recruits a higher percentage of lymphoid DC (CD8{alpha}+CD11bdull/-). However, AdGM also expanded the lymphoid DC population, albeit to a much lesser extent than the myeloid population. This contrasts with a previous reports in which pegylated GM-CSF reportedly expanded only the CD11bbright population but not CD11bdull/- DC (12). Even more striking, virtually all of the GM-CSF-recruited DC were CD4-. This is distinctly different from Flt3L-recruited or normal DC and further contrasts with DC recruited by GM-CSF protein injections which reportedly do not have reduced CD4+ staining (36). Other distinctive phenotypic characteristics of AdGM-recruited DC included high expression of the Fc{gamma}III/II receptor CD16/CD32 but minimal B220 staining. The functional importance of each of these phenotypic differences remains unresolved.

Beside the phenotypic differences, there are numerous functional differences between DC recruited by endogenous overexpression of GM-CSF and previous reports using daily GM-CSF protein injections. DC recruited by endogenous GM-CSF overexpression had increased Ag capture in vitro and in vivo consistent with its effects reported previously on peritoneal macrophages (34). These findings are also compatible with those of Daro et al. (12) who showed that DC from mice treated with pegylated GM-CSF captured and processed Ag more efficiently than controls. However, unlike Daro et al. (12), we found that DC recruited by endogenous GM-CSF actually had lower Ag uptake than Flt3L-recruited DC. Furthermore, DC recruited by endogenous GM-CSF stimulated allogeneic and Ag-specific T cells to a greater extent than controls. Conversely, DC recruited by either recombinant or pegylated GM-CSF protein did not have enhanced immunostimulation (12, 36). The functional differences that we observed using DC recruited by endogenous GM-CSF secretion compared with protein injections may simply reflect the disparity in the phenotype and secretory profile of the DC subsets recruited.

Although the production of IL-12 and several other cytokines by DC has been investigated, there are few data regarding DC secretion of TNF-{alpha}. In mice, TNF-{alpha} matures bone marrow-derived DC. By virtue of its maturational effects, TNF-{alpha} renders bone marrow-derived DC more effective in generating antitumor immunity after adoptive transfer (37). However, the effects appeared to be due primarily to maturation and not TNF-{alpha} per se because CpG and CD40, potent stimulators of DC maturation, achieve even greater antitumor effects (37, 38). Although the maturational effects of TNF-{alpha} on DC are known, the importance of TNF-{alpha} production by DC has not been defined. In fact, there have been scarce reports of any constitutive TNF-{alpha} secretion by DC subsets. Lu et al. (39) reported that various murine liver DC populations secrete high levels of TNF-{alpha} only when stimulated by high-dose LPS (10 µg/ml) for 48 h. Human Langerhans cells can also be induced to produce TNF-{alpha} by virus or LPS (40, 41, 42, 43, 44, 45). Human monocyte-derived DC produce TNF-{alpha} upon exposure to LPS or Escherichia coli (46, 47). In our system, normal splenic DC only produced low levels of TNF-{alpha} while Flt3L DC did not make detectable quantities. This is consistent with the report by Pulendran et al. (48), who demonstrated that CD8{alpha}+ splenic DC do not secrete TNF-{alpha} while CD8{alpha}- DC subsets produce very low levels (<100 pg/ml by 24 h).

This appears to be the first report of unstimulated DC producing markedly elevated quantities of TNF-{alpha}. The regulation of TNF-{alpha} secretion by GM-CSF-recruited DC is uncertain. The consequences of TNF-{alpha} production by DC on their capacity to stimulate effector cells have also not been delineated. We showed that the enhanced allostimulatory capacity of GM-CSF-recruited DC is partially mediated by their high TNF-{alpha} secretion. TNF-{alpha} also accounted for the increased ability of GM-CSF-recruited DC to stimulate NK cells. Thus, TNF-{alpha} may enable GM-CSF-recruited DC to link innate and adaptive immunity. The minimal production of IFN-{gamma} by GM-CSF-recruited DC compared with controls may also be a relevant factor in their enhanced immunostimulation. DC production of IFN-{gamma} has been shown to mediate tolerogenic effects on T cells in certain DC subsets (49).

