Junctional adhesion molecules mediate transendothelial migration of dendritic cell vaccine in cancer immunotherapy

Accepted Manuscript Junctional adhesion molecules mediate transendothelial migration of dendritic cell vaccine in cancer immunotherapy Seung-Eon Roh, Yideul Jeong, Myeong-Ho Kang, Yong-Soo Bae PII:

S0304-3835(18)30487-7

DOI:

10.1016/j.canlet.2018.07.029

Reference:

CAN 14003

To appear in:

Cancer Letters

Received Date: 26 April 2018 Revised Date:

21 June 2018

Accepted Date: 21 July 2018

Please cite this article as: S.-E. Roh, Y. Jeong, M.-H. Kang, Y.-S. Bae, Junctional adhesion molecules mediate transendothelial migration of dendritic cell vaccine in cancer immunotherapy, Cancer Letters (2018), doi: 10.1016/j.canlet.2018.07.029. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

Junctional adhesion molecules mediate transendothelial migration of dendritic cell vaccine in cancer immunotherapy

RI PT

Seung-Eon Roh1*, Yideul Jeong*, Myeong-Ho Kang and Yong-Soo Bae† Department of Biological Sciences, Science Research Center (SRC) for Immune Research on Non-lymphoid Organ (CIRNO), Sungkyunkwan University, Jangan-gu, Suwon, Gyeonggi-do 16419, South Korea

SC

ABSTRACT

In vitro generated dendritic cells (DCs) have been studied in cancer

M AN U

immunotherapy for decades. However, the detailed molecular mechanism underlying transendothelial migration (TEM) of DC vaccine across the endothelial barrier to regional lymph nodes (LNs) remains largely unknown. Here, we found that junctional adhesion molecule (JAM)-Like (JAML) is involved in the TEM of mouse bone marrowderived DCs (BMDCs). Treatment with an anti-JAML antibody or JAML knock-down significantly reduced the TEM activity of BMDCs, leading to impairment of DC-based

TE D

cancer immunotherapy. We found that the interaction of JAML of BMDCs with the coxsackie and adenovirus receptor of endothelial cells plays a crucial role in the TEM of BMDCs. On the other hand, human monocyte-derived DCs (MoDCs) did not express

EP

the JAML protein but still showed normal TEM activity. We found that MoDCs express only JAM1 and that the homophilic interaction of JAM1 is essential for MoDC TEM across a HUVEC monolayer. Our findings suggest that specific JAM family members

AC C

play an important role in the TEM of in vitro-generated mouse and human DCs from the inoculation site to regional LNs in DC-based cancer immunotherapy.

Keywords: Mouse bone marrow-derived dendritic cells (BMDCs), human monocyte-derived dendritic cells (MoDCs), TEM, JAM, JAML, DC immunotherapy

ACCEPTED MANUSCRIPT

Junctional adhesion molecules mediate transendothelial migration of dendritic cell

RI PT

vaccine in cancer immunotherapy

SC

Seung-Eon Roh1*, Yideul Jeong*, Myeong-Ho Kang and Yong-Soo Bae†

M AN U

Department of Biological Sciences, Science Research Center (SRC) for Immune Research on Non-lymphoid Organ (CIRNO), Sungkyunkwan University, Jangan-gu, Suwon, Gyeonggi-do 16419, South Korea

1

Present Address: Department of Neuroscience, Johns Hopkins University School of

TE D

Medicine, 725 North Wolfe St. Baltimore, MD 21205, USA.

†

EP

*These authors contributed equally to this work.

Corresponding author: Professor, Department of Biological Science; Director of CIRNO,

AC C

Research Complex Bldg 1, Sungkyunkwan University. Tel: 82-31-299-4149; Fax: 82-31-2907087; E-mail: [email protected]

-1-

ACCEPTED MANUSCRIPT

ABSTRACT In vitro generated dendritic cells (DCs) have been studied in cancer immunotherapy for decades. However, the detailed molecular mechanism underlying

RI PT

transendothelial migration (TEM) of DC vaccine across the endothelial barrier to regional lymph nodes (LNs) remains largely unknown. Here, we found that junctional adhesion molecule (JAM)-Like (JAML) is involved in the TEM of mouse bone marrow-

SC

derived DCs (BMDCs). Treatment with an anti-JAML antibody or JAML knock-down significantly reduced the TEM activity of BMDCs, leading to impairment of DC-based

M AN U

cancer immunotherapy. We found that the interaction of JAML of BMDCs with the coxsackie and adenovirus receptor of endothelial cells plays a crucial role in the TEM of BMDCs. On the other hand, human monocyte-derived DCs (MoDCs) did not express the JAML protein but still showed normal TEM activity. We found that MoDCs express

TE D

only JAM1 and that the homophilic interaction of JAM1 is essential for MoDC TEM across a HUVEC monolayer. Our findings suggest that specific JAM family members play an important role in the TEM of in vitro-generated mouse and human DCs from

AC C

EP

the inoculation site to regional LNs in DC-based cancer immunotherapy.

Keywords: Mouse bone marrow-derived dendritic cells (BMDCs), human monocyte-derived dendritic cells (MoDCs), TEM, JAM, JAML, DC immunotherapy

Abbreviation

APC, antigen presenting cell; BMDC, bone marrow-derived DC; CAR, coxsackie and adenovirus receptor; CCL, chemokine ligand; CCR, chemokine receptor; CTL, cytotoxic T lymphocyte; DC, dendritic cell; GM-CSF, granulocyte-macrophage colony-stimulating factor; HUVEC, human umbilical vein endothelial cell; IgSF,

immunoglobulin superfamily; imDC, immature dendritic cell; JAM, junctional adhesion molecule; JAML, junctional adhesion molecule-like; LN, lymph node; LPS, lipopolysaccharide; mDC, mature dendritic cell; MoDC, monocyte-derived DC; pDC, plasmacytoid DC; TEM, trans-endothelial migration -2-

ACCEPTED MANUSCRIPT

Introduction Dendritic cells (DCs) serve as professional antigen-presenting cells (APCs), which induce adaptive immunity and tolerance [1-3]. DCs circulate in blood vessels in immature

RI PT

form or as DC progenitors. Immature DCs localize in peripheral tissues such as the airways, skin epidermis and in the interstitial tissue of solid organs. Immature DCs or their precursors can also be recruited to the sites of inflammation and capture harmful stimuli, such as pathogens, foreign antigens [4, 5] or danger signals produced by damaged cells in response to

SC

ionizing radiation [6]. When DCs mature under pro-inflammatory conditions or via

M AN U

pathogen/damage-associated molecular pattern stimulation, the antigen-captured DCs migrate to regional lymph nodes (LNs) where they activate naive or antigen-specific T cells [3, 7]. DC trafficking is accompanied by multi-step events; for example, chemotaxis, rolling adhesion, tight adhesion, and/or transendothelial migration (TEM). These events are probably mediated by tightly regulated molecular interactions between the cells and tissues.

TE D

TEM is an indispensable process in DC migration, transmigrating from the blood vessels to peripheral tissues and, upon maturation, reverse-transmigrating into the blood or lymphatic

EP

vessels, traveling to the LNs [8]. TEM is associated with complex interactions between adhesion molecules. Platelet endothelial cell adhesion molecule-1 and β-1/2 integrins have

AC C

been reported to be involved in the TEM of myeloid DCs and plasmacytoid DCs (pDCs) across the activated endothelium in response to the chemokines CXCL12 and CCL5 [9]. However, the mechanisms underlying the TEM of in vitro-generated DCs have not yet been characterized even though bone marrow-derived DCs (BMDCs) [10-14] and monocytederived DCs (MoDCs) [15-17] have been studied in DC-based cancer immunotherapy for over two decades. Junctional adhesion molecule (JAM) is an intercellular adhesion molecule that belongs -3-

ACCEPTED MANUSCRIPT to the immunoglobulin superfamily (IgSF). It has been reported that pDC migration to regional LNs is mediated by interactions between integrins on pDCs and JAM-A (or JAM1) on high endothelial venules (HEVs) of regional LNs [18]. Another study showed that the

RI PT

vascular endothelial-JAM (VE-JAM)/JAM2, which is restricted to the HEV of the tonsils or LNs, interacts with T cell, natural killer (NK) cells, or DC probably through a JAM3 molecule [19]. Junctional adhesion molecule-like (JAML) was also identified as a member of

SC

the JAM family in retinoic acid-treated acute promyelocytic leukemic cells [20, 21]. JAML promotes leukocyte TEM through homophilic and heterophilic interactions [22]. JAML is

M AN U

exclusively expressed in neutrophils, monocytes, and some types of T cells [23] and is involved in the TEM of monocytes and neutrophils [21, 22, 24-26]. JAML mediates neutrophil and monocyte TEM by interacting with the coxsackie and adenovirus receptor (CAR), which is expressed in tight junctions of endothelial cells (ECs) [27-29]. However,

TE D

JAML has not yet been investigated in relation to DC TEM.

