PeproTech Inc.
Originally published online as doi:10.1189/jlb.1105694 on March 30, 2006

Published online before print March 30, 2006
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
jlb.1105694v1
79/6/1271    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
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
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Harui, A.
Right arrow Articles by Basak, S. K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Harui, A.
Right arrow Articles by Basak, S. K.
(Journal of Leukocyte Biology. 2006;79:1271-1278.)
© 2006 by Society for Leukocyte Biology

Adenoviral-encoded antigens are presented efficiently by a subset of dendritic cells expressing high levels of {alpha}vß3 integrins

Airi Harui*, Michael D. Roth*,{dagger}, Darshni Vira*, Mihir Sanghvi*, Hiroyuki Mizuguchi{ddagger} and Saroj K. Basak*,1

* Division of Pulmonary & Critical Care Medicine and
{dagger} Jonsson Comprehensive Cancer Center, David Geffen School of Medicine at University of California Los Angeles; and
{ddagger} National Institute of Biomedical Innovation, Ibaraki, Osaka, Japan

1Correspondence: Division of Pulmonary & Critical Care, Department of Medicine, UCLA School of Medicine, Los Angeles, CA 90095-1690. E-mail: sbasak{at}mednet.ucla.edu


arrow
ABSTRACT
 
Dendritic cells (DC) play a central role in antigen presentation and are often targeted by adenoviral (Ad)-based gene therapy. However, DC lack the coxsackie-Ad receptor, and little is known about the process by which they acquire and present Ad-encoded antigens. We examined the expression of {alpha}{nu}ß3 integrins (CD51/CD61) on mouse bone marrow-derived DC (BM-DC) and their susceptibility to transduction by Ad vectors. Less than 10% of BM-DC precursors expressed CD51, but expression increased over time in culture with granulocyte macrophage-colony stimulating factor (GM-CSF)/interleukin (IL)-4. After 7 days, 28 ± 1.7% of CD11c+ DC expressed high levels of CD51 (CD51hi), and the remaining DC expressed low levels of CD51 (CD51lo). CD51hi CD express higher major histocompatibility complex type 1 (MHC I); however, both of the DC subsets expressed similar levels of MHC II and costimulatory molecules. When exposed to a first-generation Ad vector, transgene expression was restricted to the CD51hi DC subset and blocked by soluble peptides expressing an arginine, glycine, aspartic acid (RGD) sequence, confirming the role of integrins in viral entry. Consistent with this, a modified Ad expressing an RGD-binding sequence in its fiber knob (Ad-RGD) transduced the CD51hi DC subset with significantly higher efficiency. When BM-DC were transduced with an Ad-expressing ovalbumin (Ad-OVA), the CD51hi subset proved superior in activating OT-I (T cell receptor-OVA) T cells. Similar to in vitro effects, systemic administration of GM-CSF/IL-4 increased the expression of CD51 on splenic DC and rendered these cells susceptible to Ad transduction. These results suggest that a limited subset of DC expressing high levels of {alpha}{nu}ß3 integrins is preferentially transduced by Ad vectors and activates CD8+ T cell responses against Ad-encoded antigens.

Key Words: gene therapy • vaccination • RGD sequence • ovalbumin • antigen presentation


arrow
INTRODUCTION
 
Dendritic cells (DC) play a central role in antigen uptake, processing, and presentation in vivo. Large numbers of DC can also be generated ex vivo when mouse bone marrow (BM) cells are cultured for 7–10 days with granulocyte macrophage-colony stimulating factor (GM-CSF) and interleukin (IL)-4 [1 ]. Once loaded with a target antigen, these BM-derived DC (BM-DC) have been administered as a cell-based vaccine in a variety of experimental animal models [2 3 4 ]. Several approaches to antigen loading have been examined, including the use of peptides, whole protein, RNA, cDNA, and transduction by a variety of vectors [5 6 7 ]. Use of adenoviral (Ad) vectors has been especially attractive, as they are easy to manufacture, express high levels of transgene antigen in dividing and nondividing cells, and are capable of stimulating antigen-specific immunity, even in the setting of immune tolerance [8 9 10 ]. However, human and murine DC lack expression of the coxsackie-Ad receptor (CAR), which mediates the high-affinity interaction between target cells and the fiber knob region of the Ad capsid [11 ]. As a result, Ad-based transduction of DC requires high multiplicities of infection (MOI) [5 , 12 ]. A secondary step in Ad transduction involves the binding of an arginine, glycine, aspartic acid (RGD) sequence expressed on the penton base of the Ad capsid with {alpha}vß3 and/or {alpha}vß5 integrin molecules expressed by target cells [13 , 14 ]. In the absence of CAR, this lower efficiency binding is thought to mediate transduction of DC [11 , 15 ]. Reverse transcriptase-polymerase chain reaction (RT-PCR) has demonstrated that {alpha}{nu} integrins (CD51) are expressed by human and murine DC [11 ], but their direct role in DC transduction and the presentation of Ad-encoded antigen have never been evaluated.

To address this issue, we used a fluorescein-activated cell sorter (FACS)-based approach to examine the distribution and regulation of CD51 on mouse BM-DC in vitro and splenic CD11c+ DC in vivo. This approach identified two distinct DC subsets: one subset expressing high levels of CD51 (CD51hi), which rendered the DC susceptible to Ad transduction, and one subset, which expresses low levels of CD51 (CD51lo) and was refractory to transduction. The capacity for CD51hi and CD51lo DC to present Ad-encoded transgene antigens was examined by sorting the two populations and testing their ability to stimulate transgenic CD8+ T cells expressing T cell receptors (TCR) specific for the encoded antigen. Our findings suggest that a limited subset of DC expresses high levels of {alpha}{nu}ß3 (CD51hi/CD61hi) integrins, regardless of whether they are BM-DC generated in vitro or fresh DC recovered from the spleen. Expression of {alpha}{nu}ß3 by DC is related directly to their susceptibility to Ad-based transduction and their capacity to stimulate antigen-specific CD8+ T cell responses.


arrow
MATERIALS AND METHODS
 
Mice and cell lines
Male C57BL/6 mice and OT-I transgenic [TCR-ovalbumin (OVA)] mice, 6–8 weeks old, were purchased from Charles River Laboratories (Wilmington, MA) and The Jackson Laboratories (Bar Harbor, ME), respectively. They were housed at the University of California Los Angeles (UCLA) animal facility, and the UCLA Animal Research Committee approved all procedures.