In addition to TNF-{alpha}, GM-CSF-recruited DC expressed high levels of IL-6 compared with controls. Both the regulation of DC secretion of IL-6 as well as its effect on DC immune function are not completely understood. Chomarat et al. (50) reported that IL-6 is a critical factor in the molecular control of APC development and that IL-6 release from fibroblasts switches the differentiation of monocytes from DC to macrophages. Nevertheless, the relevance of IL-6 produced by the DC or monocytes themselves in this process in not defined. Grohmann et al. (49) reported that IL-6 plays a critical role in mediating the effects of CD40 ligation in CD8{alpha}+ DC and enhancing their immunogenicity. The role of IL-6 included down-regulation of the IFN-{gamma}R expression on CD8{alpha}+ DC and correlated with the reduced ability of these DC to initiate T cell apoptosis in vitro. However, considering that the DC recruited by GM-CSF were largely CD8{alpha}-, the potential role of IL-6 in mediating their enhanced immunostimulatory function requires further study.

Considering that DC recruited by endogenous GM-CSF have enhanced T cell and NK cell stimulatory properties, it was somewhat surprising that recruitment of large numbers of these DC did not confer tumor protection in any of the models tested. However, it is possible that GM-CSF-recruited DC only acquire enhanced immunostimulatory function after in vitro culture where they became more mature and express higher levels of costimulatory and accessory molecules (51, 52). In contrast, these DC may be relatively quiescent in situ. Similarly, it is uncertain whether recruited DC actually produce activating cytokines such as TNF-{alpha} in vivo or only after isolation. The failure of GM-CSF-recruited DC to stimulate tumor protection raises considerable interest in the potential for substances that can directly activate DC in vivo to harness the immune potential of this massive DC recruitment. Merad et al. (53) recently reported that systemic administration of immunostimulatory DNA activates DC in vivo and augments the antitumor effects of Flt3L. Application of other in vivo DC activators to our model holds considerable promise for the immunotherapy of disease.


   Footnotes
 
1 This work was supported in part by National Institutes of Health Grant CA94503. Back

2 Address correspondence and reprint requests to Dr. Ronald P. DeMatteo, Memorial Sloan-Kettering Cancer Center, Box 203, 1275 York Avenue, New York, NY 10021. E-mail address: dematter{at}mskcc.org Back

3 Abbreviations used in this paper: DC, dendritic cell; PALS, periarteriolar lymphatic sheath; AdGM, adenovirus encoding the GM-CSF transgene; Flt3L, Flt3 ligand; AdFlt3L, Ad encoding the Flt3L transgene; AdGFP, Ad encoding green fluorescent protein. Back