We found that JAML is highly expressed in mouse BMDCs, but not in in vivo DCs. In in vitro and in vivo TEM assays, we found that JAML and CAR play an important role in

EP

BMDC adhesion to ECs and TEM via intermolecular interactions. However, human MoDCs did not express JAML but instead had normal TEM capacity. Alternative to JAML, human

AC C

MoDCs were found to express large amounts of JAM1, which is involved in MoDC TEM via a homophilic interaction.

-4-

ACCEPTED MANUSCRIPT

Materials and Methods Mice and cell lines BALB/c and C57BL/6 mice (6–8 weeks old) were purchased from Koatech (Gyeonggi,

RI PT

Republic of Korea) and maintained in a specific pathogen-free (SPF) condition at the Animal Maintenance Facility of Sungkyunkwan University according to the University Animal Care and Use Guidelines. The mouse brain microvascular EC line (bEND.3), kindly provided by Y.

SC

S. Jung (Aju University School of Medicine, Suwon, Korea), was grown in Dulbecco’s modified Eagle’s medium (DMEM, HyCloneTM, Fisher Scientific) containing 10% fetal

M AN U

bovine serum (FBS) and penicillin/streptomycin as previously described [30]. Human umbilical vascular ECs (HUVEC obtained from American Type Culture Collection, ATCC Rockville, MD, USA) were cultured in sterile endothelial growth medium (EGM-2, Lonza

Reagents and antibodies

TE D

Ltd, Basel, Switzerland) supplemented with EGM-2 SingleQuots (Lonza).

Mouse and human granulocyte-macrophage colony-stimulating factor (GM-CSF) and human interleukin (IL)-4 were provided by CreaGene Inc. (Seongnam, Korea). Mouse and human

EP

tumor necrosis factor-α (TNF-α) were obtained from BioLegend (San Diego, CA, USA), and

AC C

mouse and human IFN-γ were obtained from R&D Systems (Minneapolis, MN, USA). Mouse and human macrophage inflammatory protein-1α and -3β (MIP-1α and MIP-3β), and monocyte chemoattractant protein-1 (MCP-1) were purchased from ProSpec (East Brunswick, NJ, USA). The mouse anti-JAML antibodies 4E10 and AMICA were obtained from BioLegend (San Diego, CA, USA) and from R&D Systems, respectively. Human JAML antibody (P-12) was purchased from Santa Cruz Biotechnology inc. (Dallas, TX, USA). Mouse anti-CAR antibody (RmcB) was obtained from Millipore (Boston, MA, USA), and anti-human JAM1 antibody (J10.4) was purchased from Santa Cruz Biotech. Cell tracking -5-

ACCEPTED MANUSCRIPT dyes were obtained; 5, 6-carboxyfluorescein succinimidyl ester (CFSE, Molecular Probes, Eugene, OR, USA), CellTrace Violet (CTV, Thermo-Fisher Scientific, Waltham, MA, USA) and 1,1′-dioctadecyl-3,3,3′,3′-tetramethylindodicarbocyanine, 4-chlorobenzenesulfonate salt

RI PT

(DiD, Thermo-Fisher Scientific). The antibodies FITC-CD8, PerCP5.5-CD45, Pe-Cy7CD11c, APC-CD19/CD3/NK1.1, APC-Cy7-MHC II and other antibodies for flow cytometry, as well as Amcyan-LIVE/DEAD cell dye, were obtained from Thermo-Fisher Scientific.

SC

Mouse BMDC and human MoDC

Mouse BMDCs were prepared from bone-marrow (BM) cells as described previously with

M AN U

minor modifications [31]. Briefly, BM cells, isolated from femurs and tibiae of 7-8 week old mice, were incubated with ACK lysing buffer (Lonza) for RBC lysis. After washing, cells were cultured in RPMI1640 (HyClone) containing 10% FBS (GIBCO), 100 U/mL penicillin, 100 µg/mL streptomycin, and 10 ng/mL mGM-CSF (CreaGene Inc, Seongnam, Korea). On

TE D

day 2, the medium was changed after vigorous washing with PBS to remove non-adherent cells. On day 4, the medium was replaced with a 1:1 mixture of fresh medium with the preexisting medium. Cells on day 6 were used as immature DCs (imDCs), which were matured

EP

by treating with lipopolysaccharide (LPS, Sigma-Aldrich, St Louis, USA) at 200 ng/mL for 24 h. The surface phenotypes and mean fluorescence intensities (MFIs) of BMDCs were

AC C

evaluated (Supplemental Fig. 1). Human MoDCs were produced from monocytes as described previously with some modifications [32]. Human peripheral blood mononuclear cells were incubated in culture plates for 1 h. Suspended cells were removed, and adherent monocytes

were

cultured

in

RPMI1640

containing

2%

autologous

serum,

penicillin/streptomycin, 1000 U/mL hGM-CSF (CreaGene Inc), and 500 U/mL hIL-4 (Ceagene Inc.) for 6 days. The resulting sells were used as immature MoDCs. Mature MoDCs were prepared by treating immature MoDCs with human TNF-α (10 ng/mL) and -6-

ACCEPTED MANUSCRIPT IFN-γ (10 ng/mL) for 24 h. The surface phenotypes and MFIs of MoDCs were examined (Supplemental Fig. 2). Mouse splenic DCs (spDCs) were isolated from the spleen of 6-8 week old C57BL/6 mice using CD11c-microbeads according to manufacturer’s instructions

RI PT

(Miltenyi Biotech, Bergisc Gladbach, Germany). Reverse transcriptase polymerase chain reaction (RT-PCR) and quantitative RT-PCR RT-PCR was performed as described previously [33] with minor modification. Cells were

SC

lysed with 500 µL of TRIzol reagent (Invitrogen, Carlsbad, CA, USA). RT-PCR was performed with an RT-PCR pre-mix kit (iNtRON Biotechnology, Daejeon, Korea) according

M AN U

to the manufacturer’s instructions. The following primer sets were used: mouse JAML (S) 5'gagtggattggctcttctcg-3'

and

(As)

5'-tcatttttgaggcggatttc-3';

human

JAML

(S)

5'-

cctctggtcttgggtggtaa-3'

and

(As)

5'-tcaaaatggcagggtttttc-3';

human

JAM1

(S)

5'-

gtgccttcagcaactcttcc-3' and (As) 5'-accagatgccaaaaaccaag-3'; human JAM2 (S) 5'and

(As)

5'-acctgcgatatccaacaga-3';

TE D

gctctgagtggaactgtgg-3'

gtactggccctgatcacgtt-3'and

(As)

5'-gctgccttgacaggagtttc-3';

human human

JAM3

(S)

5'-

JAM4

(S)

5'-

tacaatacgctgctgctgct-3' and (As) 5'-aggatttgggagggagagaa-3'; mouse β-actin (S) 5'-

EP

gtatgcctcggtcgtacca-3' and (As) 5'-cttctgcatcctgtcagcaa-3'; human GAPDH (S) 5'aatatccgttgtggatct-3' (As) 5'-tccaccacccttcctgta-3'. The quantitative RT-PCR analysis was

AC C

performed using the Fast SYBR Green Master Mix kit (Thermo-Fisher Scientific) with the same primers used in RT-PCR for JAM family genes. Relative expression was presented as mRNA abundance compared to GAPDH, housekeeping gene. Western blot analysis Western blot analysis was performed as described previously [33]. In brief, cells were lysed with a lysis buffer containing 50 mM Tris–HCl (pH 7.4), 150 mM NaCl, 1 mM DTT, 30 mM NaF, 10 mM Na3VO4, 0.5% NP-40, and a protease inhibitor cocktail (Pierce Protein -7-

ACCEPTED MANUSCRIPT Research Products, Rockford, IL, USA). Whole cell lysates were normalized with the Bradford protein assay (Bio-Rad, Hercules, CA, USA), and 40-120 µg of the lysates were separated on 8%-12% SDS-PAGE and then transferred to a polyvinylidene difluoride

RI PT

membrane (Millipore). The membranes were incubated with an optimal concentration of primary antibody overnight at 4°C, additionally incubated with horseradish peroxidase (HRP)-conjugated secondary antibody for 45 min, and visualized using an enhanced

SC

chemiluminescence (ECL) detection kit (Amersham Biotech, NJ, USA). Flow cytometric analysis