Cytokines and in vitro generation of BM-DC
Recombinant mouse GM-CSF (specific activity ≥5x106 units/mg) and IL-4 (specific activity ≥1x107 units/mg) were acquired from Peprotech (Rocky Hill, NJ). BM-DC were generated from murine BM as described previously [1 ]. Briefly, BM cells were harvested and depleted of natural killer (NK), B, and T cells and granulocytes by treatment with monoclonal antibodies (mAb; PharMingen, San Diego, CA) and complement (Sigma Chemical Co., St. Louis, MO). DC precursors were cultured overnight in RPMI 1640 (Irvine Scientific, Santa Ana, CA) with 10% heat-inactivated fetal calf serum (FCS; Omega Scientific, Tarazana, CA), HEPES (Sigma Chemical Co.), 2-mercaptoethanol (2-ME; Sigma Chemical Co.) and penicillin-streptomycin (Sigma Chemical Co.). The next day, adherent cells were removed, and 20 ng/ml each GM-CSF and IL-4 was added. Cell cultures were incubated for 7–9 days with a media change every 3 days.

Cytokine-induced generation of DC in situ
DC were also generated in situ by treating animals with a continuous in vivo infusion of GM-CSF and IL-4 as described previously [4 ]. Briefly, GM-CSF and IL-4 were delivered by a mini-osmotic pump for 7 days (10 µg each/day). After 7 days, spleen cells from control and cytokine-treated animals were isolated, and CD11c+ DC were purified using magnetic beads coated with anti-CD11c mAb according to the manufacturer’s specifications (Miltenyi Biotec Inc., Auburn, CA). The resulting DC population was cultured briefly overnight with GM-CSF and IL-4 (10 ng each/ml) prior to further analysis.

FACS analysis of DC subsets
DC were identified by their expression of CD11c (Caltag Laboratories, Burlingame, CA) and then analyzed for their concurrent expression of {alpha}{nu} integrin (CD51, BD PharMingen) and ß3 integrin (CD61, BD PharMingen) by standard, multichannel FACS analysis using a FACSCalibur cytometer (Becton Dickinson, San Jose, CA) and FCS Express analysis software (De Novo Software, Ontario, Canada). CD11c+ DC were also analyzed for the expression of several DC markers including major histocompatibility complex (MHC) classes I and II, CD80, and CD86 (Caltag Laboratories).

Ad vectors
Five different Ad vectors were used: three first-generation E1(-)/E3{Delta} Ad vectors expressing green fluorescent protein (GFP; Ad-GFP) or OVA (Ad-OVA) or no transgene (Ad-RR5) and two RGD fiber mutant Ad expressing the same transgenes (Ad-RGD-GFP and Ad-RGD-OVA). All of these vectors have been described previously [15 , 16 ]. Ad-RGD vectors express an RGD sequence inserted into the HI loop region of their fiber knob. Ad vectors were propagated in 293 cells, purified by CsCI density centrifugation, dialyzed, and stored at –80°C. Viral particle titers were determined spectrophotometrically by the methods of Mittereder et al. [17 ], and ~100 viral particles were equivalent to one plaque-forming unit for the Ad-GFP vector.

Transduction of DC and blocking by RGD peptides
DC were transduced with Ad as described previously [1 ]. Briefly, 2.5–5.0 x 105 DC were exposed to a suspension of Ad at different MOI (up to 2000 viral particles per target cell) in 200 µl medium containing half phosphate-buffered saline and half RPMI 1640, supplemented with 2% fetal bovine serum. Cultures were incubated for 2 h at 37°C in a humidified incubator. An additional 800 µl complete media containing GM-CSF and IL-4 (20 ng each/ml) were then added, and cultures were incubated for another 48 h before analysis. GFP expression was evaluated by FACS.

In selected experiments, exogenous peptides expressing an RGD sequence (RGD-4C; AnaSpec, San Jose, CA) were used as blocking agents to evaluate the role of {alpha}v integrins in mediating viral entry and transduction [18 ]. BM-DC (5x105) were preincubated with 20 µM or 200 µM RGD-4C peptide at 4°C for 60 min. Ad vectors (Ad-GFP or Ad-RGD-GFP) were then added at a MOI of 1000 and incubated at 4°C for the first hour, followed by a 37°C incubation for the second hour. Cells were treated with 1 mM trypsin-EDTA (Invitrogen, Calsbad, CA) for 2 min at 37°C and then DNase (0.02 mg/ml, Sigma Chemical Co.) for 10 min at 37°C to disrupt and degrade vectors, which had not been internalized. After additional washing, cells were incubated for a final 40–48 h prior to assessing transduction efficiency by FACS.

