|Home | About | Journals | Submit | Contact Us | Français|
Dendritic cells present exogenous proteins to MHC class I restricted CD8+ T cells. This function does not require endogenous antigen synthesis within DC, providing the potential to elicit CD8+ T cell responses to immune complexes, inactivated microbes, dying cells and proteins like ovalbumin. In mice, the CD8+ or DEC-205+ DC are specialized for cross-presentation, and this subset can be increased 10 fold in numbers following Flt3L treatment in vivo. Therefore we studied cross-presentation by abundant Flt3L DC using HIV gag protein. When enriched by positive selection with anti-CD11c beads, cells from Flt3L mice are not only more abundant but are more highly enriched in CD11c high DC, particularly the DEC-205+ subset. DC cross-present HIV gag to primed CD8+ T cells, but when the antigen is delivered within an antibody to DEC-205 receptor, cross-presentation becomes 100 fold more efficient than non-targeted antigen. This finding requires gag to be engineered into anti-DEC antibody, not just mixed with antibody. Flt3L DC are a valuable tool to study cross-presentation, since their use overcomes the obstacle posed by the low number of cross-presenting DC in the steady state. These findings support future experiments to use Flt3L to enhance presentation of DC-targeted vaccines.
Induction of strong CD8+ T cell responses is a major goal in the development of preventive and therapeutic vaccines against persistent viruses and tumors. Dendritic cells (DC) can initiate CD8+ T cell responses through either direct priming or cross-priming. Direct priming refers to the generation of peptide-MHC class I complexes from endogenously synthesized proteins, while cross-priming involves processing of exogenous proteins acquired from the extracellular environment. Cross-priming has been shown to be important in initiating MHC class I- restricted responses to tumors, peripheral self, viral and bacterial antigens .
The efficiency of cross-presentation can be enhanced when dendritic cells (DC) take up antigens through different receptor-mediated pathways particularly antigen-antibody complexes [2-4], dying cells [5, 6], and proteins targeted within antibodies to specific receptors like DEC-205 , DC-SIGN , MMR ; Langerin , LOX1 , and CLEC9A . In our research, we have been studying cross-presentation by genetically engineering the sequences for protein antigens into the heavy chain of monoclonal antibodies (mAb) to DC receptors, particularly DEC-205 .
Much of the research in this field has used ovalbumin (OVA) as a tool and the CD8+ OT-I TCR transgenic line specific for OVA presented on H-2Kb molecules. This is because the SIINFEKL peptide from OVA is presented efficiently, with only picomolar levels of peptide being active when DC are the antigen presenting cells . Nevertheless, DC can cross-present other proteins like malaria circumsporozoite protein [13-16], HIV gag [17, 18], HSV and influenza proteins [19-21], and tumor antigens [12, 22-24], although this cross-presentation is often studied with TCR transgenic T cells. Vaccine science requires that the cross-presentation pathway be extended to candidate vaccine antigens and non-transgenic T cells.
In mouse lymphoid tissues, a subset of DC, marked by expression of CD8α homodimer and the DEC-205/CD205 endocytic receptor, is more active in cross-presentation [4, 6, 25]. Here we have studied these cross-presenting DC using mice exposed to Flt3L, which is known to expand several different subsets of DC including a relatively high proportion of CD8+ DC, increasing total numbers of this subset ~ 30 fold [26, 27].
With the most commonly studied protein, OVA, concentrations higher than 50 μg/ml of protein are required to detect cross-presentation [9, 13, 25, 28-34]. Here we test whether cross-presentation of an exogenous HIV protein antigen, the gag p24 protein, to primed CD8+ T cells can be enhanced by receptor mediated uptake. We will show that Flt3L DC efficiently cross-present HIV gag via the DEC-205 receptor, about 100 times more effective than non-targeted gag. 0.1 μg/ml of anti-DEC gag is as effective as 25 μg/ml of gag, suggesting that this strategy for increasing DC numbers be used to identify cross-presentation mechanisms and improve protein vaccines.