M AN U

To examine JAM expression on mouse/human DCs and to evaluate the TEM capacity of DCs and other cells, flow cytometry was performed as described previously [31]. Cells were harvested and stained with FACS antibodies diluted in FACS flow (BD Bioscience) for 20 min at 4°C. Antibodies used were PE-labeled anti-mouse CD11c (BioLegend), FITC-labeled

TE D

anti-human HLA-DR, PE-labeled anti-mouse JAML (BioLegend), PE-labeled anti-human AMICA (JAML) (R&D Systems), and anti-human JAM1 antibodies (Santa Cruz Biotechnology). spDCs were stained with FITC-CD8, PerCP5.5-CD45, Pe-Cy7-CD11c,

EP

APC-CD19/CD3/NK1.1, APC-Cy7-MHC II and Amcyan-LIVE/DEAD cell dye. Each splenic DC subset was negatively and positively gated with CD3/CD19/NK1.1 and CD45+

AC C

Live+ CD11c+ MHC II+ single cells, respectively. When necessary, FITC-conjugated secondary anti-mouse IgG was also used. After washing, cells were analyzed by FACSCanto II and FACS Diva software (BD Bioscience, San Jose, CA, USA). For the analysis of mouse BMDC phenotypes, antibodies of FITC-CD86, PE-CD80, PerCP-Cy5.5-CD40, APC-Cy7MHC II, Pacific Blue-MHC I were obtained from Biolegend, and PE-Cy7-CD11c and Amcyan-LIVE/DEAD cell dye were obtained from Thermo-Fisher Scientific. For human

-8-

ACCEPTED MANUSCRIPT Mo-DCs, antibodies of FITC-CD83, FITC-CD40, FITC-HLA-DR/DP/DQ, PE-CD80, PECD86, and APC-CD11c were obtained from BD Bioscience.

Confocal microscopy

RI PT

Confocal microscopy was performed as described previously with minor modifications [34, 35]. The cells were stained with PE-conjugated anti-mouse JAML antibody, fixed in paraformaldehyde for 5 min, and permeabilized with 0.2% Triton X-100 for 10 min. Next, the

SC

cells were then stained with DAPI (4', 6-diamidino-2-phenylindol) solution for 5 min. The cells were observed and their images were captured at low and high magnifications under a

M AN U

Laser Scanning Confocal Microscope (Carl Zeiss LSM 700, Oberkochen, Germany) and Zen software.

Transendothelial migration (TEM) assay

A TEM assay was performed as previously described with some modifications [8, 36]. The

TE D

0.5 µm size polycarbonate transwell plates (Corning, Sigma-Aldrich) were coated with 0.1% gelatin, and bEND.3 (or HUVEC) cells were cultured until confluence and were pre-activated

EP

by treating with TNF-α (25 ng/mL) for 6 h. The chemokines MIP-1α (for immature BMDCs), and MIP-3β (for mature BMDCs) were added into the lower chambers in order to initiate

AC C

migration. Immature or mature BMDC/MoDCs (1×104 cells) were added into the upper chamber and incubated for 4 h. Cells that migrated into the lower chambers were harvested, stained with mouse PE-conjugated CD11c antibody (FITC-conjugated CD1a antibody for human MoDCs and PE-conjugated CD14 for human THP-1 and monocyte), and assessed by flow cytometry. Cell adhesion assay and immunocytochemistry To evaluate the role of JAML and CAR in BMDC adhesion to the EC monolayer, bEND.3 -9-

ACCEPTED MANUSCRIPT cells were cultured until confluence on a cover glass in culture plates and treated with TNF-α (25 ng/mL) for 6 h. Then, the cell monolayers were incubated with control IgG or anti-CAR antibodies (5 µg/mL) for 1 h. Immature or mature BMDCs (5×104 cells) pre-incubated with

RI PT

control IgG or anti-JAML (10 µg/mL) were incubated with the bEND.3 cell monolayer for 1 h. After several washings, cells were fixed with 4% paraformaldehyde at 4°C for 20 min, and samples were stained with a PE-conjugated anti-CD11c antibody. The coverslip was mounted

SC

on a glass slide and visualized by fluorescence microscopy. Transfection of siRNA

M AN U

JAML siRNA obtained from Genolution Inc. (Seoul, Korea) was transfected into 5-day BMDCs in serum-free Opti-MEM (GIBCO) using an electroporator (Bio-Rad) and supplied cuvettes (Bio-Rad). Cells were then incubated in the presence or absence of TNF-α (10 ng/mL) and IFN-γ (10 ng/mL) for 2 days in cytokine-containing RPMI medium (FBS 10%).

Chemotaxis

TE D

Cells were assessed by Western blot assay, FACS analysis, and TEM assay.

EP

In vitro chemotaxis of BMDCs was assessed in a 24-transwell chamber as described [13]. Immature and mature BMDCs were treated with control isotype (IgG) or anti-JAML

AC C

antibodies (5 µg/mL). The cells were then washed twice with PBS and resuspended in serumfree RPMI. The BMDCs (1 x 104 cells, 0.1mL) were placed in the upper chamber and 600 µl serum-free RPMI containing CCL19 (300 ng/mL) was added to the lower chamber. After 4 h incubation at 37°C, the number of cells that migrated from the upper chamber to the lower chamber was measured by flow cytometry. In vivo migration of BMDC after injection - 10 -

ACCEPTED MANUSCRIPT BMDCs were treated with control IgG or anti-JAML antibody (5 ug/mL) for 1 hr. IgG or anti-JAML antibody-treated BMDCs were labeled with DiD or CTV respectively for 20 min. DiD-labeled anti-JAML antibody-treated BMDCs and CTV-labeled IgG-treated BMDCs

RI PT

were mixed together (each 1 x 106 cells/mouse in 0.1 mL) and subcutaneously injected into the right flank region of C57BL/6 mice. After 24 h, inguinal draining LNs were obtained, and labeled cells were analyzed by flow cytometry.

SC

DC-based tumor immunotherapy

Tumor immunotherapy was performed with tumor-bearing C57BL/6 mice as described

M AN U

previously [37]. E·G7 cells (ovalbumin-expressing mouse thymoma cell line, 5 x 105 cells/mouse) were injected into the right flank region of C57BL/6 mice. On days 3 and 10 after tumor inoculation, the mice were vaccinated with IgG-treated BMDCs or anti-JAML antibody-treated BMDCs that had been pulsed with OVA257-264 peptide for 1 h. Tumor growth

TE D

was monitored every 3 days using slide calipers. The volume of tumors was calculated as (A2 x B)/2, where A is the short axis and B is the long axis. Tetramer assay

EP

As described previously [37], CD8+ T cells isolated from the spleens of mice vaccinated with

AC C

IgG-treated BMDCs or anti-JAML-treated BMDCs were stained with H-2Kb SIINFEKL Pentamer (ProImmune Ltd., Oxford, UK), followed by flow cytometry analysis. Cytotoxic T lymphocyte (CTL) assay A CTL assay was performed as described [14, 37]. Briefly, CD8+ T cells were isolated from the spleens of vaccinated mice using a CD8a+ T cell isolation kit (Miltenyi Biotec). Target cells (E·G7) were stained with DiD and co-cultured with isolated CD8+ T cells at various ratios for 1 day. After fixation-compatible viability dye (FVD, Thermo-Fisher Scientific) - 11 -

ACCEPTED MANUSCRIPT staining, CTL activities were assessed in the DiD-labeled target cells by flow cytometry. Statistical analysis Statistical analysis was performed using GraphPad Prism 5.0 software. Comparisons of group

RI PT

means were calculated using Student’s t-test. A P value <0.05 was considered statistically significant. Data are representative of at least three independent experiments. All analyses were performed with GraphPad Prism (Version5.0, GraphPad Software, Inc), and are

AC C

EP

TE D

M AN U

SC

expressed as mean ± s.e.m.

- 12 -

ACCEPTED MANUSCRIPT

Results BMDCs express JAML JAML is known to be exclusively expressed in neutrophils, monocytes, and some T cells [23].