Antigen-specific stimulation of OT-I T cells
BM-DC and the CD51hi and CD51lo subsets, which had been transduced with Ad-OVA or control Ad-RR5 vector (MOI 1000 particles/cell), were evaluated after 48 h of culture for their ability to present transgene antigen and activate OVA-specific CD8+ T cell responses using transgenic OT-I T cells, which were isolated from lymph nodes of OT-I mice, and the CD8+ subset was purified by negative depletion using anti-B220, anti-Gr.1, anti-NK1.1, and anti-CD4 mAb and complement. Assays were performed in a round-bottomed 96-well culture plate. OVA-specific CD8+ T cells (2x105) were cocultured with transduced DC at DC:T ratios of 1:25, 1:50, 1:100, and 1:200. Cells were cultured in 200 µl RPMI 1640 containing 10% FCS and 10–4 M 2-ME at 37°C for 3 days. Cultures were then pulsed with 1.25 uCi (46.2 kBq) [H3]-thymidine, and cells were harvested 12 h later. T cell proliferation was determined by counting each sample in a liquid scintillation ß-counter. To evaluate specific BM-DC subsets, Ad-transduced cells were stained with fluorescent anti-CD11c and anti-CD51 mAb. The CD11c+/CD51hi and CD11c+/CD51lo cells were then sorted using a FACSVantage SE cell sorter (Becton Dickinson).


arrow
RESULTS
 
Changes in integrin {alpha}v expression during DC differentiation
BM-DC progenitors were prepared by depleting lineage-specific cells from the marrow, and the CD11c+ cells were examined for expression of CD51 by FACS at the beginning of culture and again on Days 5, 7, and 9 of exposure to GM-CSF and IL-4 (Fig. 1 ). At the start of culture, only 5–12% of cells were CD51hi. By Day 5, this had increased to 20 ± 2.3% of the total DC population, and by Day 7, 28 ±1.7% of the cells were CD51hi. Changes in CD51 expression between Days 7 and 9 were minor, and there were no further increases observed, even with extended culture. These results indicate that BM-DC precursors are heterogeneous with respect to expression of the {alpha}v integrin, and the subpopulation expressing high levels of CD51 increases during the initial stages of DC culture, but CD51hi DC remain as a minor population.


Figure 1
View larger version (16K):
[in this window]
[in a new window]
  		Figure 1. The percentage of CD11c+ BM precursors and CD11c+ DC expressing high levels of (CD51hi) when cultured in vitro with GM-CSF and IL-4. Mouse BM was depleted of NK cells, B cells, T cells, and granulocytes by treatment with mAb and complement. Adherent cells were depleted after an overnight culture, and the remaining cells were cultured with 20 ng/ml each GM-CSF and IL-4. FACS analysis was performed on culture on Days 0, 5, 7, and 9 with dual staining for CD11c and CD51. Mean values ± SE from three individual experiments.

Coexpression of integrin {alpha}v and ß3 chains on DC correlates with susceptibility to Ad transduction
The integrin {alpha} and ß chains heterodimerize to form functional integrin molecules. By RT-PCR, DC express {alpha}vß3 and {alpha}vß5, two distinct integrins that bind RGD sequences [11 , 15 ]. In this case, 100% of the DC expressing high levels of CD51 coexpressed high levels of CD61 on the cell surface (CD51hi/CD61hi), consistent with the expression of {alpha}vß3 (Fig. 2 ). Less than one-third of the total CD11c+ DC was positive, and the majority of the DC expressed only low levels of CD51 and CD61 (CD51lo/CD61lo).


Figure 2
View larger version (21K):
[in this window]
[in a new window]
  		Figure 2. Expression pattern for CD51 ({alpha}v) and CD61 (ß3) on DC subsets. BM-DC were harvested after 7 days of culture with GM-CSF/IL-4 and stained with fluorescent antibodies to detect CD11c, CD51, and CD61. (A) Two-dimensional dot-plot demonstrating two distinct CD11c+ DC subsets expressing high or low levels of CD51. (B) CD51 and CD61 expression on the CD51hi DC subset (solid histogram). (C) CD51 and CD61 expression on the CD51lo DC subset (solid histogram). Empty histograms represent control antibody staining of DC. Representative experiment, n = 6.

When the entire BM-DC population was cultured with a first-generation Ad-GFP vector, there was a concentration-dependent expression of GFP 48 h later. MOI of 25, 250, and 1000 resulted in 1–2%, 3–6%, and 7–15% of the target DC population being transduced, respectively. Higher transduction efficiencies ranging between 50% and 80% could be achieved at higher MOI of 2000–6000 virus particles/cell; however, exposure to this level of virus resulted in cytopathic effects and the subsequent loss of 20–30% of the transduced DC (data not shown). To characterize the relationship between integrin expression and transduction efficiency, cells that had been cultured with a MOI of 1000 were counterstained with fluorescent-labeled anti-CD11c and anti-CD51 mAb and evaluated by FACS (Fig. 3A 3B 3C ). Close to 100% of the GFP expression resided in the CD51hi DC subset, with minimal expression by the CD51lo subset. Hypothesizing that transduction was mediated by the interaction of {alpha}vß3 on DC with RGD sequences on the Ad capsid, we repeated these studies with the Ad-RGD-GFP vector. At the same MOI of 25, 250, and 1000, this vector produced superior transduction rates (2–4%, 15–22%, and 42–46% of the DC, respectively). Similar to results with the first-generation Ad, GFP expression resided primarily in the CD51hi subset (Fig. 3D 3E 3F) . Essentially, 100% of the CD51hi DC were transduced by Ad-RGD-GFP at a MOI of 1000 compared with only 30% with Ad-GFP. Furthermore, the level of expression was increased approximately tenfold over that produced by the first-generation Ad-GFP. The CD51lo subset was transduced at a significantly lower rate and intensity. Increasing the MOI of either vector to 2000 further increased GFP expression, but the pattern remained the same, and the overwhelming majority of GFP was detected in the CD51hi subset. These results indicate a marked and preferential transduction of CD51hi DC, which can be enhanced by targeting the Ad to interact with {alpha}vß3 and/or by increasing the viral load used for transduction.


Figure 3
View larger version (42K):
[in this window]
[in a new window]
  		Figure 3. Expression of the GFP transgene by CD51hi and CD51lo DC subsets following transduction by Ad-GFP or Ad-RGD-GFP. BM-DC were harvested after 7 days of culture with GM-CSF/IL-4 and transduced with Ad-GFP (A–C) or Ad-RGD-GFP (D–F) at a MOI of 1000 or 2000 (solid histograms). FACS analysis was used to detect GFP expression in the CD11c+/CD51hi (B, E) and CD11c+/CD51lo (C, F) DC subsets. Empty histograms represent background fluorescence by untransduced DC subsets. Values represent mean fluorescence intensity (MFI). Representative experiment, n = 8.