We used anti-CD11c coated magnetic beads to enrich splenic CD11c+ DC from CxB6F1 mice injected with B16 Flt3L secreting melanoma cells. As previously reported, Flt3L greatly expanded total DC numbers ~10 fold [26, 35, 36]. In our experiments 4-5 ×106 cells were isolated with anti-CD11c beads from one spleen of an untreated mouse, whereas ~ 60-70 ×106 cells could be enriched from the spleen of a mouse harboring a growing B16-Flt3L melanoma. When the DC from untreated and Flt3L mice were compared by flow cytometry in surface antigen expression (Suppl. Fig. 1 for gating conditions), the populations from Flt3L mice were much more enriched in CD11c high cells, ~95% vs. ~25% in untreated mice (Fig. 1). The CD11c-selected population from Flt3L treated mice was PDCA-1low, B220low but CD11chigh, indicating a paucity of plasmacytoid DC (Fig. 1). The majority of the cells expressed MHC class II molecules, while the representation of the CD8α and DEC-205 subset was increased relative to untreated mice. However in three different experiments we observed that this increase was biased toward DEC-205+ DC over CD8+ cells, possibly because of the presence of a recently described CD8α- CD24hi population of precursors to CD8α+ DC .
These results extend prior work on Flt3L  showing that the expansion induced by this hematopoietin allows for an increase in the purity of CD11c selected cells that are also enriched in DEC-205+ DC.
Since the DC generated from Flt3L treated mice expressed intermediate levels of I-Ab molecules (Fig. 1), we expected that the cells would be functionally immature. A distinguishing feature of mature DC is their capacity to stimulate naïve T cells in an allogenic mixed leukocyte reaction (MLR). Thus we tested whether DC isolated from Flt3L-treated mice exhibited allo-stimulatory activity, comparing it to untreated spleen DC (Fig. 2A). We evaluated the DC with or without direct in vivo activation with poly IC, which is known to mature DC . Accordingly, Flt3L or untreated CxB6 F1 mice were injected with 50 μg of poly IC, and 15 h later, DC were enriched with anti-CD11c beads, fixed with para-formaldehyde to prevent further maturation in culture, washed, and added in graded doses to allogeneic T cells isolated from C57BL/6 mice, which had been labeled with CFSE. In Fig. 2A, we show that both DC populations, upon in vivo maturation, stimulated CD4+ and CD8+ T cells to proliferate. We also found that the Flt3L-mobilized DC had a stronger stimulating capacity being about 5 times more active than normal DC. This result can be explained by the enrichment in CD11chigh cells in DC preparations from Flt3L treated mice. Populations from Flt3L mice were almost entirely CD11chigh and MHC IIhigh DC, whereas populations selected with anti-CD11 beads from normal spleen were roughly 30% CD11chigh and MHC IIhigh DC and clearly had contaminating DX5+ and CD19+ NK and B cells (Fig. 1).
To deliver HIV antigens to DC, Trumpfheller et al. cloned HIV gag p24 protein within the heavy chain of a mAb specific for DEC-205 endocytic receptor . We tested the ability of the Flt3L DC to mediate presentation to HIV specific CD8+ and CD4+ T cells after in vitro pulsing of DC with anti-DEC-205 HIV gag p24 fusion mAb (anti-DEC-p24) or a control Ig-p24 fusion antibody. Specifically, CD11c+ populations were isolated from Flt3L treated and non treated mice. Then the DC were cultured overnight with poly IC in the presence of the indicated sources of gag antigen, washed to remove excess antigen and added to HIV gag specific T cells for 6 h. We found (Fig. 2B) that both DC populations stimulated IFN-γ production from HIV-gag primed CD8+ and CD4+ T cells isolated from mice primed with Adenovirus-gag p24 and boosted with anti-DEC-p24 and poly IC. Adenovirus-p24 primarily primes CD8+ T cells, while anti-DEC-p24 along with poly IC induces CD4+ T cells as described by Trumpfheller et al. . The response to DC pulsed with anti-DEC-p24 was similar to a pool of pre-processed HIV gag 15 mer peptides, and anti-DEC-p24 antibody resulted in stronger responses than control Ig-p24 (Fig. 2B). There was a small difference between the DC from Flt3L treated mice and those from normal mice in their ability to stimulate antigen-primed T cells. Thus, Flt3L treatment induced high numbers of CD11c+ DC that were functionally comparable to normal DC in their capacity to process and present in vitro anti-DEC-p24 mAb.
Altogether these results demonstrate that CD11c selected DC isolated from Flt3L-mobilized mice were functional in inducing MLR and in exogenous MHC class I and class II antigen-presentation pathways.