Received for publication April 19, 2002. Accepted for publication July 9, 2002.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Kelsall, B. L., W. Strober. 1996. Distinct populations of dendritic cells are present in the subepithelial dome and T cell regions of the murine Peyer’s patch. J. Exp. Med. 183:237.[Abstract/Free Full Text]
  2. Anjuere, F., P. Martin, I. Ferrero, M. L. Fraga, G. M. del Hoyo, N. Wright, C. Ardavin. 1999. Definition of dendritic cell subpopulations present in the spleen, Peyer’s patches, lymph nodes, and skin of the mouse. Blood 93:590.[Abstract/Free Full Text]
  3. Maraskovsky, E., K. Brasel, M. Teepe, E. R. Roux, S. D. Lyman, K. Shortman, H. J. McKenna. 1996. Dramatic increase in the numbers of functionally mature dendritic cells in Flt3 ligand-treated mice: multiple dendritic cell subpopulations identified. J. Exp. Med. 184:1953.[Abstract/Free Full Text]
  4. Pulendran, B., J. Lingappa, M. K. Kennedy, J. Smith, M. Teepe, A. Rudensky, C. R. Maliszewski, E. Maraskovsky. 1997. Developmental pathways of dendritic cells in vivo: distinct function, phenotype, and localization of dendritic cell subsets in FLT3 ligand-treated mice. J. Immunol. 159:2222.[Abstract/Free Full Text]
  5. Vremec, D., G. J. Lieschke, A. R. Dunn, L. Robb, D. Metcalf, K. Shortman. 1997. The influence of granulocyte/macrophage colony-stimulating factor on dendritic cell levels in mouse lymphoid organs. Eur. J. Immunol. 27:40.[Medline]
  6. Leenen, P. J., K. Radosevic, J. S. Voerman, B. Salomon, N. van Rooijen, D. Klatzmann, W. van Ewijk. 1998. Heterogeneity of mouse spleen dendritic cells: in vivo phagocytic activity, expression of macrophage markers, and subpopulation turnover. J. Immunol. 160:2166.[Abstract/Free Full Text]
  7. Ohteki, T., T. Fukao, K. Suzue, C. Maki, M. Ito, M. Nakamura, S. Koyasu. 1999. Interleukin 12-dependent interferon {gamma} production by CD8{alpha}+ lymphoid dendritic cells. J. Exp. Med. 189:1981.[Abstract/Free Full Text]
  8. Maldonado-Lopez, R., T. De Smedt, P. Michel, J. Godfroid, B. Pajak, C. Heirman, K. Thielemans, O. Leo, J. Urbain, M. Moser. 1999. CD8{alpha}+ and CD8a. J. Exp. Med. 189:587.[Abstract/Free Full Text]
  9. Sousa, C., G. Yap, O. Schulz, N. Rogers, M. Schito, J. Aliberti, S. Hieny, A. Sher. 1999. Paralysis of dendritic cell IL-12 production by microbial products prevents infection-induced immunopathology. Immunity 11:637.[Medline]
  10. De Smedt, T., E. Butz, J. Smith, R. Maldonado-Lopez, B. Pajak, M. Moser, C. Maliszewski. 2001. CD8{alpha}- and CD8{alpha}+ subclasses of dendritic cells undergo phenotypic and functional maturation in vitro and in vivo. J. Leukocyte Biol. 69:951.[Abstract/Free Full Text]
  11. Huang, L. Y., C. Sousa, Y. Itoh, J. Inman, D. E. Scott. 2001. IL-12 induction by a Th1-inducing adjuvant in vivo: dendritic cell subsets and regulation by IL-10. J. Immunol. 167:1423.[Abstract/Free Full Text]
  12. Daro, E., B. Pulendran, K. Brasel, M. Teepe, D. Pettit, D. H. Lynch, D. Vremec, L. Robb, K. Shortman, H. J. McKenna, et al 2000. Polyethylene glycol-modified GM-CSF expands CD11bhighCD11chigh but not CD11blowCD11chigh murine dendritic cells in vivo: a comparative analysis with Flt3 ligand. J. Immunol. 165:49.[Abstract/Free Full Text]
  13. Suss, G., K. Shortman. 1996. A subclass of dendritic cells kills CD4 T cells via Fas/Fas-ligand-induced apoptosis. J. Exp. Med. 183:1789.[Abstract/Free Full Text]
  14. Kronin, V., K. Winkel, G. Suss, A. Kelso, W. Heath, J. Kirberg, H. von Boehmer, K. Shortman. 1996. A subclass of dendritic cells regulates the response of naive CD8 T cells by limiting their IL-2 production. J. Immunol. 157:3819.[Abstract]
  15. Steinman, R. M., M. Pack, K. Inaba. 1997. Dendritic cells in the T-cell areas of lymphoid organs. Immunol. Rev. 156:25.[Medline]
  16. Traver, D., K. Akashi, M. Manz, M. Merad, T. Miyamoto, E. G. Engleman, I. L. Weissman. 2000. Development of CD8{alpha}-positive dendritic cells from a common myeloid progenitor. Science 290:2152.[Abstract/Free Full Text]
