www.elsevier.com/locate/yviro

Virology 324 (2004) 297–310

Despite differences between dendritic cells and Langerhans cells in

the mechanism of papillomavirus-like particle antigen uptake,

both cells cross-prime T cells

Mengyong Yan,a Judy Peng,a Ibtissam A. Jabbar,a Xiaosong Liu,a Luis Filgueira,b

Ian H. Frazer,a and Ranjeny Thomasa,*

aCentre for Immunology and Cancer Research, Princess Alexandra Hospital, University of Queensland, Brisbane, QLD 4102, Australiab Institute of Anatomy, Zurich, Switzerland

Received 10 November 2003; returned to author for revision 11 December 2003; accepted 24 March 2004

Available online 14 May 2004

Abstract

As human papillomavirus-like particles (HPV-VLP) represent a promising vaccine delivery vehicle, delineation of the interaction of VLP

with professional APC should improve vaccine development. Differences in the capacity of VLP to signal dendritic cells (DC) and Langerhans

cells (LC) have been demonstrated, and evidence has been presented for both clathrin-coated pits and proteoglycans (PG) in the uptake

pathway of VLP into epithelial cells. Therefore, we compared HPV-VLP uptake mechanisms in human monocyte-derived DC and LC, and

their ability to cross-present HPV VLP-associated antigen in the MHC class I pathway. DC and LC each took up virus-like particles (VLP).

DC uptake of and signalling by VLP was inhibited by amiloride or cytochalasin D (CCD), but not by filipin treatment, and was blocked by

several sulfated and non-sulfated polysaccharides and anti-CD16. In contrast, LC uptake was inhibited only by filipin, and VLP in LC were

associated with caveolin, langerin, and CD1a. These data suggest fundamentally different routes of VLP uptake by DC and LC. Despite these

differences, VLP taken up by DC and LC were each able to prime naive CD8+ T cells and induce cytolytic effector T cells in vitro.

D 2004 Elsevier Inc. All rights reserved.

Keywords: Dendritic cells; Human papillomavirus like particle; T cells

Introduction monocyte-derived DC but not LC generated from peripheral

Dendritic cells (DC) and Langerhans cells (LC) play a

crucial role in priming MHC class I restricted T cell

responses as they are the only APCs that are capable of

stimulating resting, naive T lymphocytes and hence initiat-

ing CTL immune responses in vivo. Papillomavirus virus

like particles (PV-VLPs) are the basis of vaccines to prevent

PV infection and are immunogenic in humans with or

without adjuvant, suggesting that they have innate capacity

to activate professional APCs. As PV-VLPs represent a

promising vaccine delivery vehicle, delineation of the inter-

action of VLP with professional APC should improve

vaccine development. Previously, human papillomavirus

like particles (HPV-VLPs) have been shown to activate

0042-6822/$ - see front matter D 2004 Elsevier Inc. All rights reserved.

doi:10.1016/j.virol.2004.03.045

* Corresponding author. Fax: +61-7-3240-5946.

E-mail address: rthomas@cicr.uq.edu.au (R. Thomas).

blood in the absence of co-stimulatory CD40 signalling

(Fausch et al., 2002, 2003). It has been suggested that lack

of presentation of HPV-VLP-associated epitopes by LC in

the absence of CD40 signalling might define an immuno-

logical escape mechanism used by HPV during natural HPV

infection (Fausch et al., 2002, 2003).

Clathrin-mediated endocytosis is the main pathway for

internalization of extracellular material in most cell types.

DCs ingest antigens by phagocytosis, macropinocytosis, and

receptor-mediated endocytosis. Macropinocytosis samples

antigen in the fluid phase and enables antigens to be endo-

cytosed independently of the requirement for specific recep-

tors (Norbury et al., 1997; Sallusto et al., 1995). Clathrin-

independent pathways, especially endocytosis via caveolae,

may also be involved in the internalization of membrane

components (glycosphingolipids and glycosylphosphatidyli-

nositol-anchored proteins), extracellular ligands such as folic

M. Yan et al. / Virology 324 (2004) 297–310298

acid, bacterial toxins, and several non-enveloped viruses

including Simian virus 40 (SV40), polyoma virus, echovirus

1, respiratory syncytial virus, and HIV (Gilbert et al., 2003;

Llano et al., 2002; Marjomaki et al., 2002; Werling et al.,

1999). For example, SV40 uses MHC class I as a receptor

and is internalized by caveolae. The uptake of these virus

particles by caveolae could trigger signal transduction in

endothelial cells, and viral particles were transported sub-

sequently to the endoplasmic reticulum, thus bypassing

other organelles (Norkin et al., 2002; Pelkmans et al.,

2002). HPV VLP uptake into human DC and the murine

C127 cell line has been demonstrated to occur through a

clathrin-dependent pathway. Furthermore, interaction with

heparan sulfate (HS) as an initial step in virus binding to

host cells has been described for entry of various viruses,

including HPV, into various cells (Joyce et al., 1999).

Heparin has been shown to inhibit binding of VLP to

epithelial cells through altered charge interactions with

glycosaminoglycan receptors (Joyce et al., 1999). Given

the previously demonstrated differences in the capacity of

VLP to signal DC and LC, and evidence for both clathrin

coated pits and HS in the uptake pathway of VLP into

epithelial cells, we compared HPV-VLP uptake mecha-

nisms in human monocyte-derived DC and LC, and their

ability to cross-present HPV VLP-associated antigen in the

MHC class I pathway. While VLP uptake mechanisms

differ in DC and LC, we show that both APC are able to

cross-present VLP-associated antigen to CD8+ T cells.

Fig. 1. Uptake of VLP by DC and LC. LCs or DCs were pulsed for 1 h at

37 jC with HPV 6bL1 VLP, either untreated or labeled with CFDA. To

visualize unlabeled VLP, permeabilized cells were incubated with HPV

6bL1 mAb (B10.5) or control antibody, and then with biotinylated rabbit

anti-mouse antibody and streptavidin-FITC. Fluorescence was (A) detected

by flow cytometry and (B) localized by confocal microscopy. (C) The

uptake of VLP by DC and LC generated for varying periods of time in the

presence of GM-CSF, IL-4 F TGF-h1 was investigated following

incubation as above with VLP-CFDA and flow cytometry. Mean DC or

LC fluorescence intensity (MFI) is shown for each time point. Error bars

represent SEM.

Results

Intracellular HPV 6bL1 VLP uptake by DC and LC

Binding of particulate antigen to receptors on DC or LC

may initiate signal transduction, resulting in cellular activa-

tion. We wished to study the interaction of virus particles

with DC and LC, using HPV VLP, which have been shown

to bind to the surface of both DC and LC, and with similar

efficiency (Fausch et al., 2002). We first compared the

intracellular location of VLP following uptake by mono-

cyte-derived DC (DC) and by monocyte-derived DC differ-

entiated in the presence of TGF-h1 (LC). Uptake and

location were established by flow cytometry and confocal

microscopy using CFDA-conjugated VLP or mAb B10.5-

specific for conformational epitopes on the VLPs (Chris-

tensen et al., 1994). Using either location technique, both

DC and LC took up VLP (Fig. 1). To assess the extent to

which VLP uptake was dependent on the duration of

monocyte differentiation to APC, DC and LC were exposed

to CFDA-VLP after 12–180 h of monocyte culture in the

presence of cytokines, and the intensity of intracellular

CFDA-conjugated VLP was assessed by flow cytometry 1

h later. The capacity for VLP uptake gradually increased

with increasing APC maturity, reaching a peak at 84 h (Fig.

1C), and decreased after 108 h.