RI PT

However, in the present study, we found that JAML is also expressed in BMDCs, as was shown by RT-PCR and Western blot (Fig. 1A) and confocal microscopy (Fig. 1B). JAML expression in BMDCs was not altered in response to maturation signals, IFN-γ and TNF-α

SC

(Fig. 1A), although JAML had been reported to be induced by chemoattractants in neutrophils [27]. Moreover, JAML expression in BMDCs was not affected by the presence of

M AN U

Toll-like receptor (TLR) ligands (poly I:C and LPS) or the chemoattractants MIP-1α and MIP-3β (data not shown). We next examined the JAML expression profiles during BMDC development from BM precursor cells. Even though JAML was weakly expressed in BM precursor cells, it was significantly induced at the protein level during DC development,

TE D

while it was constitutively expressed at the mRNA level (Fig. 1C left). Interestingly, even though JAML expression was induced maximally on day 4 and then decreased during DC development, as determined by Western blot analysis, the surface expression of JAML was

EP

maintained from the DC precursor stage on day 4 up until the mature BMDC stage (Fig. 1C right). JAML expression was not different in immature and mature BMDCs, although CC

AC C

chemokine receptor (CCR)-7 (CCR7) expression was significantly enhanced when BMDCs were matured (Fig. 1D). However, JAML expression was undetectable in several DCs isolated from the spleen and LNs (Fig. 1E).

JAML is involved in BMDC TEM Chemokine receptor expression on the cell surface determines to where the cell migrate. Immature DCs express CCR-1, 4, and 5, which are responsive to the C-C motif chemokine - 13 -

ACCEPTED MANUSCRIPT ligand 3 (CCL3), known as MIP-1α. When mature, DCs rearrange the chemokine receptor repertoire and express CCR7, which is responsive to CCL19 (MIP-3β) and CCL21[4]. Previous studies have revealed that JAML plays an important role in the TEM of neutrophils

RI PT

and monocytes [27, 28]. To investigate whether JAML is involved in BMDC TEM, we established an in vitro TEM assay using TNF-α pre-activated bEND.3 cell monolayer on transwell plates. The monolayer supported DC TEM efficiently (Fig. 2A). The addition of

SC

chemokines (MIP-1α for immature DCs and MIP-3β for mature DCs) increased the TEM activity, and mature DCs transmigrated more efficiently than immature DCs (Fig. 2A).

M AN U

Transmigration of both immature and mature DCs was reduced in the presence of a JAMLneutralizing antibody in a dose-dependent manner (Fig. 2B). However, the TEM of mature DCs compared with immature DCs was more greatly affected by JAML neutralization, suggesting that JAML plays a more important role in the TEM of mature DCs (Fig 2B). A

TE D

JAML-neutralizing antibody did not affect CCR7 expression (Fig 2C) or chemotactic migration (Fig. 2D) of either immature or mature DCs. To further confirm the role of JAML in DC TEM, JAML mRNA was specifically knocked-down with JAML siRNA. Both

EP

immature and mature DCs transfected with si-jaml showed reduced TEM activity (Fig. 2E). When DCs were subcutaneously inoculated into mice after treatment with a JAML-

AC C

neutralizing antibody, the number of DCs that migrated into regional LNs was significantly reduced compared with the IgG-treated control (Fig. 2F). Taken together, our results suggest that JAML plays an important role in the TEM of BMDCs to regional LNs when a DC-based vaccine is inoculated.

BMDC TEM requires interactions between BMDC JAML and endothelial CAR JAML belongs to the IgSF which includes CAR, platelet endothelial cell adhesion molecule, - 14 -

ACCEPTED MANUSCRIPT and JAM1 (or JAM-A), 2 (or JAM-B), 3 (or JAM-C), and 4. They share structural and functional features and, in some cases, interact with each other to contribute to cell adhesion and migration [20]. Among the IgSF molecules, CAR has been proposed as a counter-

RI PT

receptor to JAML, and the interaction between the two molecules has been reported as important in neutrophil and monocyte diapedesis [27, 28]. To investigate whether CAR is important in DC TEM, we performed the TEM assay in the presence of an anti-CAR

SC

neutralizing antibody. As expected, DC TEM was reduced in a dose-dependent manner in both immature and mature DCs. Addition of anti-JAML and anti-CAR antibodies further

M AN U

lowered the TEM of both BMDCs (Fig. 3A). To directly demonstrate that the JAML-CAR interaction is important during BMDC transmigration, we performed the adhesion assay with a bEND.3 cell monolayer pre-activated with TNF-α. The numbers of immature and mature BMDCs that were attached to the bEND.3 monolayer were counted under a fluorescence

TE D

microscope. When BMDC and the bEND.3 monolayer were treated with anti-JAML and antiCAR monoclonal antibodies, respectively, the number of attached cells was significantly reduced compared to the control group (Fig. 3B). These results suggest that CAR and JAML

AC C

diapedesis.

EP

may interact with each other in BMDC TEM, as had been shown in neutrophil and monocyte

JAML plays an important role in DC immunotherapy in tumor-bearing mice. In DC immunotherapy in mice, in vitro-generated, antigen-pulsed BMDCs are injected into mice subcutaneously. These BMDCs migrate to regional LNs by TEM and induce anti-tumor T cell immunity (CTL), which inhibits tumor growth. Administration of OVA-pulsed BMDCs was effective in inhibiting E·G7 tumor growth in mice, but the BMDC-mediated inhibition of tumor growth was significantly impaired by the treatment of BMDCs with anti-JAML - 15 -

ACCEPTED MANUSCRIPT antibody before administration (Fig. 4A). In the following immunological analysis in vaccinated mice, E·G7-specific CTL activity was significantly reduced in mice vaccinated with anti-JAML Ab-treated BMDCs compared to control mice vaccinated with IgG-treated

RI PT

BMDCs (Fig. 4B). Similarly, the OVA-specific, tetramer-positive CD8+ T cell population was significantly reduced in mice vaccinated with anti-JAML Ab-treated BMDCs (Fig. 4C). These results suggest that JAML expression in BMDCs plays an important role in inducing

SC

tumor-specific CTL response by facilitating DC migration to regional LNs during DC

M AN U

immunotherapy.

Human MoDCs do not express JAML but show normal TEM capacity We next examined whether or not human DCs express JAML. We found that THP-1 and human monocytes express JAML, but human MoDCs do not express JAML on their surface,

TE D

even though they express the mRNA (Fig. 5A). JAML expression was significantly downregulated after one day during six days of DC development from monocytes (Fig. 5B). The treatment of DCs with MG132 did not reconstitute JAML protein expression (data not

EP

shown), implying that JAML expression is controlled at the translational level rather than by post-translational degradation. Even though JAML was not expressed in human MoDCs, both

AC C

immature and mature MoDCs showed normal TEM capacity compared to THP-1 and monocytes in HUVEC monolayer transwell plates (Fig. 5C), which suggests that MoDCs have other molecules for TEM.

TEM of human MoDCs is mediated by homophilic interaction of JAM1, in contrast to monocyte TEM Since it was shown that MoDCs do not express JAML, we investigated if other JAM family - 16 -

ACCEPTED MANUSCRIPT members are expressed in MoDCs. Out of the JAM family of proteins, MoDCs expressed only JAM1 (JAM-A), and JAM1 expression was markedly enhanced during differentiation from monocytes, while monocytes expressed JAM1, 2, and 3 (Fig. 6A and 6B). The role of

RI PT

JAM1 in DC TEM is still under debate. It has been shown that DC infiltration in LNs is increased in JAM1-deficient mice [38]. On the contrary, it has also been reported that the blocking of JAM-A results in the reduction of TEM of splenic pDCs through lymphatic ECs

SC

in vitro [18]. However, these findings were reported in mice. We assessed the role of JAM1 in the TEM of human MoDCs using a HUVEC monolayer system. Blocking JAM1 with an

M AN U

anti-JAM1 antibody significantly inhibited the TEM capacity of mature MoDCs in a dosedependent manner (Fig. 6C). JAM1 is known to mediate platelet adhesion to activated endothelium through a homophilic interaction [39]. In our experiments, JAM1 expression was also observed in the HUVEC system (Fig. 6B). In the TEM assay using a HUVEC monolayer, MoDC TEM was reduced by 55% when JAM1 in MoDCs was blocked, but was

TE D

reduced by 40% when HUVEC was blocked by 10 µg/mL of an anti-JAM1 antibody (Fig. 6C). This difference may be due to less accessibility of the antibody to JAM1 that is localized

EP

in the tight junctions of the intercellular HUVEC monolayer. Additive inhibitions were observed when MoDCs and HUVECs were simultaneously blocked with an anti-JAM1

AC C

antibody (Fig. 6C). Overall, the data imply that the homophilic interaction of JAM1 contribute to the TEM of MoDCs. It has been reported that JAM-A accelerates lesion formation and monocyte infiltration in atherosclerosis-prone mice [40]. Subsequently, monocytic JAML was found to play a critical role in regulating monocyte TEM [28]. To investigate whether JAML and JAM1 are cooperatively involved in the TEM of monocytes or not, we assessed TEM of monocytes with both blocking antibodies. The neutralization of either JAML or JAM1 resulted in a significant reduction of monocyte TEMs (Fig. 6D). - 17 -

ACCEPTED MANUSCRIPT Furthermore, simultaneous blocking with JAML and JAM1 antibodies additively decreased the TEM of monocytes (Fig. 6D), suggesting that JAM1 and JAML communally contribute to

AC C

EP

TE D

M AN U

SC

RI PT

monocyte TEM in contrast to the TEM of MoDCs.