  17. Martinez, D. H., P. Martin, C. F. Arias, A. R. Marin, C. Ardavin. 2002. CD8{alpha}+ dendritic cells originate from the CD8a- dendritic cell subset by a maturation process involving CD8a, DEC-205, and CD24 up-regulation. Blood 99:999.[Abstract/Free Full Text]
  18. Pulendran, B., J. L. Smith, G. Caspary, K. Brasel, D. Pettit, E. Maraskovsky, C. R. Maliszewski. 1999. Distinct dendritic cell subsets differentially regulate the class of immune response in vivo. Proc. Natl. Acad. Sci. USA 96:1036.[Abstract/Free Full Text]
  19. Sallusto, F., A. Lanzavecchia. 1994. Efficient presentation of soluble antigen by cultured human dendritic cells is maintained by granulocyte/macrophage colony-stimulating factor plus interleukin 4 and downregulated by tumor necrosis factor {alpha}. J. Exp. Med. 179:1109.[Abstract/Free Full Text]
  20. Larsen, C. P., S. C. Ritchie, R. Hendrix, P. S. Linsley, K. S. Hathcock, R. J. Hodes, R. P. Lowry, T. C. Pearson. 1994. Regulation of immunostimulatory function and costimulatory molecule (B7-1 and B7-2) expression on murine dendritic cells. J. Immunol. 152:5208.[Abstract]
  21. Metcalf, D., K. Shortman, D. Vremec, S. Mifsud, L. Di Rago. 1996. Effects of excess GM-CSF levels on hematopoiesis and leukemia development in GM-CSF/max 41 double transgenic mice. Leukemia 10:713.[Medline]
  22. Maraskovsky, E., B. Pulendran, K. Brasel, M. Teepe, E. R. Roux, K. Shortman, S. D. Lyman, H. J. McKenna. 1997. Dramatic numerical increase of functionally mature dendritic cells in FLT3 ligand-treated mice. Adv. Exp. Med. Biol. 417:33.[Medline]
  23. Bronte, V., D. B. Chappell, E. Apolloni, A. Cabrelle, M. Wang, P. Hwu, N. P. Restifo. 1999. Unopposed production of granulocyte-macrophage colony-stimulating factor by tumors inhibits CD8+ T cell responses by dysregulating antigen-presenting cell maturation. J. Immunol. 162:5728.[Abstract/Free Full Text]
  24. Hanada, K., R. Tsunoda, H. Hamada. 1996. GM-CSF-induced in vivo expansion of splenic dendritic cells and their strong costimulation activity. J. Leukocyte Biol. 60:181.[Abstract]
  25. Kielian, T., E. Nagai, A. Ikubo, C. A. Rasmussen, T. Suzuki. 1999. Granulocyte/macrophage-colony-stimulating factor released by adenovirally transduced CT26 cells leads to the local expression of macrophage inflammatory protein 1{alpha} and accumulation of dendritic cells at vaccination sites in vivo. Cancer Immunol. Immunother. 48:123.[Medline]
  26. DeMatteo, R. P., G. Chu, M. Ahn, E. Chang, C. F. Barker, J. F. Markmann. 1997. Long-lasting adenovirus transgene expression in mice through neonatal intrathymic tolerance induction without the use of immunosuppression. J. Virol. 71:5330.[Abstract]
  27. Wang, J., D. P. Snider, B. R. Hewlett, N. W. Lukacs, J. Gauldie, H. Liang, Z. Xing. 2000. Transgenic expression of granulocyte-macrophage colony-stimulating factor induces the differentiation and activation of a novel dendritic cell population in the lung. Blood 95:2337.[Abstract/Free Full Text]
  28. Sallusto, F., M. Cella, C. Danieli, A. Lanzavecchia. 1995. Dendritic cells use macropinocytosis and the mannose receptor to concentrate macromolecules in the major histocompatibility complex class II compartment: downregulation by cytokines and bacterial products. J. Exp. Med. 182:389.[Abstract/Free Full Text]
  29. Lattime, E. C., O. Stutman. 1991. Thymic lymphomas mediate non-MHC-restricted, TNF-dependent lysis of the murine sarcoma WEHI-164. Cell. Immunol. 136:69.[Medline]
  30. Carbone, F. R., S. J. Sterry, J. Butler, S. Rodda, M. W. Moore. 1992. T cell receptor alpha-chain pairing determines the specificity of residue 262 within the Kb-restricted, ovalbumin257–264 determinant. Int. Immunol. 4:861.[Abstract/Free Full Text]
  31. DeMatteo, R. P., J. F. Markmann, K. F. Kozarsky, C. F. Barker, S. E. Raper. 1996. Prolongation of adenoviral transgene expression in mouse liver by T lymphocyte subset depletion. Gene Ther. 3:4.[Medline]
  32. Fernandez, N. C., A. Lozier, C. Flament, P. Ricciardi-Castagnoli, D. Bellet, M. Suter, M. Perricaudet, T. Tursz, E. Maraskovsky, L. Zitvogel. 1999. Dendritic cells directly trigger NK cell functions: cross-talk relevant in innate anti-tumor immune responses in vivo. Nat. Med. 5:405.[Medline]