The means of uptake of 6bL1 VLP differs between DC

and LC

Major antigen uptake pathways used by APC include

macropinocytosis and phagocytosis, involving receptor-

mediated uptake via clathrin-coated pits, or caveolae. Cyto-

M. Yan et al. / Virology 324 (2004) 297–310 299

chalasin D (CCD) blocks phagocytosis and amiloride blocks

macropinocytosis. Pretreatment of DC with either CCD or

amiloride blocked uptake of VLP. In contrast, pretreatment

of LC with CCD or amiloride had no significant effect on

uptake (Fig. 2). DC and LC were both sensitive to amiloride-

mediated inhibition of macropinocytosis, as amiloride

blocked uptake of dextran-FITC, taken up by macropinocy-

tosis and mannose receptor-mediated processes, by both DC

and LC (Fig. 2) (Sallusto et al., 1995). However, DC, but not

LC, were sensitive to the effects of CCD when taking up

dextran-FITC, most likely because CCD blocks the initial

steps of endocytosis but does not affect the recycling endo-

some pathway, active in LC (Salamero et al., 2001). Taken

together, these data support the hypothesis that macropino-

cytosis and phagocytosis mediate uptake of HPV 6bL1 VLP

by DC, but not by LC. Moreover, the lack of inhibition by

CCD suggests involvement of the recycling endosome

pathway in VLP uptake in LC.

The processing pathway of HPV 6bL1 in LC includes

langerin+ compartments

To pursue the hypothesis that VLPs traffic to the recy-

cling endosome pathway after uptake by LC, we examined

the collocation of VLP with CD1a and with langerin. CD1a

is an early endosome marker that does not traffic to late

endosomes and is part of the recycling endosome pathway,

and langerin is also shown to be associated with recycling

endosomes and Birbeck granules. All LC prepared from PB

Fig. 2. HPV 6bL1 VLP uptake by DC and LC is different. (A) DC or LC were pre

10 Ag/ml) for 30 min 37 jC before exposure to CDFA-VLPs (A) or dextran-FITC c

cells from three donors.

monocytes expressed CD1a. Although the intensity of

langerin expression varied from low to high, approximately

95% of LC and only rare DCs expressed langerin (Fig. 3A).

Co-localization of CD1a and langerin with VLP was ob-

served. Both langerinhi and langerinlo LCs took up VLP.

Confocal microscopy indicated that VLPs were co-localized

to the same compartment as CD1a and langerin, in langerin+

LC (Figs. 3B and C).

Caveolae are associated with the recycling endosomal

compartment. To further assess the association of caveolae

with VLP uptake, expression of caveolin-1 by DC and LC

was first examined. By immunofluorescence microscopy,

caveolin-1 was identified in a vesicular pattern within DC

and LC, and within permeabilized cells by flow cytometry

(Fig. 4A and data not shown). Although VLPs were

predominantly distributed within endosomes, some co-lo-

calized with caveolin-1 in DC and LC (Fig. 4A). The effect

of filipin, which destroys caveolar structures, on VLP

uptake was therefore examined. Filipin had no significant

inhibitory effect on HPV 6bL1 VLP uptake by DC but

did block uptake of the positive control cholera toxin-

Alexa Red by DC and LC, indicating that caveolae were

functional (Figs. 4B and C) (Nichols, 2002). In contrast,

VLP uptake was significantly blocked by filipin pretreat-

ment of LC, confirming a role for caveolin-mediated

uptake of VLP by LC, but not DC. Lack of involvement

of caveolae in VLP uptake by DC resembles the

response of C127 cells to bovine papillomavirus particles

(Day et al., 2003).

incubated with either amiloride (5 or 20 mM) or cytochalasin D (CCD, 2 or

ontrol (B). VLP uptake is shown as mean fluorescence intensityF SEM for

Fig. 3. Co-localization of 6bL1-CFDA, langerin, and CD1a within LC. LCs were incubated with 6bL1 VLP-CFDA for 1 h at 37 jC. After fixation and

permeabilization, cells were (A) incubated with mouse antihuman langerin, then biotinylated anti-mouse Ig and streptavidin-PE, and subjected to flow

cytometry. Cells stained with control mAb alone are shown in the top panel. DCs were stained with langerin (solid line) or control (dotted line) mAb only. (B)

Permeabilized cells were stained with mouse antihuman langerin mAb, then biotinylated anti-mouse Ig and streptavidin-Texas Red for confocal microscopy.

(C) LCs were exposed to HPV 6bL1 VLP-CFDA for 30 min, washed, and held for 1 h at 37 jC. Cells were fixed, permeabilized, and incubated with anti-CD1a

lyzed by confocal microscopy (C).

M. Yan et al. / Virology 324 (2004) 297–310300

Polysaccharides are associated with HPV 6bL1 VLP uptake

by DC but not by LC

Surface heparan sulfate is required for the uptake of

HPV 16, HPV 11, and HPV 6b VLPs by a range of cell

types (Giroglou et al., 2001; Joyce et al., 1999). To study

followed by biotinylated anti-mouse Ig and streptavidin-Texas-Red, and ana

the effect of heparin on HPV 6bL1 VLP entry to LC and

DC, VLPs were mixed with various concentrations of the

sulfated polysaccharides high molecular weight (HMW)

heparin and fucoidin, or a non-sulfated polysaccharide,

mannan, before incubation with DC or LC. Both heparin

and fucoidin could block more than 70% of VLP uptake by

Fig. 4. HPVVLP associate with caveolae. (A) DC and LCwere exposed to HPV 6bL1VLP, fixed and permeabilized, incubated with rabbit antihuman caveolin-1

polyclonal antibody, biotinylated anti-rabbit Ig, and streptavidin-Texas Red, then with HPV 6bL1 mAb (B10.5), and finally with biotinylated rabbit anti-mouse

antibody and streptavidin-FITC, then analyzed by immunofluorescence microscopy. Original magnification is �400. (B) DC and LC were preincubated with

various concentrations of filipin for 15 min, then exposed to CDFA-VLP. VLP uptake was assessed by flow cytometry. Data show the meanF SEM of the percent

inhibition by filipin of uptake of VLP by three donors. (C) DC and LCwere preincubated with either 10 Ag/ml filipin or no inhibitor and then with 0.5 Ag/ml Alexa

Red-labeled cholera toxin for 30 min. Red fluorescence of control cells treated with filipin alone (purple histogram), cells treated with cholera toxin alone (dotted

line), and cells treated with filipin followed by cholera toxin (solid green line) is shown for DC and LC, with percent inhibition indicated on each plot.

M. Yan et al. / Virology 324 (2004) 297–310 301

DC, but had no effect on, or increased VLP uptake by LC.

At higher concentrations, mannan blocked about 40% of

HPV 6bL1 VLP uptake by DC but did not block LC uptake.

Titration of heparin showed that 1000 U was sufficient to

block uptake of VLP by DC, but had little or no effect on

uptake by LC in the presence of TGF-h1 (Figs. 5A and C).

The data indicate that some sulfated polysaccharides are

sufficient to prevent uptake of HPV 6bL1 VLP by DC, and

there may be an additional contribution by non-sulfated

polysaccharides. In contrast, while a proportion of DC and

LC both expressed cell surface heparan sulfate as detected

by anti-syndecan-1 mAb (Fig. 5B), neither heparan sulfate

nor heparinase, which cleaves heparan sulfate (HS) from the

cell surface, affected the uptake of VLP by DC or LC (Fig.

5A). Thus, the mechanism of VLP uptake by DC appears to

be distinct from the cell surface HS-dependent mechanism

utilized by epithelial cells (Joyce et al., 1999), as uptake of

VLP by DC is not blocked by heparinase, and LC uptake of

VLP is not blocked by polysaccharides in the presence

of TGF-h1.

Fig. 5. Polysaccharides are associated with HPV 6bL1 VLP uptake by DC, but not by LC. (A) DCs or LCs were pretreated with heparinase I (3 U/ml) for 30

min at 37 jC and then exposed to VLPs. Alternatively, HMW heparin (1000 or 2500 U) or heparin sulfate (HS, 5 Ag/ml) was mixed with VLP, and DC or LC

exposed to the treated VLP. Data represent the mean F SEM percent inhibition of VLP uptake in the presence of the shown inhibitor, for LC or DC from three

donors. (B) DCs or LCs were stained with rat antihuman Syndecan-1 (solid line) or isotype control mAb (dotted line), followed by biotinylated goat anti-rat Ig,

and then streptavidin-FITC. (C) Inhibition of VLP uptake by sulfated or non-sulfated polysaccharides. DCs or LCs were incubated with different concentrations

of heparin, fucoidin, or mannan for 15 min at 37 jC before addition of HPV 6bL1 VLP-CFDA. After 1-h incubation, cells were washed and analyzed by flow

cytometry. Data represent the mean F SEM percent inhibition of VLP uptake in the presence of the shown inhibitor, for LC or DC from three donors.