Effective T cell stimulation only occurs if in parallel to antigen uptake, DC undergo maturation, a process that can be triggered by different pathogens or mimics of microbial agonists, such as poly IC for double stranded RNA or lipopolysaccaride respectively. In vitro studies have indicated that DC maturation leads to enhanced cross-presentation as well as expression of the costimulatory molecules required for activation of CD8+ T cells [30, 39-42]. To confirm the capacity of maturation to promote cross-presentation, we enriched splenic CD11c+ DC from mice treated with B16 Flt3L melanoma cells, added different sources of HIV gag protein or peptides for 5 h, and then cultured the cells overnight without or with 0.1 μg/ml of LPS or 25 μg/ml of poly IC as maturation stimuli. After extensive washing, the cells were added to HIV gag specific T cells in the presence of BFA, and cross-presentation was assessed 6 h later by intracellular cytokine staining for IFN-γ secretion by CD8+ T cells.
In three different experiments (Fig. 3A), LPS or poly IC treatment improved cross-presentation relative to PBS treated DC, although the latter could undergo spontaneous maturation in culture as previously reported [43-46]. Although these results cannot be used to assess if immature Flt3L-mobilized DC have any cross-presenting function for stimulation of IFN-γ production from T cells, the results indicate that anti-DEC-p24 mAb is efficiently cross-presented by more mature Flt3L DC, and all our subsequent studies used poly IC to stimulate the antigen-pulsed Flt3L DC.
Splenic CD8α+ DEC-205+ DC are specialized to cross-present cell-associated and protein antigens [4, 19, 21, 25, 47]. In Fig. 1 we have shown that CD8α+ DEC-205+ cell numbers expanded considerably after Flt3L treatment with a profound increase skewed toward the DEC-205+ subset (Fig. 1). To verify that DEC-205+ DC were responsible for DEC-205 mediated cross-presentation of HIV gag, CD11c+ DEC-205+ and CD11c+ DEC-205- cells were purified and sorted by FACS from the spleens of mice inoculated with B16 Flt3L melanoma cells (Methods and Suppl. Fig. 1). Sorted cells were pulsed with anti-DEC-p24, then poly IC was added. After 15 h, graded doses of antigen-pulsed and matured DEC-205+ and DEC-205- cells were added to HIV gag primed T cells, and IFN-γ secretion was assessed 6 h later.
Both DC subsets were able to stimulate CD8+ T cells to secrete IFN-γ after in vitro incubation with a pool of HIV gag p24 15 mer peptides (Fig. 3B). However, when we analyzed cross-presentation of gag protein within anti-DEC-p24 mAb, only the DEC-205+ DC efficiently cross-presented HIV gag to primed CD8+ T cells over a wide range of DC to T cell doses, 1:10-1:90 (Fig. 3B). In contrast, there was no presentation of control Ig-p24, indicating that the DEC+ DC subset selectively cross-presents DEC-targeted protein.
With these results, we proceeded to carry out quantitative assays to assess the efficiency of HIV gag p24 cross-presentation by Flt3L-mobilized DC.
To better evaluate the efficiency of cross-presentation by Flt3L-mobilized DC, we carried out more detailed studies of antigen and DC dose. When we pulsed Flt3L CD11c+ DC with different doses of anti-DEC-p24 vs control Ig-p24 overnight, along with maturation by poly IC, escalating doses of anti-DEC-p24 resulted in an increased percentage of IFN-γ+ CD8+ T cells, reaching a plateau at only 1 μg/ml of the fusion antibody (Fig. 4A). When used to stimulate HIV gag primed T cells at a DC:T cell ratio of 1:3, the Flt3L DC pulsed with 0.1 μg/ml of anti-DEC-p24 induced higher levels of IFN-γ secretion by CD8+ T cells than DC pulsed with 1 μg/ml control Ig-p24 (Fig. 4A). In Fig. 4B, we displayed the same results but from 4 different experiments.
Next, we assessed the role of DC dose. Flt3L DC were pulsed with a fixed dose of control Ig-p24 vs increasing concentrations of anti-DEC-p24 or gag peptides. After maturation with poly IC, DC were added in graded numbers to HIV gag primed/boosted T cells. In vitro cross-presentation of anti-DEC-p24 by Flt3L DC occurred in a DC-dose dependent manner. Flt3L DC pulsed with anti-DEC-p24 stimulated IFN-γ secretion of CD8+ T cells over a range of DC to T cell doses from 1:3 up to 1:90 (Fig. 4C, left panel). A similar DC dose dependence was obtained when we analyzed stimulation of CD4+ T cells after anti-DEC-p24 in vitro pulsing (Fig. 4C, right panel).