Role of RGD in Ad transduction
The role of {alpha}vß3 integrins in the transduction process was assessed further by competitive inhibition with soluble RGD peptides (Fig. 4 ). DC were transduced with Ad-GFP (Fig. 4A) or Ad-RGD-GFP (Fig. 4C) in the presence or absence of RGD peptides at different concentrations (20 µM or 200 µM). In both cases, there was a concentration-dependent decrease in the percentage of target cells expressing GFP. At an RGD concentration of 20 µM, transduction by Ad-GFP was inhibited 57%, and Ad-RGD-GPF was inhibited 72%. At 200 µM, inhibition increased to 68% and 82%, respectively (B, D).


Figure 4
View larger version (29K):
[in this window]
[in a new window]
  		Figure 4. Exogenous RGD peptide blocks transduction of DC by Ad and Ad-RGD in a concentration-dependent manner. BM-DC were harvested after 7 days of culture with GM-CSF/IL-4 and cocultured with RGD peptide at 20 µM or 200 µM for 1 h (4°C), followed by infection with Ad-GFP or Ad-RGD-GFP at a MOI of 1000. After 48 h, cells were stained with CD11c and evaluated for the percentage of DC expressing GFP by FACS analysis (A, C). The percentage of DC transduced under each condition was used to determine the magnitude of inhibition produced by different concentrations of RGD-blocking peptide (B, D). Representative experiment, n = 3.

Efficient presentation of transgene antigen is mediated by the DC subset expressing high levels of {alpha}vß3
CD51hi and CD51lo DC were evaluated for their expression of cell surface markers (CD11c, CD51, CD61) and molecules centrally involved in antigen presentation (MHC classes I and II, CD80, and CD86) by FACS (Fig. 5 ). Although the CD51hi subset expressed three- to fourfold higher levels of CD11c, CD51, and CD61, they also expressed 2.2 ± 0.05-fold higher MHC I than the CD51lo DC subset. There were slight differences in the expression of other markers (MHC II, CD80, CD86) when the two subsets were compared on multiple occasions, but these were minor and not reproducible from experiment to experiment.


Figure 5
View larger version (32K):
[in this window]
[in a new window]
  		Figure 5. Expression of cell surface integrins and activation markers on DC subsets before and after transduction by Ad-GFP. BM-DC were harvested after 7 days of culture with GM-CSF/IL-4 and stained with fluorescent-labeled anti-CD11c and anti-CD51 mAb in combination with anti-CD61, anti-MHC I, anti-MHC II, anti-CD80, or anti-CD86 mAb. (A) Relative expression (MFI) of markers on CD11c+/CD51hi and CD11c+/CD51lo DC prior to transduction. (B) Relative expression (MFI) of markers on CD11c+/CD51hi and CD11c+/CD51lo DC 48 h following transduction with Ad-OVA. Individual histograms demonstrating relative expression of CD11c, MHC I, and CD86 are shown for the CD11c+/CD51hi and CD11c+/CD51lo DC subsets on control (A) or transduced cells (B). Values represent MFI. Representative experiment, n = 3.

The same analysis was repeated after exposing the cells to Ad-GFP. Although the general pattern of marker expression was similar to that observed prior to viral exposure, there was a 33–42% reduction in the expression of CD51 in both of the DC subsets, consistent with ligand-induced down-regulation of this integrin receptor. In addition, there was a significant increase in the expression of MHC I (97±20% increase) and CD86 (64±22% increase) in the CD51hi DC subset, suggesting a low level of viral-associated activation. Both DC subsets experienced a modest decrease in MHC II expression following viral exposure, but there were no significant changes in the expression of CD61 or CD80.

To measure antigen-presenting cell (APC) activity, DC were transduced with Ad-OVA at a MOI of 1000, cultured for 48 h, and then sorted into CD51hi and CD51lo subsets. Compared with untransduced DC or DC transduced with the Ad-RR5 vector, which does not express a transgene (data not shown), there was a significant stimulation of OT-1 proliferation by Ad-OVA-transduced CD11c+ DC at all of the DC:T cell ratios (Fig. 6A ). Although the CD51hi and CD51lo DC subsets stimulated the proliferation of OT-1 cells, stimulation by CD51hi DC at any given DC:T cell ratio was two- to threefold higher than the response to CD51lo DC (Fig. 6B) . Neither untransduced DC, whether sorted or not (Fig. 6A and 6B) , or DC transduced with the Ad-RR5 vector (data not shown) stimulated OT-I OVA-specific T cells at any given DC:T cell ratio.


Figure 6
View larger version (16K):
[in this window]
[in a new window]
  		Figure 6. Activation of transgenic OT-I T cells by the CD51hi and CD51lo DC subsets following Ad-OVA transduction. BM-DC were harvested after 7 days of culture with GM-CSF/IL-4, transduced with Ad-OVA at a MOI of 1000, and cultured for 48 h prior to coculture for 3 days with 2 x 105 OVA-specific CD8+ T cells (from OT-I mice) at various DC:T ratios. Proliferation was determined by [3H]-thymidine incorporation and is expressed in counts per minute (cpm). (A) Unsorted, control DC and DC transduced with Ad-OVA (DC-Ad-OVA) were used to stimulate OT-I T cells. (B) Control DC and DC transduced with Ad-OVA (DC-Ad-OVA) were stained with fluorescent-labeled anti-CD11c and anti-CD51 and FACS-sorted into CD11c+/CD51hi and CD11c+/CD51lo subsets prior to coculture with OT-I cells. Results are the mean ± SE of triplicate cultures. P < 0.05, comparing proliferation induced by transduced and control DC at all DC:T cell ratios and when comparing transduced CD51hi to transduced CD51lo DC. Representative experiment, n = 3.