M. Yan et al. / Virology 324 (2004) 297–310302

CD16, mannose receptor, and DC-HIL are associated with

uptake of 6bL1 VLP by DC but not by LC

A number of antigen recognition molecules are defined

on the surface of APC. High expression of MMR is found

on DC and lower expression on LC (Turville et al., 2002).

CD16 is expressed by some DC, but not by LC, and ligation

of CD16 may be associated with signalling of NFnB(Pengal et al., 2003). DC-associated, HS proteoglycan-

dependent integrin ligand (DC-HIL) (Shikano et al.,

2001b) is expressed at low levels intracellularly by DC,

but not by LC (Fig. 6B). Because the mechanism of VLP

uptake by DC appears to be distinct from cell surface HS,

we attempted to define any role for these receptors in the

uptake of VLP by DC. Anti-CD16 mAb and anti-MMR

blocked about 20% and 40% of VLP uptake by DC,

respectively, whereas antibody to the control membrane

antigen a-6 integrin or control Ig had no consistent effect.

In contrast, anti-MMR mAb blocked uptake of HPV 6bL1

VLP by LC to a lesser extent, about 20%, commensurate

with the lower levels of MMR expression by LC. To

determine whether DC-HIL plays a role in the uptake of

HPV 6bL1 VLP by DC or LC, a DC-HIL-Fc fusion protein

was used to block the uptake of VLP. Uptake of VLP by

DC, but not LC, was partially blocked (50%) by DC-HIL,

but not by a control Fc fusion protein (Fig. 6A). Apparent

stimulation of VLP uptake in DC by control-Fc was not

consistent between experiments. These data together sug-

gest that the polysaccharide receptors MMR and DC-HIL

and, albeit modestly, the FcR CD16 contribute to VLP

uptake by DC. In contrast, only the MMR polysaccharide

receptor contributes to VLP uptake by LC.

Heparin blocks RelB translocation and enhanced APC

function of DC after exposure to 6bL1 VLP

DCs, but not LCs, have been shown to be activated

after uptake of HPV16-VLP, leading to upregulation of co-

stimulatory molecules and cytokine production (Fausch et

al., 2002; Lenz et al., 2001). DC activation is associated

with RelB nuclear translocation and DNA binding activity

(O’Sullivan and Thomas, 2002; Thompson et al., 2002).

To determine whether blocking VLP uptake by DC via

proteoglycan (PG) receptor-mediated phagocytosis would

block DC activation, DCs were pretreated with heparin or

with CCD before incubation with VLP. Cell surface

activation markers and binding of DNA to RelB were

assessed. As neither cell surface marker upregulation nor

Fig. 6. Uptake of VLP by candidate HPV receptors on DC and LC. (A) DC and LC were incubated with potential inhibitors of VLP cellular receptors,

including antihuman CD16 mAb, antihuman-mannose receptor (MMR) mAb, anti-a-6 integrin, and DC-HIL-Fc fusion protein, or with appropriate control

antibodies (mIg, control Fc), washed, and pulsed with HPV 6bL1 VLP-CFDA. After 1-h incubation, cells were washed and analyzed by flow cytometry.

Data represent mean inhibition of VLP uptake by DC or LC in the presence of the shown inhibitor. (B) Expression of CD16, MMR, and DC-HIL by DC and

LC. DC or LC were incubated with mouse isotype control Ig (filled histogram), antihuman CD16, or antihuman MMR mAb (open histogram) and then

analyzed by flow cytometry. For the detection of DC-HIL, DCs or LCs were fixed and permeabilized and incubated with control-Fc (filled histogram) or

DC-HIL-Fc (open histogram). Expression was analyzed in each case by flow cytometry. Representative of three experiments (*P < 0.05 compared with

control-Fc).

M. Yan et al. / Virology 324 (2004) 297–310 303

RelB DNA binding was induced by VLP uptake in LC,

the data presented here are for DC only. Heparin blocked

the enhanced expression of CD86 and CD83 induced by

VLP (Fig. 7A). Furthermore, heparin and CCD, but not

anti-CD16—consistent with its modest capacity to block

VLP uptake—blocked RelB DNA binding and cellular

activation (Fig. 7B). The data are consistent with the

hypothesis that a proteoglycan receptor-mediated phago-

cytic pathway conveys the signalling by VLP to DC for

RelB activation.

Fig. 7. Inhibition of receptor-mediated phagocytosis of VLP by DC inhibits DC function. (A) DCs or LCs were pulsed with VLP, with or without heparin, for

24 h at 37 jC, and expression of HLA-DR, CD83, and CD86 was determined by flow cytometry. *P < 0.05 compared with control DC. (B) DCs were

preincubated with inhibitors of VLP binding as shown, washed, and then exposed to HPV 6bL1 VLP in the presence or absence of 1000 U/ml HMW heparin.

RelB-DNA binding, as a measure of DC activation, was detected in nuclear extracts by ELISA. Data represent the mean F SEM of duplicate analyses. *P <

0.05 compared to control DC.

M. Yan et al. / Virology 324 (2004) 297–310304

Cross-priming of VLP antigen by DC and LC

The difference between DC and LC in the uptake and

traffic of VLP and in cellular activation in response to VLP

might indicate differences in their capacity for cross-pre-

sentation of the VLP antigen. Previously in vitro-generated

DC and LC have been demonstrated to have equivalent

capacity for T cell priming (Caux et al., 1997; Geissmann et

al., 1999). Therefore, we examined the capacity of DC and

LC to cross-prime naive CD8+ T cells to a VLP-associated

MHC class I-restricted antigenic epitope and to cross-

present the VLP antigen to primed CD8+ T cell lines. We

utilized VLP chimeric for the HLA B8-restricted FLR

peptide of EBNA1 (Burrows et al., 1992a, 1992b). Naive

CD8+ T cells from EBV seronegative donors were primed in

vitro by DC or LC, previously loaded with chimeric 6bL1-

FLR VLP. After restimulation in vitro, the function and

specificity of T cell lines cross-primed by VLP-loaded DC

or LC were evaluated. T cells primed by HPV 6bL1-FLR

VLP-loaded DC produced IFN-g in response to FLR

peptide, but not to irrelevant peptide GB906. T cells primed

by HPV 6bL1-FLR VLP-loaded LC also produced IFN-g,

but required a higher concentration of peptide for activation

(Figs. 8A and B). DC-primed CD8+ T cells demonstrated

more nonspecific lytic activity than LC-primed CD8+ T

cells. CTL activity was further assayed using the HLA B8+

B lymphoma cell line BL30A pulsed with either FLR

peptide or irrelevant peptide GB906 as targets. BL30A

incubated with FLR peptide, but not GF906, was lysed by

T cell lines initially cross-primed by either DC or LC loaded

with HPV 6bL1-FLR VLP (Figs. 8C and D). In general,

FLR-specific killing relative to the background nonspecific

lytic activity of both T cell lines was low. However, both DC

and LC loaded with HPV VLP-FLR could effectively

present VLP-encoded MHC class I epitope to CD8+-resting

T cells for priming.

Discussion

HPV-VLPs are candidate vaccines both for the prevention

of HPV infection (Suzich et al., 1995) and as an immuno-

genic delivery system for antigen incorporated into the PV-

VLP (Kaufmann et al., 2001; Peng et al., 1998; Rudolf et al.,

Fig. 8. Cross-priming of CD8+ T cells by DC and LC loaded with chimeric VLP. HLA-B8+ CD8+ T cells were exposed over 3 weeks in vitro to DC or LC

loaded with chimeric HPV 6bL1 expressing the FLR B8-restricted CTL epitope. T cell lines generated from loaded DC (A) or LC (B) were tested by ELISPOT

for IFN-g production in response to FLR peptide or a control peptide G8906. T cell lines generated from loaded DC (C) or LC (D) were similarly tested for

CTL activity using as targets HLA B8+ B cell line BL30A loaded with either FLR or control peptide. Representative of two experiments with error bars

representing means F SEM.