Together the results indicate that a low concentration of anti-DEC-p24 antibody mediates presentation of gag protein to both HIV specific CD8+ and CD4+ T cells in a DC-dose dependent fashion.
Previous studies showed that only relatively high doses of antigens e.g., 50-100 μg/ml OVA, are able to bring about cross-presentation through the exogenous MHC-I pathway. We therefore wondered if DEC targeting could increase presentation of protein that was not coupled to the anti-DEC antibody. We first compared cross-presentation of anti-DEC-p24 mAb with soluble HIV gag p24 protein. In Fig. 5A we found that cross-presentation of soluble HIV gag protein to specific CD8+ T cells by Flt3L DC took place in a dose dependent manner, and that 25 μg/ml of gag was comparable to 1 μg/ml of anti-DEC-gag (or 0.25 μg/ml gag protein within the antibody). These results demonstrate that cross-presentation of anti-DEC-p24 by Flt3L DC improves efficiency almost 100 times relative to non-targeted gag protein.
Then to rule out a boosting effect of the anti-DEC antibody itself, we analyzed CD8+ T cell responses to Flt3L DC pulsed in vitro with anti-DEC-p24 fusion mAb, unconjugated anti-DEC-205 mAb, the combination of unconjugated anti-DEC-205 with HIV gag protein, or the soluble HIV protein alone. As shown in Fig. 5B, the combination of unconjugated mAb with soluble HIV protein did not enhance gag presentation.
We conclude that introduction of a protein within anti-DEC-205 antibody greatly enhances the capacity of Flt3L-mobilized DC to cross-present a protective microbial antigen, HIV gag.
The limited number of DC isolated from mouse spleen has often restricted their use for functional studies, particularly cross-presentation, where the relevant CD8α+ or DEC-205+ DC [4, 19-21, 47] represent a relatively small fraction of CD11c high DC. Flt3L is a regulator of hematopoietic cell development and increases the number of peripheral DC in various tissues of mice [26, 36]. It is now evident that the receptor, Flt-3 or Flt-2 or CD135, is a marker for committed progenitors of DC that form in the bone marrow and then continue to respond to Flt3L after migration via the blood into spleen and lymph nodes [48-53]. Interestingly the expansion in DC numbers was skewed towards the expansion of DEC-205+ cells over CD8+ DC. The CD8- DEC-205+ DC may represent precursors to CD8+ DEC-205+ DC . One emphasis of our current study is that when mice are exposed to Flt3L, it becomes much more feasible to study cross-presentation, since the numbers and purity of cross-presenting DEC-205+ DC are greatly expanded.
The majority of studies of cross-presentation emphasize the clonal expansion of CD8+ TCR transgenic T cells. Polyclonal T cells from HIV gag primed mice also respond to cross-presenting DC, although the efficiency of presentation was increased by targeting the gag protein to the DEC-205 receptor. Previously we found that the targeting of HIV gag protein to the DEC-205 receptor on monocyte-derived human DC enhanced cross-presentation to gag-specific CD8+ T cells from individuals infected with HIV-1 . In that study, we also compared different receptors, DEC-205, MMR and DC-SIGN, and obtained evidence that antigen delivery via DEC-205 was superior to the other endocytic receptors for expanding gag-specific CD8+ T cells . Here our current studies were directed to DEC-205 targeting, where DEC-205 represents one receptor that is clearly expressed in vivo on most T cell area DC of human lymph nodes . Other DC receptors and surface products might be expected to mediate improved presentation with antibody-targeted antigens. Here we provide quantitative information on the capacity of DC to present antigen via DEC-205, which also takes place in DC preparations that are abundant and highly enriched. Flt3L induced expansion of DC should greatly facilitate studies of cross-presentation and set the stage for the use of Flt3L to enhance the efficacy of DEC-205 targeted vaccines in vivo.
Immune responses require DC maturation induced by microbial molecular patterns, in particular by agonists for microbial pattern recognition receptors . These agonists may directly enhance the intracellular mechanism for cross-presentation [29, 40, 42, 56]. The role of maturation is difficult to study with isolated DC because these undergo what is termed “spontaneous” maturation in culture. Nevertheless, we were able to show that maturation of Flt3L DC, either with LPS or with poly IC, improved cross-presentation of anti-DEC-p24.