- 18 -

ACCEPTED MANUSCRIPT

Discussion The migration and accurate positioning of DCs are essential for their immunological functions. They originate from bone marrow and localize to peripheral tissues, such as the

RI PT

skin and mucosa. DCs can also be recruited quickly to inflammatory sites in response to chemoattractants. Upon encountering foreign antigens or pathogen signals, DCs migrate to the draining LN in which they prime naive T cells [4, 5, 7].

SC

During leukocyte migration, TEM is a crucial step in which JAM family members are deeply involved. JAM-1, JAM-2, and JAM-3 have been suggested to be involved in

M AN U

transmigration of monocytes, neutrophils, and some T cells through homophilic and heterophilic interactions [22]. While the roles of JAM family members in monocyte and neutrophil diapedesis have been well investigated, their roles in DC functioning, particularly in in vitro-generated DCs, have not yet been fully elucidated. Among the JAM family, JAML

TE D

has been reported as a new member. The importance of JAML in TEM of monocytes and neutrophils led us to examine the expression of JAM molecules in in vitro-generated DCs. In the present study, we found that JAML is expressed in both immature and mature

EP

mouse BMDCs (Fig. 1A, B). Previous reports have revealed that inflammatory stimulation upregulates JAML expression in neutrophils and monocytes [29, 41], suggesting that

AC C

inflammation-induced JAML upregulation is related to increased diapedesis. However, BMDCs did not upregulate JAML expression in response to pro-inflammatory (TNF-α, IFNγ) or pathogenic signals (LPS, poly I:C) or to other chemokines (MCP-1, MIP-1α, MIP-3β) (data not shown). Other agents that upregulate JAML expression in BMDCs remain to be investigated. We expected that JAML would also be expressed in a subset of in vivo-isolated DCs for TEM activity. However, none of those DCs expressed JAML (Fig. 1E), suggesting that in vitro-generated BMDC TEM might be different from in vivo DC TEM. - 19 -

ACCEPTED MANUSCRIPT DCs interact with the extracellular matrix and ECs for transmigration and utilize specific chemokine receptor-ligand pathways to determine their destination: CCR2-CCL2 (MCP-1α), CCR5-CCL5 (RANTES), and CCR6-CCL20 (MIP-3α) for immature DCs; CCR7-CCL19

RI PT

(MIP-3β) and CCL21 (SLC) for mature DCs. Moreover, during inflammatory conditions, ECs increase integrin expression on their surface and are more amenable to diapedesis. As expected, BMDC TEM increased in the presence of chemokines and TNF-α-preactivated EC

SC

monolayers (Fig. 2A). Even though there was not much difference in JAML expression between immature and mature BMDCs (Fig. 1), mature BMDCs showed higher TEM activity

M AN U

than immature BMDCs (Fig. 2A), suggesting that molecules other than JAML are involved in the TEM of mature BMDCs. JAML neutralization or JAML gene knock-down decreased the TEM of BMDCs in both in vitro TEM and in vivo LN migration (Fig. 2), indicating that JAML plays an important role in BMDC migration to regional LNs after inoculation with a

TE D

DC vaccine.

JAM family members, CAR, and endothelial cell-specific adhesion molecules mediate adhesion and migration of various immune cells and stem cells through homophilic and

EP

heterophilic interactions [27, 29, 41, 42]. Among them, CAR, a tight junction-related adhesion molecule, had been suggested to interact with JAML on neutrophils [27] and

AC C

monocytes [28]. When CAR was neutralized, BMDC TEM decreased in a dose- dependent manner (Fig. 3A). Additive inhibition was observed when CAR and JAML were simultaneously blocked (Fig. 3), which suggests that CAR also interact with JAML for the TEM of BMDCs.

DC-based cancer immunotherapy has been studied over the past two decades [43]. However, the mechanism of in vitro-generated DC TEM in DC-based cancer immunotherapy remains largely unknown. In the present study, we found that JAML plays a crucial role in - 20 -

ACCEPTED MANUSCRIPT BMDC transmigration to regional LNs and in the induction of antigen-specific CTL responses in DC immunotherapy in mice (Fig. 4). Human primary monocytes and THP-1 cells expressed JAML (Fig. 5), as previously

RI PT

reported [27, 28]. However, human MoDCs did not express JAML protein on their surface, even though JAML mRNA was detected in both immature and mature MoDCs (Fig. 5A). JAML expression on monocytes rapidly disappeared during the early stage of MoDC

SC

generation from monocytes in the presence of GM-CSF and IL-4 (Fig. 5B). However, MoDCs showed normal TEM activity even in the absence of JAML expression (Fig. 5C).

M AN U

MoDCs expressed only JAM1 from the JAM family of proteins (Fig. 6A and 6B). JAM1 has been reported to mediate leukocyte transmigration [44]. However, the role of JAM1 in DC TEM, particularly in MoDC TEM, was then unknown. In the present study, we found that the homophilic interaction of JAM1 between MoDCs and HUVECs played a crucial role in

TE D

MoDC TEM (Fig. 6C). We also found that JAM1 and JAML were involved in monocyte TEM, which differs from what was observed in MoDC TEM (Fig. 6D). To date, a number of studies have been conducted to improve the efficacy of DC

EP

vaccines in DC-based cancer immunotherapy [45]. One of these efforts is to deliver a DC vaccine more efficiently to regional LN, to induce strong anti-tumor immunity. For this

AC C

purpose, studies have investigated ectopic expression of CCR7 in DC vaccine [46] or topical application of an immune response modifier such as imiquimod (a toll-like receptor 7 ligand) at the administration site of DC vaccine [47]. However, these studies aimed to find a way to improve the chemotaxis of DC vaccines or to increase the viability of the administered DC vaccines independent of the TEM activity of DC vaccine. Apart from chemotaxis, TEM is also a crucial step in DC vaccine migration to regional LN after administration, but the mechanism of in vitro-generated DC TEM remains largely unknown. In the present study, we - 21 -

ACCEPTED MANUSCRIPT found that JAML plays a crucial role in the TEM of mouse BMDCs and that homophilic interaction of JAM1 is involved in MoDC TEM across the endothelial barrier to regional LNs. Our findings provide a better understanding of TEM of in vitro-generated DCs, which could

AC C

EP

TE D

M AN U

SC

associated JAM molecules for cancer immunotherapy.

RI PT

lead to the development of more effective DC vaccine through modification of TEM-

- 22 -

ACCEPTED MANUSCRIPT

Acknowledgement We thank Mohammad Alam Miah and Cheol-Hee Yoon for their helpful assistance and YiSook Jung (Aju University School of Medicine, Suwon, Korea) for her excellent technical

RI PT

support. This work was supported by a National Research Foundation (NRF) grant funded by the Korea Ministry of Science and ICT (SRC-2017R1A5A1014560) and in part by the Bio &

Conflicts of Interest:

AC C

EP

TE D

M AN U

No potential conflicts of interest were disclosed.

SC

Medical Technology Development Program of NRF (2012M3A9B402826).

- 23 -

ACCEPTED MANUSCRIPT

References [1] J. Banchereau, R.M. Steinman, Dendritic cells and the control of immunity, Nature, 392 (1998) 245-252.

RI PT

[2] R.M. Steinman, The dendritic cell system and its role in immunogenicity, Annu Rev Immunol, 9 (1991) 271-296. [3] R.M. Steinman, J. Banchereau, Taking dendritic cells into medicine, Nature, 449 (2007) 419-426.

SC

[4] D. Alvarez, E.H. Vollmann, U.H. von Andrian, Mechanisms and consequences of dendritic cell migration, Immunity, 29 (2008) 325-342. [5] L. Flores-Romo, In vivo maturation and migration of dendritic cells, Immunology, 102 (2001) 255-262.

M AN U

[6] I.V. Mavragani, Z. Nikitaki, M.P. Souli, A. Aziz, S. Nowsheen, K. Aziz, E. Rogakou, A.G. Georgakilas, Complex DNA Damage: A Route to Radiation-Induced Genomic Instability and Carcinogenesis, Cancers, 9 (2017) 91. [7] G.J. Randolph, V. Angeli, M.A. Swartz, Dendritic-cell trafficking to lymph nodes through lymphatic vessels, Nat Rev Immunol, 5 (2005) 617-628.