M. Yan et al. / Virology 324 (2004) 297–310 305

1999). As delineation of the interaction of VLP with profes-

sional APC should improve vaccine development, we com-

pared the uptake mechanism of VLP by DC and LC. A

number of mechanisms have been established by which DC

can take up antigen. Macropinocytosis results in the sam-

pling of antigen in the fluid phase and enables antigens to be

endocytosed independently of the requirement for specific

receptors (Norbury et al., 1997; Sallusto et al., 1995). Uptake

via receptors that are concentrated in clathrin-coated pits

may be involved for some substances (Chuang et al., 1997).

By following the uptake of VLP-CFDA in both DC and LC,

we studied the intracellular traffic of VLP once it was taken

up by cells. Several differences in the handling of VLP by

DC and LC emerged. DC uptake of VLP was inhibited by

cytochalasin D and amiloride, indicating that uptake was

mediated by phagocytosis and the formation of clathrin-

coated pits, in keeping with previous observations using

HPV16-VLP (Fausch et al., 2003).

In contrast to DC, the uptake of VLP by LC was clathrin-

independent as cytochalasin D and amiloride had no effect

on uptake. In LC, caveolae appear to be involved, as the

uptake of HPV 6bL1 VLP was blocked in a large part by

filipin, which extracts cholesterol from caveolae, thereby

destabilizing raft architecture (Venkatesan et al., 2003).

While some co-localization of VLP and caveolin-1 oc-

curred, VLPs appear to translocate rapidly through caveolae

into the recycling endosome pathway. VLP co-localized

with CD1a+ early endosomes and langerin in LC. CD1a is

included in recycling vesicles that deliver internalized

proteins back to the cell surface rather than transporting

them more deeply into the late endocytic system (Sugita et

al., 1999). Langerin, a protein implicated in Birbeck granule

biogenesis, is also predominantly found in the endosomal

recycling compartment and in Birbeck granules. It has been

shown that Langerin traffics between the plasma membrane,

the early or sorting endosomes and the ER but not to late

endosomes, and is associated with the CD1a recycling

compartment (Mc Dermott et al., 2002). Thus, VLP may

predominantly traffic in the endosomal recycling pathway in

LC. While the mechanism of MHC loading with the VLP

antigen in this compartment is not yet known, previous

studies demonstrate that influenza antigen loading of na-

scent class II may occur directly from the early endosome

under acidic conditions (Sinnathamby and Eisenlohr, 2003).

These results differ from those previously published—

perhaps due to differences in the method of blocking with

filipin or differences in uptake between HPV16L1 and HPV

6bL1 VLP (Fausch et al., 2003).

Animal cell membranes abundantly express proteoglycans

(PGs), which are made up of a protein core with one or more

M. Yan et al. / Virology 324 (2004) 297–310306

covalently attached GAG chains that bind several different

protein ligands (Rostand and Esko, 1997). Heparan sulfate

(HS) and other GAGs are included in the proteoglycans

(Rostand and Esko, 1997; Stringer and Gallagher, 1997).

Viruses in their cellular entry process exploit this anatomical

feature of the cell membrane. Interaction with HS as an initial

step in virus binding to host cells has been described for entry

of various viruses into epithelial cells during either single or

multistep virus entry processes (Liu and Thorp, 2002). DCs

have been shown to be activated after uptake of VLP, leading

to upregulation of co-stimulatory molecules and cytokine

production (Fausch et al., 2002; Lenz et al., 2001). DC

activation is associated with RelB nuclear translocation and

DNA binding activity (O’Sullivan and Thomas, 2002;

Thompson et al., 2002). In the current studies, RelB DNA

binding was inhibited by CCD and heparin, indicating that

NFnB activation by VLP is triggered through cell surface

GAG signalling and receptor-mediated phagocytosis. Hepa-

rin inhibits binding of VLP through altered charge inter-

actions with GAG receptors (Joyce et al., 1999). Such

interactions might also explain the variable increase in VLP

uptake by LC in the presence of heparin. Unexpectedly, the

DC GAG receptor for HPV VLP appears not to be HS, as in

epithelial cells, as although a proportion of DC expressed

HS, heparinase had no effect on VLP uptake. We therefore

explored the characteristics of the DC receptor for VLP. The

highly sulfated polysaccharide fucoidin also inhibited HPV

VLP uptake by DC, and the non-sulfated polysaccharide

mannan had weaker inhibitory effects on the uptake of HPV

VLP. A recent study suggested that heparin-related binding

of DC to endothelial cells was mediated by a novel receptor,

DC-associated, HS PG-dependent integrin ligand (DC-HIL).

HS ligand binding to DC-HIL is also blocked by heparin and

fucoidin (Shikano et al., 2001a). Here we show that the

soluble fusion protein DC-HIL-Fc, which could bind to the

DC-HIL extracellular domain (ECD), also has an inhibitory

effect on the uptake of HPV VLP by DC. These data suggest

that DC-HIL is a candidate receptor for the initial binding

and uptake of VLP by DC. Both HS and hyaluronan have

been shown to activate the NFnB pathway in DC via TLR4

(Karin and Ben-Neriah, 2000; Lyakh et al., 2000). DC

activation through DC-HIL has not previously been exam-

ined (Shikano et al., 2001a).

FcgRIII (CD16) was expressed by only a proportion of

DC, and blockade of CD16 had modest effects on VLP

uptake byDC. Although ligation of Fcg receptors may induce

activation of NFnB, the current expression and blocking

studies indicate that this is not the primary receptor mediating

VLP RelB-dependent signals to monocyte-derived DC

(Alonso et al., 2000). Mannose receptors have been shown

to be involved in the binding of several viruses, as well as

yeast VLP to DC and in the activation of monocyte-derived

DC (Milone and Fitzgerald-Bocarsly, 1998; Tsunetsugu-

Yokota et al., 2003). Consistent with the effects of mannan,

anti-mannose receptor mAb partially blocked the uptake of

HPV 6bL1 VLP by DC and LC. The data raise the possibility

of co-receptor function for CD16 or integrin-recognition

RGD motifs displayed by GAGs or mannose receptors with

DC-HIL (Shikano et al., 2001a). A similar situation exists for

adenovirus, in which aVh5 serves as a co-receptor, particu-

larly for infectivity of adenovirus, along with the primary

attachment receptor HS (Summerford et al., 1999). It is not

yet elucidated which receptors are responsible for the initial

binding of VLP by LC. However, we have determined at least

that CD16, a-6 integrin, and DC-HIL are not responsible

receptors, but that the MMR is partially involved in the

uptake of VLP by LC.

Previously published data suggested that after taking up

chimeric HPV16 L1-E7, DCs, but not LCs, were able to

cross-prime E7 peptide-specific CD8+ T cells. We used the

same methodology to prime CD8+ T cells with DCs or LCs

that were pulsed with chimeric HPV 6bL1-FLR. While we

could confirm the previously published data, additional dose

titration of the FLR peptide indicated that T cell lines

generated in response to LC loaded with chimeric VLP

could also respond with specific IFN-g production in

response to higher doses of peptide. In contrast to previous

studies (Fausch et al., 2002, 2003), we found that CD8+ T

cells stimulated by either LC or DC loaded with chimeric

VLP could act as FLR-specific CTL. CTL analysis indicates

that LCs cross-prime VLP-associated epitopes to resting T

cells even if not previously signalled with CD154. The most

likely explanation for the difference in results with previous

studies is that CTL assays were carried out in the current

studies over a range of effector to target ratios and peptide

concentrations rather than a single ratio or concentration,

allowing for greater sensitivity of detection of the relative

capacity of DC and LC to cross-present VLP-associated

antigens. Thus, the current data suggest that, after uptake,

processing of VLP-associated antigen by LC might be less

efficient than by DC. Nevertheless, both APC cross-present

VLP-associated antigen in vitro. Confirmation of this ca-

pacity is required in vivo.