CD51 expression increases in vivo in response to systemic GM-CSF and IL-4
Systemic administration of GM-CSF and IL-4 has been shown to expand and activate DC in vivo and enhance immune responses to an Ad-based vaccination [4 ]. Hypothesizing that cytokine-mediated changes in the CD51hi DC subset might play a role in this effect, we examined splenic CD11c+ DC for their expression of CD51 and susceptibility to Ad transduction (Fig. 7 ). In control mice, DC expressing CD51 represented <5% of the total CD11c+ population, and treatment with GM-CSF and IL-4 increased this percentage by three- to fourfold (Fig. 7A) . These effects from a 7-day infusion of cytokines were similar in magnitude to those observed when BM-DC precursors were exposed to GM-CSF and IL-4 for 7 days in vitro (Fig. 1) .


Figure 7
View larger version (17K):
[in this window]
[in a new window]
  		Figure 7. Systemic treatment with GM-CSF/IL-4 in vivo increases the CD51hi DC subset and their susceptibility to transduction by Ad-RGD-GFP. GM-CSF and IL-4 (10 µg/day each) were delivered by a mini-osmotic pump for 7 days in vivo, and DC were purified from control and cytokine-treated mice using magnetic beads coated with anti-CD11c mAb. (A) DC from control and cytokine-treated mice were stained with anti-CD11c and anti-CD51 to enumerate the percentage of CD11c+ DC that expressed high levels of CD51 (% CD11c+/CD51hi). (B) DC from control and GM-CSF/IL-4-treated mice were transduced with Ad-RGD-GFP at a MOI of 4000, cultured for 48 h, and the expression of GFP on CD11c+ DC was determined by FACS analysis. Values represent MFI of the populations expressing GFP. Representative experiment, n = 2.

Purified CD11c+ DC from the spleens of control and cytokine-treated animals were then transduced in vitro with Ad-RGD-GFP at a MOI of 1000 (Fig. 7B) . In control DC, only a minor fraction of the CD51hi cells expressed GFP after 48 h. In contrast, transduction of DC from GM-CSF/IL-4-treated mice was increased dramatically, and GFP expression increased by up to sevenfold. These results suggest that exposure to GM-CSF/IL-4 and expression of CD51 mediate similar roles in regulating the transduction of DC in vivo as they do in vitro.


arrow
DISCUSSION
 
DC can be expanded from precursors in vitro and differentiated into potent APC. However, to be transformed into an effective cell-based vaccine, DC must first be loaded with antigens. Antigen loading is a crucial factor in determining vaccine efficiency, directly impacting on the avidity, magnitude, and breadth of the T cell response [19 , 20 ]. Viral vectors have several advantages over peptides, whole protein, and other gene-based approaches for antigen loading. Not only do they generate an intracellular protein that can be optimally processed into immunogenic MHC classes I and II peptides, but they result in sustained antigen production, the magnitude of which relates directly to the capacity for T cell activation [1 , 20 ]. Although Ad vectors are used frequently for inserting transgenes into DC, it is almost paradoxical that these target cells lack expression of CAR, the primary Ad receptor [15 ]. This paradox led us to investigate {alpha}v integrins as the alternative pathway by which DC are transduced. It is suprising that we found that high levels of {alpha}vß3 are expressed on only a minor subpopulation of CD11c+ DC. As a result, Ad-based transduction directs antigen loading to the CD51hi subset, which in turn, becomes the primary DC population responsible for T cell activation.

The normal sequence for Ad-based transduction involves a high-affinity binding between the fiber knob region of the vector and CAR expressed by the target cell [14 , 21 ]. After this interaction, RGD sequences, which reside in the penton base of the vector, interact with {alpha}{nu}ß3 and/or {alpha}{nu}ß5 integrin molecules on the target cell to promote internalization. These integrin receptors are heterodimers containing {alpha} and ß chains. Several investigators have used RT-PCR and/or antibody approaches to confirm that DC lack CAR yet express {alpha}{nu}, ß3, and ß5 [12 , 15 , 22 ]. However, the heterogeneity of these receptors on DC and the impact of this heterogeneity on viral transduction have not been addressed previously. Similar to our findings, Brossart and co-workers [12 ] reported that CD51 was expressed at low levels on the immortalized C57BL/6 murine DC line JAWS II and that expression was modulated by exposure to cytokines or maturation signals. Similarly, staining of human monocyte-derived DC with antibodies against {alpha}{nu}ß3 demonstrated that only a subpopulation expressed this integrin [22 ] and that transduction was limited to a fraction of the total cells.

We used multiparameter FACS analysis and cell sorting to directly assess the impact of integrin heterogeneity on viral transduction and antigen presentation. In the presence of GM-CSF and IL-4, the BM-DC differentiated from their precursor into myeloid DC (CD11b+/CD8–), and a gradual increase in the expression of CD51 was observed. Although others have observed that stimulation with lipopolysaccharide modestly or significantly increased the expression of CD51, we did not observe any effect from this maturation stimulus (data not shown). FACS also demonstrated that the CD51hi cells uniformly express CD61, suggesting the presence of {alpha}{nu}ß3 heterodimers. Transcriptional regulation of CD51 and CD61 has been studied extensively and likely explains these results. De Nichilo and Burns [23 ] first noted differential regulation of mRNA encoding for ß3 and ß5 depending on whether macrophages were cultured with M-CSF, which selectively increased ß5, or GM-CSF, which selectively increased the expression of ß3. Huang and co-authors [18 ] reported identical findings with human monocytes and related the change in integrin receptor expression directly to a change in susceptibility to Ad-based transduction. Kitazawa and co-workers [24 ] assessed several cytokines for their effects on murine BM-derived macrophages. Although GM-CSF increased expression of ß3 mRNA by approximately twofold, IL-4 had the strongest effect and resulted in a fivefold increase. As such, it appears that exposure of BM precursors to GM-CSF and IL-4 not only drives their differentiation into DC but also their expression of {alpha}{nu}ß3 integrins. The reason why this effect is limited to a subset of CD11c+ DC remains to be determined.