  33. Kurlander, R. J., D. M. Ellison, J. Hall. 1984. The blockade of Fc receptor-mediated clearance of immune complexes in vivo by a monoclonal antibody (2.4G2) directed against Fc receptors on murine leukocytes. J. Immunol. 133:855.[Abstract]
  34. Burger, J. A., S. M. Baird, H. C. Powell, S. Sharma, D. J. Eling, T. J. Kipps. 2000. Local and systemic effects after adenoviral transfer of the murine granulocyte-macrophage colony-stimulating factor gene into mice. Br. J. Haematol. 108:641.[Medline]
  35. Kamath, A. T., J. Pooley, M. A. O’Keeffe, D. Vremec, Y. Zhan, A. M. Lew, A. D’Amico, L. Wu, D. F. Tough, K. Shortman. 2000. The development, maturation, and turnover rate of mouse spleen dendritic cell populations. J. Immunol. 165:6762.[Abstract/Free Full Text]
  36. O’Keeffe, M., H. Hochrein, D. Vremec, J. Pooley, R. Evans, S. Woulfe, K. Shortman. 2002. Effects of administration of progenipoietin 1, Flt-3 ligand, granulocyte colony-stimulating factor, and pegylated granulocyte- macrophage colony-stimulating factor on dendritic cell subsets in mice. Blood 99:2122.[Abstract/Free Full Text]
  37. Brunner, C., J. Seiderer, A. Schlamp, M. Bidlingmaier, A. Eigler, W. Haimerl, H. A. Lehr, A. M. Krieg, G. Hartmann, S. Endres. 2000. Enhanced dendritic cell maturation by TNF-{alpha} or cytidine-phosphate- guanosine DNA drives T cell activation in vitro and therapeutic antitumor immune responses in vivo. J. Immunol. 165:6278.[Abstract/Free Full Text]
  38. Labeur, M. S., B. Roters, B. Pers, A. Mehling, T. A. Luger, T. Schwarz, S. Grabbe. 1999. Generation of tumor immunity by bone marrow-derived dendritic cells correlates with dendritic cell maturation stage. J. Immunol. 162:168.[Abstract/Free Full Text]
  39. Lu, L., C. A. Bonham, X. Liang, Z. Chen, W. Li, L. Wang, S. C. Watkins, M. A. Nalesnik, M. S. Schlissel, A. J. Demestris, J. J. Fung, S. Qian. 2001. Liver-derived DEC205+B220+CD19- dendritic cells regulate T cell responses. J. Immunol. 166:7042.[Abstract/Free Full Text]
  40. Nickoloff, B. J., G. D. Karabin, J. N. Barker, C. E. Griffiths, V. Sarma, R. S. Mitra, J. T. Elder, S. L. Kunkel, V. M. Dixit. 1991. Cellular localization of interleukin-8 and its inducer, tumor necrosis factor-{alpha} in psoriasis. Am. J. Pathol. 138:129.[Abstract]
  41. Enk, A. H., S. I. Katz. 1992. Early molecular events in the induction phase of contact sensitivity. Proc. Natl. Acad. Sci. USA 89:1398.[Abstract/Free Full Text]
  42. Schreiber, S., O. Kilgus, E. Payer, R. Kutil, A. Elbe, C. Mueller, G. Stingl. 1992. Cytokine pattern of Langerhans cells isolated from murine epidermal cell cultures. J. Immunol. 149:3524.
  43. Oxholm, A., M. Diamant, P. Oxholm, K. Bendtzen. 1991. Interleukin-6 and tumour necrosis factor {alpha} are expressed by keratinocytes but not by Langerhans cells. APMIS 99:58.[Medline]
  44. Larrick, J. W., V. Morhenn, Y. L. Chiang, T. Shi. 1989. Activated Langerhans cells release tumor necrosis factor. J. Leukocyte Biol. 45:429.[Abstract]
  45. Sprecher, E., Y. Becker. 1992. Detection of IL-1{beta}, TNF-{alpha}, and IL-6 gene transcription by the polymerase chain reaction in keratinocytes, Langerhans cells and peritoneal exudate cells during infection with herpes simplex virus-1. Arch. Virol. 126:253.[Medline]
  46. Verhasselt, V., C. Buelens, F. Willems, D. De Groote, N. Haeffner-Cavaillon, M. Goldman. 1997. Bacterial lipopolysaccharide stimulates the production of cytokines and the expression of costimulatory molecules by human peripheral blood dendritic cells: evidence for a soluble CD14-dependent pathway. J. Immunol. 158:2919.[Abstract]
  47. Huang, Q., D. Liu, P. Majewski, L. C. Schulte, J. M. Korn, R. A. Young, E. S. Lander, N. Hacohen. 2001. The plasticity of dendritic cell responses to pathogens and their components. Science 294:870.[Abstract/Free Full Text]
  48. Pulendran, B., P. Kumar, C. W. Cutler, M. Mohamadzadeh, T. Van Dyke, J. Banchereau. 2001. Lipopolysaccharides from distinct pathogens induce different classes of immune responses in vivo. J. Immunol. 167:5067.