The current studies examined the uptake by and antigen

presentation by DC and LC of HPV-L1-VLP. From these

studies, it is possible to make inferences or novel hypotheses

regarding the mechanism of action of VLP vaccines, but not

natural HPVinfection. Our results support the hypothesis that

in vivo, LC and dermal DC are both likely to cross-present

VLP antigen in draining LN after contacting and taking up

VLP vaccines injected into skin. Simultaneous cross-priming

of VLP by both DC and LC may actually be synergistic, as

DCs activated by VLP are more likely to generate CD154+ T

helper cells, which are required for full LC activation

(Geissmann et al., 1999). While LCs were not deliberately

signalled through CD40 in vitro in the current studies, it is

possible that a small proportion of CD4+ T cells in the

cultures may have been sufficient to provide help through

CD154 ligation of LC cross-priming chimeric VLP. Our data

provide a testable model for the mechanism of CTL induction

by PV-VLP in vivo. The ability of LC to take up a large

amount of VLP but not to become activated in the presence of

M. Yan et al. / Virology 324 (2004) 297–310 307

TGF-h1 supports the idea that LCs may remain immature

even after uptake of VLP at epithelial sites in vivo. However,

after uptake of VLP in the skin, LCs may migrate to draining

lymph nodes where levels of TGF-h1 are reduced. This

migration could be signalled by a6-integrin following VLP

uptake into LC (Evander et al., 1997; Price et al., 1997).

Therefore, a drop in the TGF-h levels upon migration out of

the skin, coupled with CD40 signalling by lymph node CD4+

T helper cells, would allow LC to cross-present VLP-

encoded antigens to CTL precursors in the draining lymph

node. In contrast, dermal DCs appear more likely to activate

CD4+ T helper cells in response to VLP as well as to drive

local cytokine and chemokine production, leading to a

dermal inflammatory response (Lenz et al., 1993). In this

regard, DCs developing in the absence of TGF-hmay play an

extremely important role in the distinction between cross-

tolerance and cross-priming as, in the absence of CD154

derived from CD4+ helper T cells, cross-tolerance might

result (Albert et al., 2001; Sotomayor et al., 1999).

Materials and methods

Generation of VLP

HPV6b-L1s were generated as described previously

(Kirnbauer et al., 1992; Park et al., 1993). The particulate

VLP band was collected after continuous cesium chloride

density gradient centrifugation and dialyzed overnight

against several changes of PBS at 4 jC. Total protein and

L1 content was estimated by 10% SDS-PAGE and anti-L1

immunoblotting and ELISA, and the presence of intact

virus particles was confirmed by electron microscopy

(Zhou et al., 1991). For the production of chimeric VLP-

FLR, a polynucleotide encoding a well-defined HLA-B8-

restricted, EBNA-3 CTL epitope FLRGRAYGL (called

FLR) (Burrows et al., 1992a) was added 3V-terminus of

the HPV 6bL1 gene by PCR using the primers (FLR

forward BamHI) 5V cgc gga tcc atg tgg cgg cct agc gac

agc 3V and (FLR reverse EcoRI) 5V ccg gaa ttc tta caa gtc

tag ccc ata cgc acg acc ccg gag aaa gcc ggc tcc cct ata tcc

act ttg taa caa 3V. PCR products were purified and ligated at

equimolar amounts to vector with T4-DNA-Ligase (BD

PharMingen, San Diego, CA). The resulting PCR product

was purified and cloned into baculovirus transfer vector

PVL1393 (BD PharMingen) via BamHI and EcoRI sites

and transformed into Escherichia coli DH a-5 cells (Life

Technology, San Diego, CA), sequenced, and transfected

into Sf9 insect cells (ATCC, Rockville MD), as previously

described (Peng et al., 1998). The presence of chimeric L1

protein in purified VLPs was confirmed by Western blot

analysis with anti-6bL1 Ab MC15 (Kulski et al., 1998), and

the structure of VLP was confirmed by electron microscopy

(EM). Endotoxin levels in VLP preparations were measured

in a Limulus-assay using a QCL-1000 Endotoxin kit

(Pierce, BA oud-Bejerland, Holland) and were in each case

<10 U/ml. HPV 6bL1 VLPs were labeled with 5- and 6-

carboxyfluorescein diacetate, succinimidyl ester (CFDA-SE

Cell Tracer Kit, Molecular Probes, Eugene, OR) by mixing

89 Al PBS (pH 8.5), 10 Ag VLP (1 mg/ml), and 10 mM

CFDA-SE. Following incubation in the dark for 4 h at RT,

VLPs were subjected to dialysis against PBS overnight

(McMillan et al., 1999).

Reagents and cell culture

Human cell culture was carried out in RPMI 1640

supplemented with 2 mM L-glutamine (Trace, Biosciences,

NSW, Australia), 100 U/ml penicillin, 100 Ag/ml strepto-

mycin, 20 mM HEPES (Gibco), 1 mM sodium pyruvate,

and 10% FBS (CSL).

Generation of APC

Peripheral blood mononuclear cells (PBMCs) depleted of

T, B, and NK cells by magnetic bead separation (MACS,

Miltenyi Biotec, Auburn, CA) were used to generate mono-

cyte-derived DC and LC in vitro as previously described

(Cavanagh et al., 1998). Enriched monocyte precursors were

cultured for 1–8 days in medium supplemented with 800 U/

ml rhGM-CSF and 400 U/ml rhIL-4 (both from Schering-

Plough, Sydney, Australia). For LC, preparation was iden-

tical except that 10 ng/ml rhTGF-h1 (Life Technology) was

added every second day.

Staining of DC and LC for flow cytometric analysis

Freshly isolated DCs and LCs were incubated with

optimal concentrations of specific mAb or mouse isotype

control antibody (DAKO, Carpinteria, CA) on ice for 30 min,

washed twice, then incubated with biotinylated anti-mouse Ig

for 30 min on ice, washed twice, then finally incubated with

streptavidin-FITC or streptavidin-PE (DAKO). For intracel-

lular staining, the cells were fixed and permeabilized by

Cytofix/Cytoperm (PharMingen) before adding mAb. Anti-

6bL1 mAb (anti-HPV 6 series, L12, B10.5, N8, and C6, and

anti-HPV 11 series, D2, H5, F6, and E38 were gifts from Dr.

Neil Christensen. Milton S. Hershey Medical Center, PA)

(Christensen et al., 1994), Mouse antihuman langerin (Immu-

nex, Marseille, France), and rabbit antihuman caveolin poly-

clonal antibody (BD PharMingen, San Jose, CA) were used.

After staining, all cells were fixed with 1% paraformaldehyde

(Sigma, St. Louis, MO) and analyzed on a FACScalibur (BD

PharMingen) using a single argon laser. All cells were fixed

with 1% paraformaldehyde (Sigma) before analysis.

Staining of DC and LC for confocal and

immunofluorescence microscopy

For confocal microscopy of DC or LC after uptake of

VLP/CFDA, cells were pulsed with VLP for 1 h at 37 jC,then washed and cytospun onto Superfrost Plus glass slides

M. Yan et al. / Virology 3308

(Electron Microscopy Sciences, Fort Washington, PA),

incubated at RT for 45 min, and finally fixed in 4%

paraformaldehyde and mounted in fluorescence mounting

medium (DAKO, Glostrup Denmark). For double staining,

freshly isolated DC and LC were pulsed with 1 Ag/ml 6bL1-

CFDAVLP as before. At different times, the cells were put

into Lab-Tek Chamber Slides (Nunc, Rochester, NY) coated

with 0.01% poly-L-Lysine (Sigma) and incubated for 10 min

at room temperature. After fixation and permeabilization

with Cytofix/Cytoperm (PharMingen), the cells were incu-

bated with optimal concentrations of anti-CD1a or anti-

Langerin or mouse isotype control antibody on ice for 30

min, washed twice, then incubated with biotinylated anti-

mouse Ig (DAKO, Carpinteria, CA) for 30 min on ice,

washed twice, and finally incubated with streptavidin-Tex-

as-Red (Amersham, Piscataway, NJ). Cells were analyzed

by confocal microscopy (LSM, Zeiss, Jena, Germany). For

double staining with caveolin-1, DC and LC were exposed

to HPV 6bL1 VLP for 1 h, fixed and permeabilized,

incubated with rabbit antihuman caveolin-1 polyclonal

antibody, biotinylated anti-rabbit Ig and streptavidin-Texas

Red, then with HPV 6bL1 mAb (B10.5), and finally with

biotinylated rabbit anti-mouse antibody and streptavidin-

FITC, then analyzed by immunofluorescence microscopy

and imaging software (Zeiss).