Next, we hypothesized that expression of CD51 would dictate susceptibility to Ad-based transduction. As previously reported [12 ], DC were relatively refractory to first-generation Ad-GFP, and transgene expression was restricted entirely to CD51hi cells at noncytolytic viral loads. Ad binding to {alpha}v integrins is mediated by its interaction with RGD sequences located on the penton base of the Ad capsid [14 ]. These residues are strericly hindered by their location, and a new generation of fiber-modified vectors, Ad-RGD, has been developed, in which an additional RGD sequence is inserted into a more favorable position on the HI loop of the fiber knob [16 , 25 ]. This has been shown by several investigators to enhance the transduction of DC and to increase the potency of Ad-based vaccines [10 , 11 , 15 , 26 , 27 ]. We hypothesized that although the Ad-RGD vector still targets {alpha}v integrins, it would preferentially transduce the CD51hi DC subset. Indeed, although transduction efficiency and GFP expression were improved dramatically in response to Ad-RGD-GFP, transgene expression was still predominantly in the CD51hi population. The limited transduction of CD51lo DC by this vector is likely a result of the increased binding affinity between the RGD sequence expressed on the fiber knob and its interaction with even low levels of {alpha}v integrins. To confirm these assumptions, we repeated transduction studies in the presence of exogenous RGD peptides, which blocked transduction by the conventional Ad-GFP and Ad-RGD-GFP vectors in a concentration-dependent manner.

It has been variably reported that DC transduction results in cell maturation and enhanced antigen presentation. At baseline, there were some differences between the CD51hi and CD51lo subsets with respect to expression of MHC class I and CD11c, which were higher in the CD51hi DC subset. Expression of CD51 decreased following exposure to Ad, suggesting ligand-induced down-regulation mediated by RGD sequences on the vectors. MHC II expression was also reduced modestly in both subsets, consistent with other reports that Ad transduction transiently suppresses the immunostimulatory potential of DC [28 ]. However, there was also a consistent increase in the expression of MHC class I molecules and CD86 on the surface of the CD51hi DC, which was not observed in the CD51lo population. The significance of these modest changes in MHC is not currently known but may represent selective activation of the CD51hi DC.

Having identified differential Ad transduction, antigen loading, and MHC expression based on the presence of CD51, we next evaluated the impact of CD51 expression on antigen-presenting activity. Ad-OVA and transgenic TCR-OVA T cells directly facilitated this analysis, as did the capacity to sort the DC into highly purified subsets. CD51hi DC were significantly more potent than CD51lo DC in stimulating transgene-specific T cells, consistent with their preferential antigen loading. However, it was surprising that the sorted CD51lo DC, despite their limited transduction, still activated OT-I T cells. One explanation is that even low levels of transduction, below the level of detection, are sufficient. Brossart and co-workers [12 ] suggested this to explain why transduced targets are killed at a substantially higher frequency than predicted by staining for expression of transduced antigen. However, FACS analysis is exquisitely sensitive at detecting GFP, and this explanation seems unlikely. Alternatively, as we transduced the entire BM-DC population together and cultured them prior to sorting, it is possible that the CD51lo subset acquired OVA from the culture medium. Cross-presentation of antigen following viral transduction has been reported previously [29 30 31 ]. Using transgenic receptor knock-out models, it has also been shown that expression of {alpha}{nu}ß3 is not essential for MHC class I cross-presentation [32 ], which provides a plausible mechanism to explain antigen presentation by the CD51lo subset under the current study conditions, but further clarification is warranted.

Finally, we were interested in determining the relationship of our findings to DC populations as they exist in situ. To address this, we administered GM-CSF and IL-4 directly to mice for 7 days. This form of cytokine therapy has been shown to enhance the number and function of DC in animals as well as in humans [4 , 33 ]. Further, we had previously demonstrated that administration of GM-CSF and IL-4 synergizes with an Ad-based vaccination to induce potent, transgene-specific, immune responses [4 ]. Based on our in vitro studies, we recovered splenic DC from control and cytokine-treated mice and examined them for their expression of CD51 and their susceptibility to Ad-based transduction. There was a striking threefold increase in the percentage of CD11c+ DC, which expressed CD51 when recovered from cytokine-treated mice. Furthermore, the increase in CD51 correlated with a heightened susceptibility to Ad transduction. These findings suggest an important synergism between the use of GM-CSF/IL-4 and Ad-based vaccination.

In summary, we have identified a unique subpopulation of CD11c+ DC, which express CD51 and are expanded in response to GM-CSF and IL-4. These CD51hi DC are preferentially transduced by Ad vectors in a manner that is dependent on the interaction between {alpha}v integrins and RGD sequences. As others have reported, Ad-RGD vectors are significantly more effective at transducing and expressing transgenes in DC, but this approach still preferentially targets the CD51hi population. It is more important that the targeted transduction of CD51hi DC leads to their predominant role in activating transgene-specific T cells. Depending on the conditions, the CD51lo subset may also participate in antigen presentation, but likely as a result of cross-presentation (under the conditions studied) or low-level transduction, which can occur at high MOI. Further studies are warranted to determine if there are other unique features that distinguish the CD51hi and CD51lo DC subsets. The differential targeting and function of these DC populations should be considered for their impact on Ad-based immunotherapy.


arrow
ACKNOWLEDGEMENTS
 
This work was supported by grants from the Tobacco-Related Disease Research Program of California (S. K. B., 10KT-0086); the American Lung Association of California (S. K. B.); the Cancer Research Program, California Department of Health Services (M. D. R., #TTP1019); the UCLA SPORE in Prostate Cancer, National Institutes of Health/National Cancer Institute (M. D. R., 5P50CA092131); and the UCLA Human Gene Medicine Grant (S. K. B.). The Jonsson Comprehensive Cancer Center/UCLA provided core facilities for flow cytometry and cell sorting.