TE D

[8] G. D'Amico, G. Bianchi, S. Bernasconi, L. Bersani, L. Piemonti, S. Sozzani, A. Mantovani, P. Allavena, Adhesion, transendothelial migration, and reverse transmigration of in vitro cultured dendritic cells, Blood, 92 (1998) 207-214.

EP

[9] G. de la Rosa, N. Longo, J.L. Rodriguez-Fernandez, A. Puig-Kroger, A. Pineda, A.L. Corbi, P. Sanchez-Mateos, Migration of human blood dendritic cells across endothelial cell monolayers: adhesion molecules and chemokines involved in subset-specific transmigration, Journal of leukocyte biology, 73 (2003) 639-649.

AC C

[10] D.S. Lim, J.H. Kim, D.S. Lee, C.H. Yoon, Y.S. Bae, DC immunotherapy is highly effective for the inhibition of tumor metastasis or recurrence, although it is not efficient for the eradication of established solid tumors, Cancer immunology, immunotherapy : CII, 56 (2007) 1817-1829. [11] D.S. Lim, M.S. Kang, J.A. Jeong, Y.S. Bae, Semi-mature DC are immunogenic and not tolerogenic when inoculated at a high dose in collagen-induced arthritis mice, European journal of immunology, 39 (2009) 1334-1343. [12] M.S. Ahmed, S.E. Byeon, Y. Jeong, M.A. Miah, M. Salahuddin, Y. Lee, S.S. Park, Y.S. Bae, Dab2, a negative regulator of DC immunogenicity, is an attractive molecular target for DC-based immunotherapy, Oncoimmunology, 4 (2015) e984550. [13] M.A. Miah, S.E. Byeon, M.S. Ahmed, C.H. Yoon, S.J. Ha, Y.S. Bae, Egr2 induced during DC development acts as an intrinsic negative regulator of DC immunogenicity, European journal of immunology, 43 (2013) 2484-2496. - 24 -

ACCEPTED MANUSCRIPT [14] M.A. Miah, C.H. Yoon, J. Kim, J. Jang, Y.R. Seong, Y.S. Bae, CISH is induced during DC development and regulates DC-mediated CTL activation, European journal of immunology, 42 (2012) 58-68.

RI PT

[15] J.H. Lee, W.Y. Tak, Y. Lee, M.K. Heo, J.S. Song, H.Y. Kim, S.Y. Park, S.H. Bae, J.H. Lee, J. Heo, K.H. Kim, Y.S. Bae, Y.J. Kim, Adjuvant immunotherapy with autologous dendritic cells for hepatocellular carcinoma, randomized phase II study, Oncoimmunology, 6 (2017) e1328335. [16] J.H. Lee, Y. Lee, M. Lee, M.K. Heo, J.S. Song, K.H. Kim, H. Lee, N.J. Yi, K.W. Lee, K.S. Suh, Y.S. Bae, Y.J. Kim, A phase I/IIa study of adjuvant immunotherapy with tumour antigen-pulsed dendritic cells in patients with hepatocellular carcinoma, British journal of cancer, 113 (2015) 1666-1676.

M AN U

SC

[17] F. Tada, M. Abe, M. Hirooka, Y. Ikeda, Y. Hiasa, Y. Lee, N.C. Jung, W.B. Lee, H.S. Lee, Y.S. Bae, M. Onji, Phase I/II study of immunotherapy using tumor antigen-pulsed dendritic cells in patients with hepatocellular carcinoma, International journal of oncology, 41 (2012) 1601-1609. [18] T. Matsutani, T. Tanaka, K. Tohya, K. Otani, M.H. Jang, E. Umemoto, K. Taniguchi, H. Hayasaka, K. Ueda, M. Miyasaka, Plasmacytoid dendritic cells employ multiple cell adhesion molecules sequentially to interact with high endothelial venule cells - molecular basis of their trafficking to lymph nodes, Int Immunol, 19 (2007) 1031-1037.

TE D

[19] T.W. Liang, H.H. Chiu, A. Gurney, A. Sidle, D.B. Tumas, P. Schow, J. Foster, T. Klassen, K. Dennis, R.A. DeMarco, T. Pham, G. Frantz, S. Fong, Vascular endothelial-junctional adhesion molecule (VE-JAM)/JAM 2 interacts with T, NK, and dendritic cells through JAM 3, J Immunol, 168 (2002) 1618-1626.

EP

[20] C. Moog-Lutz, F. Cave-Riant, F.C. Guibal, M.A. Breau, Y. Di Gioia, P.O. Couraud, Y.E. Cayre, S. Bourdoulous, P.G. Lutz, JAML, a novel protein with characteristics of a junctional adhesion molecule, is induced during differentiation of myeloid leukemia cells, Blood, 102 (2003) 3371-3378.

AC C

[21] C. Weber, L. Fraemohs, E. Dejana, The role of junctional adhesion molecules in vascular inflammation, Nat Rev Immunol, 7 (2007) 467-477. [22] G. Bazzoni, The JAM family of junctional adhesion molecules, Curr Opin Cell Biol, 15 (2003) 525-530. [23] A.C. Luissint, P.G. Lutz, D.A. Calderwood, P.O. Couraud, S. Bourdoulous, JAM-Lmediated leukocyte adhesion to endothelial cells is regulated in cis by alpha4beta1 integrin activation, J Cell Biol, 183 (2008) 1159-1173. [24] B.A. Imhof, M. Aurrand-Lions, Adhesion mechanisms regulating the migration of monocytes, Nat Rev Immunol, 4 (2004) 432-444. [25] W.A. Muller, Leukocyte-endothelial-cell interactions in leukocyte transmigration and the inflammatory response, Trends Immunol, 24 (2003) 327-334. - 25 -

ACCEPTED MANUSCRIPT [26] C. Weber, Novel mechanistic concepts for the control of leukocyte transmigration: specialization of integrins, chemokines, and junctional molecules, J Mol Med, 81 (2003) 4-19.

RI PT

[27] K. Zen, Y. Liu, I.C. McCall, T. Wu, W. Lee, B.A. Babbin, A. Nusrat, C.A. Parkos, Neutrophil migration across tight junctions is mediated by adhesive interactions between epithelial coxsackie and adenovirus receptor and a junctional adhesion molecule-like protein on neutrophils, Molecular biology of the cell, 16 (2005) 2694-2703. [28] Y.L. Guo, R. Bai, C.X. Chen, D.Q. Liu, Y. Liu, C.Y. Zhang, K. Zen, Role of junctional adhesion molecule-like protein in mediating monocyte transendothelial migration, Arteriosclerosis, thrombosis, and vascular biology, 29 (2009) 75-83.

SC

[29] P. Verdino, D.A. Witherden, W.L. Havran, I.A. Wilson, The molecular interaction of CAR and JAML recruits the central cell signal transducer PI3K, Science, 329 (2010) 12101214.

M AN U

[30] S.L. Park, Y.M. Kim, J.H. Ahn, S.H. Lee, E.J. Baik, C.H. Moon, Y.S. Jung, Cadmium stimulates the expression of vascular cell adhesion molecule-1 (VCAM-1) via p38 mitogenactivated protein kinase (MAPK) and JNK activation in cerebrovascular endothelial cells, J Pharmacol Sci, 110 (2009) 405-409. [31] Y.E. Choi, H.N. Yu, C.H. Yoon, Y.S. Bae, Tumor-mediated down-regulation of MHC class II in DC development is attributable to the epigenetic control of the CIITA type I promoter, Eur J Immunol, 39 (2009) 858-868.

TE D

[32] J.H. Ahn, Y. Lee, C. Jeon, S.J. Lee, B.H. Lee, K.D. Choi, Y.S. Bae, Identification of the genes differentially expressed in human dendritic cell subsets by cDNA subtraction and microarray analysis, Blood, 100 (2002) 1742-1754.

EP

[33] C.H. Yoon, E.S. Lee, D.S. Lim, Y.S. Bae, PKR, a p53 target gene, plays a crucial role in the tumor-suppressor function of p53, Proceedings of the National Academy of Sciences of the United States of America, 106 (2009) 7852-7857. [34] T.I. Arnon, R.M. Horton, I.L. Grigorova, J.G. Cyster, Visualization of splenic marginal zone B-cell shuttling and follicular B-cell egress, Nature, 493 (2013) 684-688.