Uptake of VLP by DC and LC

105 DC or LC were incubated with 1 Ag/ml VLP-CFDA

for 1 h. LCs were incubated with VLP in the presence of

TGF-h1 (+TGF-h1). For blocking experiments, DC or LC

were pretreated with amiloride (Sigma), or cytochalasin D

(CCD) (Sigma) for 15–30 min, washed three times, and

then pulsed with VLP-CFDA for 1 h. For the heparin and

fucoidin blocking experiments, 1 Ag VLP-CFDAwas mixed

with heparin (David Bull Laboratories, Victoria, Australia)

or fucoidin (Sigma) before incubation with APC. For the

heparinase treatment, DC or LC were washed once with

PBS. Heparinase (forms I and II) (Sigma) was diluted in

digestion buffer (20 mM Tris–HCl, 50 mM NaCl, 4 mM

CaCl2, pH 7.5, containing 0.01% BSA) and added to cells at

a final concentration of 1 or 3 units/well at 37 jC for 1 h.

Control cells were treated with buffer without enzyme. Cells

were washed twice before incubation with VLP-CFDA. For

the antibody blocking, DC or LC were treated with 10 mg/

ml or anti-CD16 mAb (BD PharMingen), anti-mannose

receptor (CD206) mAb (BD PharMingen), or mouse isotype

control (DAKO). After washing with PBS, the cells were

incubated with 6bL1-CFDA for 10 min at 4 jC and then 1

h at 37 jC. For the detection of cell surface heparan sulfate

proteoglycan, DC or LC were stained with rat antihuman

Syndecan-1 or isotype control mAb, followed by biotiny-

lated goat anti-rat Ig and then streptavidin-FITC and ana-

lyzed by flow cytometry. In control experiments, APC were

incubated with dextran-FITC or cholera toxin Hexa-Red (a

gift from Dr. R. Parton, University of Queensland).

Preparation of DC-HIL-Fc Fusion protein and its blocking

of HPV VLP uptake by DC and LC

DC-HIL-Fc consists of extracellular domain (ECD) of

DC-HIL and the Fc region of human IgG1. The cDNA of

this protein was a gift from Dr. K Ariizumi (University of

Texas Southwestern Medical Centre, USA), which was

cloned into the pSecTagB vector (pSTB-DC-HIL-Fc) (Shi-

kano et al., 2001a). The vector and control vector for Fc

fusion protein TNF-a-Fc were transfected into COS-1 cells

by lipofectamine (Invitrogen, Carlsbad, CA) according to

the manufacturer’s instructions. After 96 h of transfection,

the culture supernatant was harvested and immunoglobulin

fusion proteins were purified by protein A-agarose beads

(Pierce, Rockford, IL) according to the manufacturer’s

recommendations. For the blocking experiments, various

concentrations of DC-HIL-Fc or TNF-a-Fc were incubated

with DC or LC for 10 min at room temperature. The cells

were washed and then incubated with HPV 6bL1-CFDA for

10 min at 4 jC and then 1 h at 37 jC, fixed and analyzed by

flow cytometry.

Priming CD8+ T cells with DC and LC after pulsing with

chimeric 6bL1-FLR

CD8+ T cells were isolated from HLA B8+ donors by

magnetic immunodepletion using CD4, HLA-DR, CD16,

CD19, and CD56 mAb (all from BD PharMingen) and a

magnetic cell separator (MACS, Miltenyi Biotec Inc.). DC

and LC were produced from plastic adherent cells as

described. 106 DC or LC were incubated with 10 Agchimeric HPV 6bL1-FLR VLP for 1 h at room temperature,

washed, and mixed with 20 � 106 autologous freshly

isolated CD8+ PB T cells. Cells were cultured in 24-well

plates (Costar, Cambridge, MA) at 2 � 106 cells/well in

complete medium for 7 days at 37 jC. Restimulations after

7 and 14 days used 0.5 � 106 DC or LC/well loaded with 10

Ag/ml chimeric VLP. DC and LC were washed with PBS,

irradiated 25 Gy, before addition to the T cell cultures. For

restimulations, the medium was supplemented with IL-2 at

50 U/ml at 2 and 4 days after restimulation. After 28 days,

effector cells were pooled and tested for IFN-g production

by ELISPOT and cytotoxic T cell assay.

Cytotoxic T cell assay

For VLP-specific CTL activity, DC or LC pulsed with

either VLP, chimeric VLP, FLR peptide, or G8906 (a gift

from Dr. Germain Fernando, Centre for Immunology and

Cancer Research, University of Queensland, Australia), or

BL30 B lymphoma cells (a gift from Dr. Scott Burrows,

QIMR, Australia) pulsed with FLR peptide, were labeled

with Na251CrO4 (Perkin-Elmer, Boston, MA), washed, and

resuspended in medium. Targets (2 � 104) were incubated

with varying numbers of effector T cells for 4 h at 37 jC. Thesupernatant was collected into Lumaplates, dried overnight,

24 (2004) 297–310

M. Yan et al. / Virology 324 (2004) 297–310 309

then counted using a gamma counter (Packard, Canberra,

Australia). Specific lysis was calculated as follows: (exper-

imental release � spontaneous release)/(maximum release �spontaneous release)� 100. The SEM for all CTL assays was

less than 10% of the plotted percent-specific lysis.

ELISPOT

MultiScreen-HA plates (Millipore, Bedford, MA) were

coated with rat anti-mouse IFN-g mAb at a concentration of

10 Ag/ml in PBS at 4 jC overnight. Coated plates were then

washed six times with PBS with 0.25% Tween 20 (PBST)

and then blocked with 200 Al PBS/5% FCS for at least 3 h at

37 jC. For CD8+ T cell lines derived after T cell stimulation

with DC or LC incubated with chimeric VLP, a total of 2.0�105 cells/well was incubated in the presence or absence of

FLR peptide or irrelevant peptide (G8906) for 40 h at 37 jC.After washing three times with PBS 0.05% Tween, biotiny-

lated antihuman IFN-g (Mabtech AB, Nacka, Sweden) was

added and incubated at room temperature for 2–4 h, followed

by streptavidin alkaline phosphatase (DAKO) for an addi-

tional 2 h. Dark spots were developed over a 30-min reaction

with 5-bromo-4-chloro-3-indolyl phosphate/nitroblue tetra-

zolium (BNIP/NBT) tablets (Sigma). The spots were counted

under a dissecting microscope using video imaging software

(Vision Dynamics, Herts, United Kingdom).

Protein extraction and NF-jB binding ELISA

Nuclear and cytoplasmic extracts were prepared as pre-

viously described (O’Sullivan and Thomas, 2002), and

protein estimations carried out using a Protein Assay kit

(Bio-Rad, Hercules, CA). RelB DNA binding was detected

by ELISA using a Mercury Transfactor p50 Kit (Clontech,

Palo Alto, CA) and anti-RelB (sc-226) from Santa Cruz

Biotech (Santa Cruz, CA), as described (O’Sullivan and

Thomas, 2002). Nuclear extracts (10 Ag) were bound to

wells coated with NF-nB consensus oligonucleotide then

incubated with anti-RelB, followed by anti-rabbit HRP-

conjugated Ig. Nuclear RelB was estimated by measuring

color development at 650 nm using a Multiskan plate reader

(Labsystems, Chicago, IL).

Statistical analysis

Paired statistical analysis was performed by using the

Student’s two-tailed t test.