Received November 28, 2005; revised February 10, 2006; accepted February 22, 2006.


arrow
REFERENCES
 
    1
  1. Harui, A., Roth, M. D., Kiertscher, S. M., Mitani, K., Basak, S. K. (2004) Vaccination with helper-dependent adenovirus enhances the generation of transgene-specific CTL Gene Ther. 11,1617-1626[Medline]
    2. 2
  2. Ribas, A., Butterfield, L. H., McBride, W. H., Jilani, S. M., Bui, L. A., Vollmer, C. M., Lau, R., Dissette, V. B., Hu, B., Chen, A. Y., Glaspy, J. A., Economou, J. S. (1997) Genetic immunization for the melanoma antigen MART-1/Melan-A using recombinant adenovirus-transduced murine dendritic cells Cancer Res. 57,2865-2869[Abstract/Free Full Text]
    3. 3
  3. Song, W., Kong, H. L., Carpenter, H., Torii, H., Granstein, R., Rafii, S., Moore, M. A., Crystal, R. G. (1997) Dendritic cells genetically modified with an adenovirus vector encoding the cDNA for a model antigen induce protective and therapeutic antitumor immunity J. Exp. Med. 186,1247-1256[Abstract/Free Full Text]
    4. 4
  4. Basak, S. K., Harui, A., Stolina, M., Sharma, S., Mitani, K., Dubinett, S. M., Roth, M. D. (2002) Increased dendritic cell number and function following continuous in vivo infusion of granulocyte macrophage-colony-stimulating factor and interleukin-4 Blood 99,2869-2879[Abstract/Free Full Text]
    5. 5
  5. Arthur, J. F., Butterfield, L. H., Roth, M. D., Bui, L. A., Kiertscher, S. M., Lau, R., Dubinett, S., Glaspy, J., McBride, W. H., Economou, J. S. (1997) A comparison of gene transfer methods in human dendritic cells Cancer Gene Ther. 4,17-25[Medline]
    6. 6
  6. Gunzer, M., Grabbe, S. (2001) Dendritic cells in cancer immunotherapy Crit. Rev. Immunol. 21,133-145[Medline]
    7. 7
  7. Jenne, L., Schuler, G., Steinkasserer, A. (2001) Viral vectors for dendritic cell-based immunotherapy Trends Immunol. 22,102-107[CrossRef][Medline]
    8. 8
  8. Wan, Y., Bramson, J., Carter, R., Graham, F., Gauldie, J. (1997) Dendritic cells transduced with an adenoviral vector encoding a model tumor-associated antigen for tumor vaccination Hum. Gene Ther. 8,1355-1363[Medline]
    9. 9
  9. Perricone, M. A., Claussen, K. A., Smith, K. A., Kaplan, J. M., Piraino, S., Shankara, S., Roberts, B. L. (2000) Immunogene therapy for murine melanoma using recombinant adenoviral vectors expressing melanoma-associated antigens Mol. Ther. 1,275-284[Medline]
    10. 10
  10. Okada, N., Masunaga, Y., Okada, Y., Mizuguchi, H., Iiyama, S., Mori, N., Sasaki, A., Nakagawa, S., Mayumi, T., Hayakawa, T., Fujita, T., Yamamoto, A. (2003) Dendritic cells transduced with gp100 gene by RGD fiber-mutant adenovirus vectors are highly efficacious in generating anti-B16BL6 melanoma immunity in mice Gene Ther. 10,1891-1902[CrossRef][Medline]
    11. 11
  11. Okada, N., Tsukada, Y., Nakagawa, S., Mizuguchi, H., Mori, K., Saito, T., Fujita, T., Yamamoto, A., Hayakawa, T., Mayumi, T. (2001) Efficient gene delivery into dendritic cells by fiber-mutant adenovirus vectors Biochem. Biophys. Res. Commun. 282,173-179[CrossRef][Medline]
    12. 12
  12. Brossart, P., Goldrath, A. W., Butz, E. A., Martin, S., Bevan, M. J. (1997) Virus-mediated delivery of antigenic epitopes into dendritic cells as a means to induce CTL J. Immunol. 158,3270-3276[Abstract]
    13. 13
  13. Wickham, T. J., Mathias, P., Cheresh, D. A., Nemerow, G. R. (1993) Integrins {alpha} v ß 3 and {alpha} v ß 5 promote adenovirus internalization but not virus attachment Cell 73,309-319[CrossRef][Medline]
    14. 14
  14. Nemerow, G. R., Stewart, P. L. (1999) Role of {alpha}(v) integrins in adenovirus cell entry and gene delivery Microbiol. Mol. Biol. Rev. 63,725-734[Abstract/Free Full Text]
    15. 15
  15. Okada, N., Masunaga, Y., Okada, Y., Iiyama, S., Mori, N., Tsuda, T., Matsubara, A., Mizuguchi, H., Hayakawa, T., Fujita, T., Yamamoto, A. (2003) Gene transduction efficiency and maturation status in mouse bone marrow-derived dendritic cells infected with conventional or RGD fiber-mutant adenovirus vectors Cancer Gene Ther. 10,421-431[CrossRef][Medline]
    16. 16
  16. Mizuguchi, H., Koizumi, N., Hosono, T., Utoguchi, N., Watanabe, Y., Kay, M. A., Hayakawa, T. (2001) A simplified system for constructing recombinant adenoviral vectors containing heterologous peptides in the HI loop of their fiber knob Gene Ther. 8,730-735[CrossRef][Medline]
    17. 17
  17. Mittereder, N., March, K. L., Trapnell, B. C. (1996) Evaluation of the concentration and bioactivity of adenovirus vectors for gene therapy J. Virol. 70,7498-7509[Abstract]
    18. 18
  18. Huang, S., Endo, R. I., Nemerow, G. R. (1995) Upregulation of integrins {alpha} v ß 3 and {alpha} v ß 5 on human monocytes and T lymphocytes facilitates adenovirus-mediated gene delivery J. Virol. 69,2257-2263[Abstract]
    19. 19
  19. Strome, S. E., Voss, S., Wilcox, R., Wakefield, T. L., Tamada, K., Flies, D., Chapoval, A., Lu, J., Kasperbauer, J. L., Padley, D., Vile, R., Gastineau, D., Wettstein, P., Chen, L. (2002) Strategies for antigen loading of dendritic cells to enhance the antitumor immune response Cancer Res. 62,1884-1889[Abstract/Free Full Text]