AC C

[35] C.H. Yoon, S.Y. Kim, S.E. Byeon, Y. Jeong, J. Lee, K.P. Kim, J. Park, Y.S. Bae, p53derived host restriction of HIV-1 replication by protein kinase R-mediated Tat phosphorylation and inactivation, Journal of virology, 89 (2015) 4262-4280. [36] E.J. Park, J.R. Mora, C.V. Carman, J. Chen, Y. Sasaki, G. Cheng, U.H. von Andrian, M. Shimaoka, Aberrant activation of integrin alpha4beta7 suppresses lymphocyte migration to the gut, J Clin Invest, 117 (2007) 2526-2538. [37] M. Ahmed, S.E. Byeon, Y. Jeong, M. Miah, M. Salahuddin, Y. Lee, S.-S. Park, Y.-S. Bae, Dab2, a negative regulator of DC immunogenicity, is an attractive molecular target for DCbased immunotherapy, Oncoimmunology, 4 (2015) 1-15. [38] M.R. Cera, A. Del Prete, A. Vecchi, M. Corada, I. Martin-Padura, T. Motoike, P. Tonetti, G. Bazzoni, W. Vermi, F. Gentili, S. Bernasconi, T.N. Sato, A. Mantovani, E. Dejana, - 26 -

ACCEPTED MANUSCRIPT Increased DC trafficking to lymph nodes and contact hypersensitivity in junctional adhesion molecule-A-deficient mice, J Clin Invest, 114 (2004) 729-738. [39] A. Babinska, M.H. Kedees, H. Athar, T. Ahmed, O. Batuman, Y.H. Ehrlich, M.M. Hussain, E. Kornecki, F11-receptor (F11R/JAM) mediates platelet adhesion to endothelial cells: role in inflammatory thrombosis, Thrombosis and haemostasis, 88 (2002) 843-850.

RI PT

[40] A. Zernecke, E.A. Liehn, L. Fraemohs, P. von Hundelshausen, R.R. Koenen, M. Corada, E. Dejana, C. Weber, Importance of junctional adhesion molecule-A for neointimal lesion formation and infiltration in atherosclerosis-prone mice, Arteriosclerosis, thrombosis, and vascular biology, 26 (2006) e10-13.

SC

[41] D.A. Weber, R. Sumagin, I.C. McCall, G. Leoni, P.A. Neumann, R. Andargachew, J.C. Brazil, O. Medina-Contreras, T.L. Denning, A. Nusrat, C.A. Parkos, Neutrophil-derived JAML inhibits repair of intestinal epithelial injury during acute inflammation, Mucosal immunology, 7 (2014) 1221-1232.

M AN U

[42] G. Nagamatsu, M. Ohmura, T. Mizukami, I. Hamaguchi, S. Hirabayashi, S. Yoshida, Y. Hata, T. Suda, K. Ohbo, A CTX family cell adhesion molecule, JAM4, is expressed in stem cell and progenitor cell populations of both male germ cell and hematopoietic cell lineages, Mol Cell Biol, 26 (2006) 8498-8506. [43] K. Palucka, J. Banchereau, SnapShot: cancer vaccines, Cell, 157 (2014) 516-516 e511.

TE D

[44] J. Jaczewska, M.H. Abdulreda, C.Y. Yau, M.M. Schmitt, I. Schubert, P.O. Berggren, C. Weber, R.R. Koenen, V.T. Moy, E.P. Wojcikiewicz, TNF-alpha and IFN-gamma promote lymphocyte adhesion to endothelial junctional regions facilitating transendothelial migration, Journal of leukocyte biology, 95 (2014) 265-274. [45] K. Palucka, J. Banchereau, Dendritic-cell-based therapeutic cancer vaccines, Immunity, 39 (2013) 38-48.

AC C

EP

[46] N. Okada, N. Mori, R. Koretomo, Y. Okada, T. Nakayama, O. Yoshie, H. Mizuguchi, T. Hayakawa, S. Nakagawa, T. Mayumi, T. Fujita, A. Yamamoto, Augmentation of the migratory ability of DC-based vaccine into regional lymph nodes by efficient CCR7 gene transduction, Gene therapy, 12 (2005) 129-139. [47] R.M. Prins, N. Craft, K.W. Bruhn, H. Khan-Farooqi, R.C. Koya, R. Stripecke, J.F. Miller, L.M. Liau, The TLR-7 agonist, imiquimod, enhances dendritic cell survival and promotes tumor antigen-specific T cell priming: relation to central nervous system antitumor immunity, Journal of immunology, 176 (2006) 157-164.

- 27 -

ACCEPTED MANUSCRIPT

Figure Legends Figure 1. JAML expression in mouse BMDCs and in in vivo DCs Immature (imDC) and mature BMDC (mDC) were prepared as described in the Materials and

RI PT

Methods. (A) JAML expression was assessed by RT-PCR, Western blot (WB), and flow cytometry in imDCs and mDCs. (B) JAML expression in imDCs and mDCs was examined under confocal microscopy (60X and 300X). BMDCs were stained with a PE-conjugated

SC

JAML antibody and then fixed with 4% formaldehyde for DAPI staining. The boxed region of the mDC image is magnified by 300X. (C) JAML expression was assessed during BMDC

M AN U

development from BM cells by RT-PCR, WB, and flow cytometry. Cells were harvested and examined at 2-day (2D), 4-day (4D), 6-day (imDC), and 7-day mDC time points. (D) CCR7 and JAML expression was assessed in imDCs and mDCs. (E) JAML expression was assessed by flow cytometry in in vitro-generated BMDCs and in DC subsets isolated from mouse

TE D

splenocytes.

Figure 2. JAML-mediated TEM of BMDCs

EP

TEM activity was assessed as described in the Materials and Methods. (A) TEM of BMDCs (imDC and mDC) was assessed in the presence or absence of target chemokines in the EC

AC C

monolayer with or without preactivation by TNF-α. (B) imDCs and mDCs were pretreated with increasing doses of an anti-JAML antibody, and TEM was assessed. (C) CCR7 and JAML expression was assessed by flow cytometry in imDCs and mDCs, which were pretreated with a JAML blocking antibody (left). Statistical data from three replicate experiments are shown (right). (D) IgG- or anti-JAML antibody-treated imDCs and mDCs were assessed for their chemotactic activity in normal transwell plates. (E) The effects of JAML knock-down on BMDC TEM were assessed. JAML knock-down was confirmed by - 28 -

ACCEPTED MANUSCRIPT Western blot analysis and flow cytometry. The TEM of JAML-silenced imDCs and mDCs was assessed. (F) In vivo migration of IgG- and anti-JAML antibody-treated BMDCs was assessed as described in the Materials and Methods. DiD-labeled anti-JAML antibody-treated

RI PT

BMDCs and CTV-labeled IgG-treated BMDC were mixed (each 1 x 106 cells/mouse in 0.1 mL) and subcutaneously injected into the right flank region of C57BL/6 mice (n=3). After 24 h, inguinal LNs were obtained, and labeled cells were analyzed by flow cytometry. Data in A,

SC

B, D, E, and F are mean ± s.e.m. of three replicate experiments or three independent

M AN U

experiments (n=3). *P＜0.05, **P<0.01, two-way ANOVA.

Figure 3. BMDC JAML and endothelial CAR involvement in BMDC TEM (A) TNF-α-activated EC (bEND.3 cell) monolayers were pre-treated with control IgG or antiCAR antibody. BMDCs (imDC and mDC) were treated with IgG or anti-JAML antibodies;

TE D

TEM activity of antibody-treated BMDCs was assessed in an EC monolayer. (B) The EC monolayer on a coverslip was treated with control IgG or anti-CAR antibodies (5 µg/mL). BMDCs were treated with control IgG or anti-JAML antibodies and then incubated with the

EP

EC monolayer for 1 h. After several washings, cells were fixed and stained with a PE-CD11c antibody and visualized under fluorescence microscopy (40×). Adherent cells were

AC C

quantitated by averaging the number of cells from 5 separate microscopic fields. Data in A and B are mean ± s.e.m. (n=3). *p＜0.05, ** p＜0.01 and *** p＜0.001, two-way ANOVA.

Figure 4. JAML and DC-based tumor immunotherapy (A) E·G7 tumor-bearing mice (n=3 per group) were vaccinated with control IgG- or antiJAML antibody-treated OVA257-264 peptide-pulsed BMDCs, and tumor growth was monitored every 3 days (left). Tumors were surgically isolated and pictured from TB mice (n=3) on day - 29 -

ACCEPTED MANUSCRIPT 28 (right). Data are mean ± s.e.m., unpaired 2-way ANOVA, **P<0.01 and ***P<0.001. (B) CTL activity was assessed in mice (n=3 per group) vaccinated with IgG- or anti-JAML antibody-treated BMDCs. (C) OVA-specific CD8+ T-cells were assessed among the CD8+ T-

RI PT

cells isolated from the spleen of vaccinated mice (n=3 per group) by an OVA-specific tetramer assay as described in the Materials and Methods. Data in B and C are mean ± s.e.m..