Traditionally cross-presentation requires high doses of antigens with the most sensitive protein, OVA, typically being used at concentrations of >50 μg/ml or 1 nM to detect a signal even with OT-I TCR transgenic reporter T cells [33, 57]. In our prior experience with mouse DC, cross-presentation of OVA to OT-I T cells under similar culture conditions to the gag experiments in our paper required concentrations of 60-250 μg/ml OVA . Here, by using the DEC-205 receptor to target the HIV gag protein to DC, we find that dose as low as 1 μg/ml of anti-DEC-p24 or 5 pM leads to strong presentation to CD8+ T cells from a polyclonal population. Also, relative to non-targeted gag p24 protein, DEC-205 mediated cross-presentation is 100 fold more efficient. The quantitative aspects of our data highlight the potential value that Flt3L mobilization and receptor mediated targeting potentially play in the use of cross-presentation to present non-replicating antigens to CD8+ T cells.
While the goal of this paper is to highlight the usefulness of Flt3L-mobilized DC in cross-presentation of a microbial protein, Flt3L mobilization may have a role in enhancing presentation of DC targeted vaccines. Others have reported that daily injections of Flt3L over a period of 9 days is able to expand DC numbers several fold in humans [58, 59]. Further experiments are required to assess if there are ways to combine DC mobilization and DC targeting to improve vaccination.
Balb/c × C57Bl/6 (C × B6) F1 mice from Harlan were maintained under specific pathogen-free conditions and used at 6-8 wk of age in accordance with Rockefeller University Animal Care and Use Committee guidelines.
Melanoma cells expressing Fms-like tyrosine kinase 3 ligand (Flt3L), were established via retroviral gene transfer  and generously provided by L. Santambrogio (Albert Einstein College of Medicine, New York, NY). B16 Flt3L melanoma cells were cultured with DMEM containing 10% FBS and 5 × 106 were injected s.c into the belly region of mice. After 15-20 days, all major splenic DC subsets had expanded >10 fold as shown previously  and reproduced here.
We purchased from BD Biosciences FITC-antibodies to CD3ε (145-2C11), B220 (RA3-6B2), MHC-II (IAb) and DX5, PE- antibodies to CD8α (53-6.7), PDCA-1 and anti-CD11c (HL3 clone), PerCP-anti-CD4 (RM4-5), PerCP-Cy5.5-anti-CD8α, PE-Cy 7 anti-IFN-γ (XMG1.2), APC-antibodies anti-IFN-γ (XMG1.2) and thy 1.2 (CD 90.2, clone 53-2.1), Alexa Fluor 700 anti-CD3 (500A2), APC-Alexa 647-antibodies to DEC-205 and B220 as well as Cytofix/Cytoperm kit and Stabilizing Fixative. Anti-CD11c beads (N418) were from Miltenyi Biotec. Rat anti MHC class II (TIB120, M5/114.15.2) was from ATCC. Anti-rat IgG Dynalbeads and Live/Dead Fixable Aqua vitality dye were from Invitrogen. Polyinosinic-polycytidylic acid (poly IC) was from Thermo Scientific. Other reagents were LPS from Escherichia coli 0127:B8 and BFA from Sigma-Aldrich.
The fusion mAb anti-DEC-p24 and the control Ig-p24 were prepared as described  and were characterized by SDS-PAGE and Western Blotting (HRP-conjugated anti-mouse IgG or anti-HIV gag). Binding of the fusion mAb to stable DEC-205 transfected CHO cells was tested by FACS analysis as described .
The cDNA for the HIV gag p24 (clade B) was fused to the sequence containing a signal peptide and a FLAG epitope tag. This construct named SF-p24 (GenBank accession number GQ304738) was cloned into pCMV expression vector and stably transfected into CHO cells, in order to produce a soluble, FLAG-tagged (SF) protein of gag p24, which was purified without endotoxin contamination from the culture supernatants of CHO/SF-p24 cells, by anti-FLAG® M1 Affinity Gel (Sigma-Aldrich, St. Louis, MO) following manufacturer's instruction.
Overlapping (staggered by 4 aminoacids) 15 mer peptides spanning the entire HIV gag p17 or p24 sequence were synthesized by H. Zebroski in the Proteomics Resource Center (The Rockefeller University). HIV gag p24 or p17 peptides were resuspended at 1 mg/ml of each peptides in 100% DMSO and added to DC at 1 μg/ml.