    20. 20
  20. Perez-Diez, A., Spiess, P. J., Restifo, N. P., Matzinger, P., Marincola, F. M. (2002) Intensity of the vaccine-elicited immune response determines tumor clearance J. Immunol. 168,338-347[Abstract/Free Full Text]
    21. 21
  21. Bergelson, J. M., Cunningham, J. A., Droguett, G., Kurt-Jones, E. A., Krithivas, A., Hong, J. S., Horwitz, M. S., Crowell, R. L., Finberg, R. W. (1997) Isolation of a common receptor for Coxsackie B viruses and adenoviruses 2 and 5 Science 275,1320-1323[Abstract/Free Full Text]
    22. 22
  22. Varnavski, A. N., Schlienger, K., Bergelson, J. M., Gao, G. P., Wilson, J. M. (2003) Efficient transduction of human monocyte-derived dendritic cells by chimpanzee-derived adenoviral vector Hum. Gene Ther. 14,533-544[CrossRef][Medline]
    23. 23
  23. De Nichilo, M. O., Burns, G. F. (1993) Granulocyte-macrophage and macrophage colony-stimulating factors differentially regulate {alpha} v integrin expression on cultured human macrophages Proc. Natl. Acad. Sci. USA 90,2517-2521[Abstract/Free Full Text]
    24. 24
  24. Kitazawa, R., Kimble, R. B., Vannice, J. L., Kung, V. T., Pacifici, R. (1994) Interleukin-1 receptor antagonist and tumor necrosis factor binding protein decrease osteoclast formation and bone resorption in ovariectomized mice J. Clin. Invest. 94,2397-2406[Medline]
    25. 25
  25. Reynolds, P., Dmitriev, I., Curiel, D. (1999) Insertion of an RGD motif into the HI loop of adenovirus fiber protein alters the distribution of transgene expression of the systemically administered vector Gene Ther. 6,1336-1339[CrossRef][Medline]
    26. 26
  26. Okada, N., Saito, T., Masunaga, Y., Tsukada, Y., Nakagawa, S., Mizuguchi, H., Mori, K., Okada, Y., Fujita, T., Hayakawa, T., Mayumi, T., Yamamoto, A. (2001) Efficient antigen gene transduction using Arg-Gly-Asp fiber-mutant adenovirus vectors can potentiate antitumor vaccine efficacy and maturation of murine dendritic cells Cancer Res. 61,7913-7919[Abstract/Free Full Text]
    27. 27
  27. Worgall, S., Busch, A., Rivara, M., Bonnyay, D., Leopold, P. L., Merritt, R., Hackett, N. R., Rovelink, P. W., Bruder, J. T., Wickham, T. J., Kovesdi, I., Crystal, R.G. (2004) Modification to the capsid of the adenovirus vector that enhances dendritic cell infection and transgene-specific cellular immune responses J. Virol. 78,2572-2580[Abstract/Free Full Text]
    28. 28
  28. Tuettenberg, A., Jonuleit, H., Tuting, T., Bruck, J., Biermann, V., Kochanek, S., Knop, J., Enk, A. H. (2004) Early adenoviral gene expression mediates immunosuppression by transduced dendritic cell (DC): implications for immunotherapy using genetically modified DC J. Immunol. 172,1524-1530[Abstract/Free Full Text]
    29. 29
  29. Sarukhan, A., Camugli, S., Gjata, B., von Boehmer, H., Danos, O., Jooss, K. (2001) Successful interference with cellular immune responses to immunogenic proteins encoded by recombinant viral vectors J. Virol. 75,269-277[Abstract/Free Full Text]
    30. 30
  30. Morelli, A. E., Larregina, A. T., Shufesky, W. J., Sullivan, M. L., Stolz, D. B., Papworth, G. D., Zahorchak, A. F., Logar, A. J., Wang, A., Watkins, S. C., Falo, L. D., Jr, Thomson, A. W. (2004) Endocytosis, intracellular sorting, and processing of exosomes by dendritic cells Blood 104,3257-3266[Abstract/Free Full Text]
    31. 31
  31. Chhabra, A., Mehrotra, S., Chakraborty, N. G., Mukherji, B., Dorsky, D. I. (2004) Cross-presentation of a human tumor antigen delivered to dendritic cells by HSV VP22-mediated protein translocation Eur. J. Immunol. 34,2824-2833[CrossRef][Medline]
    32. 32
  32. Schulz, O., Pennington, D. J., Hodivala-Dilke, K., Febbraio, M., Reis e Sousa, C. (2002) CD36 or {alpha}vß3 and {alpha}vß5 integrins are not essential for MHC class I cross-presentation of cell-associated antigen by CD8 {alpha}+ murine dendritic cells J. Immunol. 168,6057-6065[Abstract/Free Full Text]
    33. 33
  33. Roth, M. D., Gitlitz, B. J., Kiertscher, S. M., Park, A. N., Mendenhall, M., Moldawer, N., Figlin, R. A. (2000) Granulocyte macrophage colony-stimulating factor and interleukin 4 enhance the number and antigen-presenting activity of circulating CD14+ and CD83+ cells in cancer patients Cancer Res. 60,1934-1941[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 J. Virol.Home page
I. Madisch, S. Hofmayer, C. Moritz, A. Grintzalis, J. Hainmueller, P. Pring-Akerblom, and A. Heim
Phylogenetic Analysis and Structural Predictions of Human Adenovirus Penton Proteins as a Basis for Tissue-Specific Adenovirus Vector Design
J. Virol., August 1, 2007; 81(15): 8270 - 8281.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
jlb.1105694v1
79/6/1271    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
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
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Harui, A.
Right arrow Articles by Basak, S. K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Harui, A.
Right arrow Articles by Basak, S. K.