SC

*P<0.05 and **P<0.01, Student t-test.

Figure 5. Human MoDCs do not express JAML protein but have normal TEM activity

M AN U

Immature and mature MoDCs were generated as described in the Materials and Methods. (A) JAML expression was assessed by RT-PCR, Western blot, and FACS analysis in monocytes, THP-1 cells, and immature (imMoDC) and mature MoDCs (mMoDC). (B) JAML expression was assessed by Western blot and flow cytometry during DC development from monocytes;

TE D

Mc-1D, monocytes on day 1 of DC development. (C) TEM activities of THP-1, imMoDCs, and mMoDCs were assessed using a TNF-α preactivated HUVEC monolayer system. Data

EP

are mean ± s.e.m. (n=3), Student t-test.

Figure 6. JAM1 homophilic interactions control human MoDC TEM

AC C

(A) JAM1, 2, 3, and 4 mRNAs in monocytes and imMoDCs were assessed by RT-PCR. These mRNA expression levels were reexamined by qRT-PCR with Fast SYBR_Green Master Mix kit. The expression level of each mRNA was calculated relative to that of GAPDH. (B) JAM1 mRNA expression was assessed in THP-1, THP-1-derived DCs, monocytes, imMoDCs, mMoDCs, and HUVECs. Surface expression of JAM1 was assessed by FACS analysis. (C) MoDCs and/or HUVEC monolayers were pre-treated with anti-JAM1; TEM activity of MoDCs was then assessed. (D) Monocytes were pretreated with anti-JAML - 30 -

ACCEPTED MANUSCRIPT and/or anti-JAM1 antibodies; TEM activity was then assessed in the HUVEC monolayer.

AC C

EP

TE D

M AN U

SC

RI PT

Data in C and D are mean ± s.e.m. (n=3). *p＜0.05 and ** p＜0.01, two-way ANOVA.

- 31 -

ACCEPTED MANUSCRIPT RNA imDC mDC

imDC mDC

JAML

10m

Merge

RI PT

imDC

60X

091014 JAML expression on BMDC im, TNF, IFN, TNF/IFN, LPS, LPS/IFN

10m

10m

2m

2m

2m

SC

10m

JAML

C BM

2D

4D

imDC

JAML β-actin Protein

D 4.16

12.8

66.2 JAML

4.83

anti-JAML-PE

2D

imDC

4D

mDC

E

mDC

IgG imDC mDC

65.7

IgG CD8a DC CD11b DC CD103 DC

20.1

BMDC In vivo DC

AC C

imDC

EP

β-actin

IgG-PE

R1 mDC

TE D

JAML

M AN U

BM cells

300X

mDC

mDC

091031 BM cells jaml

RNA

10m

10m

mDC

β-actin

CCR7

DAPI

JAML

60X

14 JAML expression on BMDC NF, IFN, TNF/IFN, LPS, LPS/IFN

B

Protein

imDC

A

JAML

Fig. 1

B

ACCEPTED MANUSCRIPT ** 30

*

**

20

IgG anti-JAML

30 20

*

0

2.88

49.2

*

20

0.2

60

IgG anti-JAML

40

20

EP % of TEM

Con IgG-PE

F

5000

IgG anti-JAML

4000

n.s

3000 2000 1000 0

imDC mDC

s.c. to iLN

* *

anti-JAML/DiD Con-IgG/CTV

0.25

10

0.21

mDC

JAML-PE

DiD

0 si-con si-jaml

10 (g/ml)

0.16

30

20

*

5

D

imDC mDC

imDC

si-jaml

β-actin

2

2.82

AC C

si-con

*

imDC

mDC

anti-JAML/DiD Con-IgG/CTV

0.57

No. of migrated cells / iLN

CCR7 JAML

*

SC

10

0

8.52

10 (g/ml)

5

RI PT

30

0

TE D

0.67

37.1

40

No. of TW migrated cells

IgG

45.7

2

IgG anti-JAML

M AN U

4.70

0.2

0

% of CCR7+ cells

mDC

64.9

13.1

MIP3(+)

+ mDC

imDC 14.2

MIP3(-)

-

% of TEM

+

MIP3(+)

MIP3(-)

MIP1(+)

MIP1(-)

MIP1(+)

-

E

*

mDC TEM

50

4.52

*

0

0

C

*

10

*

10

imDC

anti-JAML

% of TEM

40

TNF-

JAML

imDC TEM

40

MIP1(-)

% of transmigrated cells

A

1000

IgG anti-JAML

800 600 400

*

200 0

imDC mDC

CTV

Fig. 2

ACCEPTED MANUSCRIPT ImDC TEM

mDC TEM

40

% of TEM

20

** * ***

10 0

0

B

**

0.2

*

** *

***

2

***

20

DC/anti-JAML

0

0

**

***

0.2

imDC

40

AC C

30

***

*

**

5

***

(g/ml)

100m

100m

EP

50

**

DC/anti-JAML EC/anti-CAR

EC/anti-CAR

TE D

mDC

*

2

M AN U

DC&EC/IgG

*

10 (g/ml)

5

30

SC

30

DC/IgG DC/anti-JAML EC/anti-CAR DC/anti-JAML & EC/anti-CAR

50

DC/IgG DC/anti-JAML EC/anti-CAR DC/anti-JAML & EC/anti-CAR

# of CD11c+ cells/MS field

% of TEM

40

RI PT

A

20

**

**

*

* *

*

10

0

M JA L

G Ig

L M JA ti- AR an C & AR C tian L M JA tian gG .I on C L M JA ti- AR an & C AR C

tian

tian

. on

C

Immature BMDC

Mature BMDC

Fig. 3

12000

Untreated OVA-DC/IgG OVA-DC/anti-JAML

10000 8000

** ***

6000

Untreated

4000

OVA-DC/IgG

2000 0 7

10 13 16 19 22 25 28

Gated on Live+Single+IineageCD45+CD3+CD8+ Cells / Spleen

OVA-DC/IgG OVA-DC/anti-JAML

OVA-DC/IgG

30

*

20

FMO

5

10 E:T ratio

15

OVA-Tetramer

8.04

15 10

**

5 0

IgG Anti-JAML

EP

0

TE D

14.5

10

OVA-DC/anti-JAML

20

AC C

% of Cytotoxicity

40

M AN U

C

B

SC

Days after injection

% of OVA-CD8 + T-cells

0

50

OVA-DC /Anti-JAML

RI PT

**

Tumor Volume (mm3)

A

ACCEPTED MANUSCRIPT

Fig. 4

ACCEPTED MANUSCRIPT R2 R2

A

Monocyte

imMoDC

mRNA R1 R1

GAPDH

RI PT

R3 R3

JAML JAML

THP-1

Protein

JAML

SC

R3 R3

R2 R2

β-actin

R3 R3

M AN U

JAML

B

mMoDC

C

R3 R3

TEM of MoDC

20

MoDC

Mc-1D

R3 R3

AC C

EP

Monocyte

15

% of TEM

β-actin

TE D

JAML

10 5 0

THP-1

imDC

mDC

JAML

R3 R3

Fig. 5

ACCEPTED MANUSCRIPT

A

B 2.5

JAM1

Relative expression

JAM1

JAM4

1.5

* *

0

0

0.2

mMoDC

HUVEC

2

JAM1

*

** *

5

**

*

10 (g/ml)

monocyte TEM

IgG anti-JAM1 anti-JAML anti-JAML& JAM1

30

% of TEM

TE D

EP *

AC C

% of TEM

DC/IgG DC/anti-JAM1 HUVEC/anti-JAM1 DC & HUVEC/anti-JAM1

10

imMoDC

D

MoDC TEM

***

Monocyte

M AN U

JAM1 JAM2 JAM3 JAM4

JAM1 JAM2 JAM3 JAM4

MoDC

Monocyte

20

THP-mDC

0.5 0.0

30

THP-1 1.0

GAPDH

C

RI PT

JAM3

GAPDH

SC

JAM2

2.0

20

** *

**

10

*

** *

*

5

10 (g/ml)

0 0

0.2

2

**

Fig. 6

ACCEPTED MANUSCRIPT Highlights JAML is involved in the TEM of mouse bone marrow-derived dendritic cells (BMDC) JAML of BMDCs interacts with CAR of endothelial cells for the TEM of BMDCs. DC-based tumor immunotherapy was significantly impaired by JAML blockades.

RI PT

Human monocyte-derived DCs (MoDCs) do not express JAML, but have a TEM activity.

AC C

EP

TE D

M AN U

SC

MoDC TEM across the endothelial barrier needs the hemophilic interaction of JAM1.