The cDNA for the HIV gag p24 (clade B) was fused to the sequence containing a signal peptide and a FLAG epitope tag. This construct named SF-p24 (GenBank accession number GQ304738) was cloned into pCMV expression vector and stably transfected into CHO cells, in order to produce a soluble, FLAG-tagged (SF) protein of gag p24. The SF-p24 protein was purified without endotoxin contamination from the culture supernatants of CHO/SF-p24 cells, by Anti-FLAG® M1 Affinity Gel (Sigma-Aldrich, St. Louis, MO) following the manufacturer's instruction.
Spleens were removed from Flt3L treated mice, cut in small fragments, and digested into single cell-suspensions with 400 U/ml collagenase D (Roche Applied Science) for 25 min at 37°C. After inhibition of collagenase with 10 mM EDTA, the cells were resuspended in PBS in 2 mM EDTA and 2% FCS. CD11c+ DC were enriched by positive selection using anti-CD11c magnetic beads and MACS columns (Miltenyi Biotec). To purify CD8α+ and CD8α- cells, CD11c+ cells were sorted on a FACSVantage (BD Biosciences) into B220- DX5- CD3- CD11chigh CD8α+ and B220- DX5- CD3- CD11chigh CD8α- fractions. The purity of the DC subsets was >95-99%. Antigen primed T cells, which were from F1 mice primed with Adenovirus gag and boosted 4-6 wks later with anti-DEC-p24 and poly IC, were enriched by excluding MHC class II+ cells using TIB120/M5/114 rat mAb and anti-rat IgG Dynalbeads.
B16 Flt3L treated or untreated CxB6 F1 mice were injected i.p. with PBS or 50 μg of poly IC. Fifteen hours later spleens were collected and collagenase digested. CxB6 DC were fixed for 20 min on ice and graded numbers were added to 3×105 CFSE labeled (Molecular Probes, Eugene, OR) C57BL/6 T cells. After 4 days of culture, samples were stained with Live/Dead Fixable Violet viability dye (Invitrogen, Carlsbad, CA), Alexa 700 anti-CD3, APC Alexa 780 anti-CD8 and PerCP Cy 5.5 anti-CD4, and acquired on a BD LSR II flow cytometer (BD Biosciences). Data were analyzed with FlowJo Software (Tree Star, Inc.).
2×106 splenic CD11c+ DC, or purified DC subsets, were first pulsed with different antigens (Results) for 5 h in 24-well culture plates in a final volume of 0.5 ml RPMI 1640 containing 5% FCS and the supernatant (3% vol/vol) from J558L cells transduced with murine GM-CSF. Then 25 μg/ml poly IC or 0.1 μg/ml of LPS was added to the cultures for 15-18h. The DC were washed three times with PBS and added to antigen-primed T cells. IFN-γ production during a 6 h DC:T cell coculture in the presence of 10 μg/ml of BFA was monitored by washing the cells, incubating 10 min at 4°C with 2.4G2 mAb to block FcγR, and staining with Live/Dead Fixable Aqua, anti-CD3, anti-CD8 and anti-CD4 mAbs for 20 min at 4°C. Cells were fixed and permeabilized 10 min with Cytofix/Cytoperm and stained with APC-conjugated anti-IFN-γ mAb for 15 min at room temperature and resuspended in stabilizing fixative. 105 live-CD3+ events were acquired using a BD LSR II flow cytometer. Data were analyzed with FlowJo Software.
Statistical significance was evaluated using two-tailed Student's t test, 95% confidence interval. Results are expressed as means ± SD. In the figures, p values of 0.05 are labeled with a single asterisk (*), 0.01 (**) or 0.001 (***). Analysis was performed with a Prism 3 program (Graphpad Sofware Inc.).
We thank Laura Santambrogio (Albert Einstein College of Medicine New York, NY) for Flt3L transduced B16 melanoma cells; Klara Velinzon for expert technical assistance with cell sorting; Henry Zebroski for synthesizing gag peptides; Juliana Idoyaga for anti-DEC-205 APC-Alexa 647; Bei Wang, Cheolho Cheong and Jaehoon Choi for their involvement in the late phase of this work; Olga Mizenina for endotoxin tests; and Judy Adams for help with graphics. This work was supported by grants from NIAID, AI13013 and AI40874 to RMS.
Conflict of interest. RMS is a paid scientific consultant to Celldex Therapeutics which is developing DEC-205-targeted vaccines.