References

Albert, M.L., Jegathesan, M., Darnell, R.B., 2001. Dendritic cell matura-

tion is required for the cross-tolerization of CD8+ T cells. Nat. Immu-

nol. 2 (11), 1010–1017.

Alonso, A., Bayon, Y., Renedo, M., Crespo, M.S., 2000. Stimulation of Fc

gamma R receptors induces monocyte chemoattractant protein-1 in the

human monocytic cell line THP-1 by a mechanism involving I kappa B-

alpha degradation and formation of p50/p65 NF-kappa B/Rel com-

plexes. Int. Immunol. 12 (4), 547–554.

Burrows, F.J., Watanabe, Y., Thorpe, P.E., 1992a. A murine model for

antibody-directed targeting of vascular endothelial cells in solid tumors.

Cancer Res. 52, 5954–5962 (November 1).

Burrows, S.R., Rodda, S.J., Suhrbier, A., Geysen, H.M., Moss, D.J., 1992b.

The specificity of recognition of a cytotoxic T lymphocyte epitope. Eur.

J. Immunol. 22 (1), 191–195.

Caux, C., Massacrier, C., Vanbervliet, B., Dubois, B., Durand, I., Cella, M.,

Lanzavecchia, A., Banchereau, J., 1997. CD34+ hematopoietic progeni-

tors from human cord blood differentiate along two independent den-

dritic cell pathways in response to granulocyte-macrophage colony-

stimulating factor plus tumor necrosis factor alpha: II. Functional analy-

sis. Blood 90 (4), 1458–1470.

Cavanagh, L.L., Saal, R.J., Grimmett, K.L., Thomas, R., 1998. Prolifera-

tion in monocyte-derived dendritic cell cultures is caused by progenitor

cells capable of myeloid differentiation. Blood 92 (5), 1598–1607.

Christensen, N.D., Kirnbauer, R., Schiller, J.T., Ghim, S.J., Schlegel, R.,

Jenson, A.B., Kreider, J.W., 1994. Human papillomavirus types 6 and

11 have antigenically distinct strongly immunogenic conformationally

dependent neutralizing epitopes. Virology 205 (1), 329–335.

Chuang, E., Alegre, M.L., Duckett, C.S., Noel, P.J., Vander Heiden, M.G.,

Thompson, C.B., 1997. Interaction of CTLA-4 with the clathrin-asso-

ciated protein AP50 results in ligand-independent endocytosis that limits

cell surface expression. J. Immunol. 159 (1), 144–151.

Day, P.M., Lowy, D.R., Schiller, J.T., 2003. Papillomaviruses infect cells

via a clathrin-dependent pathway. Virology 307 (1), 1–11.

Evander, M., Frazer, I.H., Payne, E., Qi, Y.M., Hengst, K., McMillan, N.A.,

1997. Identification of the alpha6 integrin as a candidate receptor for

papillomaviruses. J. Virol. 71 (3), 2449–2456.

Fausch, S.C., Da Silva, D.M., Rudolf, M.P., Kast, W.M., 2002. Human

papillomavirus virus-like particles do not activate Langerhans cells: a

possible immune escape mechanism used by human papillomaviruses.

J. Immunol. 169 (6), 3242–3249.

Fausch, S.C., Da Silva, D.M., Kast, W.M., 2003. Differential uptake and

cross-presentation of human papillomavirus virus-like particles by den-

dritic cells and Langerhans cells. Cancer Res. 63 (13), 3478–3482.

Geissmann, F., Revy, P., Regnault, A., Lepelletier, Y., Dy, M., Brousse, N.,

Amigorena, S., Hermine, O., Durandy, A., 1999. TGF-beta 1 prevents

the noncognate maturation of human dendritic Langerhans cells.

J. Immunol. 162 (8), 4567–4575.

Gilbert, J.M., Goldberg, I.G., Benjamin, T.L., 2003. Cell penetration and

trafficking of polyomavirus. J. Virol. 77 (4), 2615–2622.

Giroglou, T., Florin, L., Schafer, F., Streeck, R.E., Sapp, M., 2001. Human

papillomavirus infection requires cell surface heparan sulfate. J. Virol.

75 (3), 1565–1570.

Joyce, J.G., Tung, J.S., Przysiecki, C.T., Cook, J.C., Lehman, E.D., Sands,

J.A., Jansen, K.U., Keller, P.M., 1999. The L1 major capsid protein of

human papillomavirus type 11 recombinant virus-like particles interacts

with heparin and cell-surface glycosaminoglycans on human keratino-

cytes. J. Biol. Chem. 274 (9), 5810–5822.

Karin, M., Ben-Neriah, Y., 2000. Phosphorylation meets ubiquitination: the

control of NF-[kappa]B activity. Annu. Rev. Immunol. 18, 621–663.

Kaufmann, A.M., Nieland, J., Schinz, M., Nonn, M., Gabelsberger, J.,

Meissner, H., Muller, R.T., Jochmus, I., Gissmann, L., Schneider, A.,

Durst, M., 2001. HPV16 L1E7 chimeric virus-like particles induce

specific HLA-restricted T cells in humans after in vitro vaccination.

Int. J. Cancer 92 (2), 285–293.

Kirnbauer, R., Booy, F., Cheng, N., Lowy, D.R., Schiller, J.T., 1992. Pap-

illomavirus L1 major capsid protein self-assembles into virus-like par-

ticles that are highly immunogenic. Proc. Natl. Acad. Sci. U.S.A. 89

(24), 12180–12184.

Kulski, J.K., Sadleir, J.W., Kelsall, S.R., Cicchini, M.S., Shellam, G., Peng,

S.W., Qi, Y.M., Galloway, D.A., Zhou, J., Frazer, I.H., 1998. Type spe-

cific and genotype cross reactive B epitopes of the L1 protein of HPV16

defined by a panel of monoclonal antibodies. Virology 243 (2), 275–282.

M. Yan et al. / Virology 324 (2004) 297–310310

Lenz, A., Heine, M., Schuler, G., Romani, N., 1993. Human and murine

dermis contain dendritic cells—Isolation by means of a novel method

and phenotypical and functional characterization. J. Clin. Invest. 92 (6),

2587–2596.

Lenz, P., Day, P.M., Pang, Y.Y., Frye, S.A., Jensen, P.N., Lowy, D.R.,

Schiller, J.T., 2001. Papillomavirus-like particles induce acute activa-

tion of dendritic cells. J. Immunol. 166 (9), 5346–5355.

Liu, J., Thorp, S.C., 2002. Cell surface heparan sulfate and its roles in

assisting viral infections. Med. Res. Rev. 22 (1), 1–25.

Llano, M., Kelly, T., Vanegas, M., Peretz, M., Peterson, T.E., Simari, R.D.,

Poeschla, E.M., 2002. Blockade of human immunodeficiency virus type

1 expression by caveolin-1. J. Virol. 76 (18), 9152–9164.

Lyakh, L.A., Koski, G.K., Telford, W., Gress, R.E., Cohen, P.A., Rice,

N.R., 2000. Bacterial lipopolysaccharide, TNF-alpha, and calcium ion-

ophore under serum-free conditions promote rapid dendritic cell-like

differentiation in CD14+ monocytes through distinct pathways that

activate NK-kappa B. J. Immunol. 165 (7), 3647–3655.

Marjomaki, V., Pietiainen, V., Matilainen, H., Upla, P., Ivaska, J., Nissinen,

L., Reunanen, H., Huttunen, P., Hyypia, T., Heino, J., 2002. Internali-

zation of echovirus 1 in caveolae. J. Virol. 76 (4), 1856–1865.

Mc Dermott, R., Ziylan, U., Spehner, D., Bausinger, H., Lipsker, D.,

Mommaas, M., Cazenave, J.P., Raposo, G., Goud, B., de la Salle, H.,

Salamero, J., Hanau, D., 2002. Birbeck granules are subdomains of

endosomal recycling compartment in human epidermal Langerhans

cells, which form where Langerin accumulates. Mol. Biol. Cell 13

(1), 317–335.

McMillan, N.A., Payne, E., Frazer, I.H., Evander, M., 1999. Expression of

the alpha6 integrin confers papillomavirus binding upon receptor-nega-

tive B-cells. Virology 261 (2), 271–279.

Milone, M.C., Fitzgerald-Bocarsly, P., 1998. The mannose receptor medi-

ates induction of IFN-a in peripheral blood dendritic cells by enveloped

RNA and DNA viruses. J. Immunol. 161, 2391–2399.

Nichols, B.J., 2002. A distinct class of endosome mediates clathrin-indepen-

dent endocytosis to the Golgi complex. Nat. Cell Biol. 4 (5), 374–378.

Norbury, C.C., Chambers, B.J., Prescott, A.R., Ljunggren, H.G., Watts, C.,

1997. Constitutive macropinocytosis allows TAP-dependent major his-

tocompatibility complex class I presentation of exogenous soluble an-

tigen by bone marrow-derived dendritic cells. Eur. J. Immunol. 27 (1),

280–288.

Norkin, L.C., Anderson, H.A., Wolfrom, S.A., Oppenheim, A., 2002.

Caveolar endocytosis of simian virus 40 is followed by brefeldin A-

sensitive transport to the endoplasmic reticulum, where the virus dis-

assembles. J. Virol. 76 (10), 5156–5166.

O’Sullivan, B.J., Thomas, R., 2002. CD40 ligation conditions dendritic cell

antigen-presenting function through sustained activation of NF-kappaB.

J. Immunol. 168 (11), 5491–5498.

Park, D.S., Selvey, L.A., Kelsall, S.R., Frazer, I.H., 1993. Human papillo-

mavirus type 16 E6, E7 and L1 and type 18 E7 proteins produced by

recombinant baculoviruses. J. Virol. Methods 45 (3), 303–318.

Pelkmans, L., Puntener, D., Helenius, A., 2002. Local actin polymerization

and dynamin recruitment in SV40-induced internalization of caveolae.

Science 296 (5567), 535–539.

Peng, S., Frazer, I.H., Fernando, G.J., Zhou, J., 1998. Papillomavirus virus-

like particles can deliver defined CTL epitopes to the MHC class I

pathway. Virology 240, 147–157.

Pengal, R.A., Ganesan, L.P., Fang, H., Marsh, C.B., Anderson, C.L.,

Tridandapani, S., 2003. SHIP-2 inositol phosphatase is inducibly

expressed in human monocytes and serves to regulate Fcgamma recep-

tor-mediated signaling. J. Biol. Chem. 278 (25), 22657–22663.

Price, A.A., Cumberbatch, M., Kimber, I., Ager, A., 1997. Alpha 6

integrins are required for Langerhans cell migration from the epider-

mis. J. Exp. Med. 186 (10), 1725–1735.

Rostand, K.S., Esko, J.D., 1997. Microbial adherence to and invasion

through proteoglycans. Infect. Immun. 65 (1), 1–8.

Rudolf, M.P., Nieland, J.D., DaSilva, D.M., Velders, M.P., Muller, M.,

Greenstone, H.L., Schiller, J.T., Kast, W.M., 1999. Induction of

HPV16 capsid protein-specific human T cell responses by virus-like

particles. Biol. Chem. 380 (3), 335–340.

Salamero, J., Bausinger, H., Mommaas, A.M., Lipsker, D., Proamer, F.,

Cazenave, J.P., Goud, B., de la Salle, H., Hanau, D., 2001. CD1a

molecules traffic through the early recycling endosomal pathway in

human Langerhans cells. J. Invest. Dermatol. 116 (3), 401–408.

Sallusto, F., Cella, M., Danieli, C., Lanzavecchia, A., 1995. Dendritic cells

use macropinocytosis and the mannose receptor to concentrate macro-

molecules in the major histocompatibility complex class II compart-

ment: downregulation by cytokines and bacterial products. J. Exp.

Med. 182, 389–400.

Shikano, S., Bonkobara, M., Zukas, P.K., Ariizumi, K., 2001a. Molecular

cloning of a dendritic cell-associated transmembrane protein, DC-HIL,

that promotes RGD-dependent adhesion of endothelial cells through

recognition of heparan sulfate proteoglycans. J. Biol. Chem. 276 (11),

8125–8134.

Shikano, S., Bonkobara, M., Zukas, P.K., Ariizumi, K., 2001b. Molecular

cloning of a dendritic cell-associated transmembrane protein, DC-HIL,

that promotes RGD-dependent adhesion of endothelial cells through

recognition of heparan sulfate proteoglycans. J. Biol. Chem. 276 (11),

8125–8134.

Sinnathamby, G., Eisenlohr, L.C., 2003. Presentation by recycling MHC

class II molecules of an influenza hemagglutinin-derived epitope that is

revealed in the early endosome by acidification. J. Immunol. 170 (7),

3504–3513.

Sotomayor, E.M., Borrello, I., Tubb, E., Rattis, F.M., Bien, H., Lu, Z., Fein,

S., Schoenberger, S., Levitsky, H.I., 1999. Conversion of tumor-specific

CD4+ T-cell tolerance to T-cell priming through in vivo ligation of

CD40. Nat. Med. 5 (7), 780–787.

Stringer, S.E., Gallagher, J.T., 1997. Heparan sulphate. Int. J. Biochem.

Cell Biol. 29 (5), 709–714.

Sugita, M., Grant, E.P., van Donselaar, E., Hsu, V.W., Rogers, R.A., Peters,

P.J., Brenner, M.B., 1999. Separate pathways for antigen presentation

by CD1 molecules. Immunity 11 (6), 743–752.

Summerford, C., Bartlett, J.S., Samulski, R.J., 1999. AlphaVbeta5 integrin:

a co-receptor for adeno-associated virus type 2 infection. Nat. Med. 5

(1), 78–82.

Suzich, J.A., Ghim, S.J., Palmer-Hill, F.J., White, W.I., Tamura, J.K., Bell,

J.A., Newsome, J.A., Jenson, A.B., Schlegel, R., 1995. Systemic im-

munization with papillomavirus L1 protein completely prevents the

development of viral mucosal papillomas. Proc. Natl. Acad. Sci.

U.S.A. 92 (25), 11553–11557.

Thompson, A.G., Pettit, A.R., Padmanabha, J., Mansfield, H., Frazer, I.H.,

Strutton, G.M., Thomas, R., 2002. Nuclear RelB+ cells are found in

normal lymphoid organs and in peripheral tissue in the context of

inflammation, but not under normal resting conditions. Immunol. Cell

Biol. 80 (2), 164–169.

Tsunetsugu-Yokota, Y., Morikawa, Y., Isogai, M., Kawana-Tachikawa, A.,

Odawara, T., Nakamura, T., Grassi, F., Autran, B., Iwamoto, A., 2003.

Yeast-derived human immunodeficiency virus type 1 p55(gag) virus-

like particles activate dendritic cells (DCs) and induce perforin ex-

pression in Gag-specific CD8(+) T cells by cross-presentation of DCs.

J. Virol. 77 (19), 10250–10259.

Turville, S.G., Cameron, P.U., Handley, A., Lin, G., Pohlmann, S., Doms,

R.W., Cunningham, A.L., 2002. Diversity of receptors binding HIV on

dendritic cell subsets. Nat. Immunol. 3 (10), 975–983.

Venkatesan, S., Rose, J.J., Lodge, R., Murphy, P.M., Foley, J.F., 2003.

Distinct mechanisms of agonist-induced endocytosis for human chemo-

kine receptors CCR5 and CXCR4. Mol. Biol. Cell 14 (8), 3305–3324.

Werling, D., Hope, J.C., Chaplin, P., Collins, R.A., Taylor, G., Howard,

C.J., 1999. Involvement of caveolae in the uptake of respiratory syn-

cytial virus antigen by dendritic cells. J. Leukocyte Biol. 66 (1), 50–58.

Zhou, J., Sun, X.Y., Stenzel, D.J., Frazer, I.H., 1991. Expression of vac-

cinia recombinant HPV 16 L1 and L2 ORF proteins in epithelial cells is

sufficient for assembly of HPV virion-like particles. Virology 185 (1),

251–257.