• Aucun résultat trouvé

LIST OF FIGURES

Figure 5. Negative selection and central T cell tolerance

II. THESIS AIM

III.1. c. Specific aim

MHCII-mediated antigen-presenting functions of pDCs can be tolerogenic or immunogenic, depending on the immunological context (Guery and Hugues, 2013).

In one hand, as mentioned previously and as shown by several studies in human and mouse, pDCs associated to tumors (in tumors and tumor-draining LNs) are rendered tolerogenic by the TME (Fig. 15). Indeed, innate pDC functions are altered, with decreased production of IFN-I, and pDC antigen presentation promote the differentiation and/or expansion of Tregs, via the expression of immunosuppressive molecules, such as IDO or ICOS-L (Aspord et al., 2013;

Conrad et al., 2012; Faget et al., 2012; Hartmann et al., 2003; Labidi-Galy et al., 2011; Sharma et al., 2007; Sisirak et al., 2012).

On the other hand, multiple studies have demonstrated that the power of pDCs can be harnessed to mount potent anti-tumor responses (Aspord et al., 2012; Liu et al., 2008a; Loschko et al., 2011b; Tel et al., 2013a; Tel et al., 2012b). In tumor-bearing mice, distal LN pDCs can be activated by a contralateral vaccination with CpG-B along with MHC-II-restricted peptide, which enhances their MHC-II-mediated antigen-presenting functions, leading to anti-tumor immunity through Th17 cell priming (Fig. 15) (Guery et al., 2014).

The original aim of this chapter A was, consequently, to determine whether TA-pDCs that harbor a tolerogenic phenotype could undergo a tolerogenic-to-immunogenic reprogramming, by injecting directly into the tumor the same treatment as the one above-mentioned, used in the contralateral setting (Fig. 15 and 16). Our goals were both to better understand MHC-II-restricted antigen-presenting functions of pDCs, as they exhibit context-dependent features, and also to test this strategy as it appeared to us as a promising immunotherapeutic approach.

Therefore, we analyzed the effect of intratumoral (i.t.) injection of CpG-B and MHC-II-restricted peptide on tumor growth, adaptive immunity and pDC phenotype, with a special focus on their MHC-II-mediated antigen-presenting functions.

68

The second aim of this investigation was to characterize the effect of this i.t. strategy on other immune cells of the innate and adaptive compartments, besides pDCs.

Figure 16. Can intratumoral administration of CpG-B along with tumor antigenic peptide reverse the tolerogenic phenotype of tumor-associated plasmacytoid dendritic cells?

A. Plasmacytoid dendritic cells (pDCs) in the tumor and tumor-draining lymph nodes (TdLNs) are maintained in a tolerogenic state (purple) by the tumor microenvironment.

B. Does intratumoral delivery of the TLR9 ligand CpG-B together with tumor antigenic peptide induce a tolerogenic-to-immunogenic conversion of tumor-associated pDCs (red)?

cDC, conventional dendritic cell; CTL, cytotoxic T lymphocyte; Th, T helper; Treg, regulatory T cell.

Drawings adapted from Guéry and Hugues, Oncoimmunology, 2015 (Guery and Hugues, 2015a).

69

III.2. Results

Intratumoral CpG-B promotes antitumoral neutrophil, cDC, and T cell cooperation without reprogramming tolerogenic pDC

Marion Humbert 1, Leslie Guéry 1, Dale Brighouse 1, Sylvain Lemeille 1 and Stéphanie Hugues 1

1 Department of Pathology and Immunology, University of Geneva Medical School, Geneva, Switzerland

Published in Cancer Research (DOI: 10.1158/0008-5472.CAN-17-2549) Objective:

Diverse mechanisms are induced by cancer immunotherapies to potentiate the strength of the immune system, in order to get rid of tumor cells. These treatments include therapeutic vaccines, which aim at inducing an efficient immune response against tumors. Therapeutic vaccines are highly dependent on the tumor microenvironment (TME), in which several immune cells harbor great levels of plasticity, leading to various outcomes with regard to the anti-tumor response.

Plasmacytoid dendritic cells (pDCs) have an immunogenic phenotype under inflammatory conditions. However, the TME negatively impact pDC functions, enhancing their immune-suppressing properties.

The aim of this article was to characterize the impact of intratumoral administration of established tumors with CpG-B in presence or absence of a major histocompatibility complex class II-restricted tumor antigenic peptide on immune cells and tumor growth. This study was primarily focused on the effect of this locally-delivered therapeutic vaccine on tumor-associated pDC functions, particularly on their role as antigen-presenting cells. Nonetheless, it also analyzed the impact of this treatment on other innate and adaptive immune cells, including neutrophils, conventional DCs and T lymphocytes, and on the interactions between these cells.

Personal contribution:

For this article, I participated in the concept and design of the study, I developed the methodology, acquired and analyzed the data, and wrote the manuscript, under the supervision of Prof. Stéphanie Hugues. Dr. Leslie Guéry participated in the concept and design of the study, and in the acquisition and analysis of data. Dale Brighouse participated in data acquisition. Dr.

Sylvain Lemeille performed the RNA sequencing computational analysis.

Tumor Biology and Immunology

Intratumoral CpG-B Promotes Antitumoral Neutrophil, cDC, and T-cell Cooperation without Reprograming Tolerogenic pDC

Marion Humbert, Leslie Guery, Dale Brighouse, Sylvain Lemeille, and Stephanie Hugues

Abstract

Cancer immunotherapies utilize distinct mechanisms to harness the power of the immune system to eradicate cancer cells. Therapeutic vaccines, aimed at inducing active immune responses against an existing cancer, are highly dependent on the immunological microenvironment, where many immune cell types display high levels of plasticity and, depending on the context, promote very different immunologic outcomes.

Among them, plasmacytoid dendritic cells (pDC), known to be highly immunogenic upon inammation, are maintained in a tolerogenic state by the tumor microenvironment. Here, we report that intratumoral (i.t.) injection of established solid tumors with CpG oligonucleotides-B (CpG-B) inhibits tumor growth. Interestingly, control of tumor growth was independent of tumor-associated pDC, which remained

refractory to CpG-B stimulation and whose depletion did not alter the efcacy of the treatment. Instead, tumor growth inhibition subsequent to i.t. CpG-B injection depended on the recruitment of neutrophils into the milieu, resulting in the activation of conventional dendritic cells, subsequent increased antitumor T-cell priming in draining lymph nodes, and enhanced effector T-cell inltration in the tumor micro-environment. These results reinforce the concept that i.t.

delivery of TLR9 agonists alters the tumor microenvironment by improving the antitumor activity of both innate and adap-tive immune cells.

Signicance: Intratumoral delivery of CpG-B disrupts the tolerogenic tumor microenvironment and inhibits tumor growth.

Cancer Res; 78(12); 3280–92.2018 AACR.

Introduction

Tumor immunity is the result of complex interactions between different cells immersed in the tumor microenvironment.

Although substantially inhibited by the tumor, effector immune responses can take place in particular conditions or experimental settings. It is not surprising, therefore, that novel therapeutic strategies are aimed at boosting antitumor immune cell responses, in particular tumor-specic T-cell priming in lymph nodes (LN).

Plasmacytoid dendritic cells (pDC) are involved in both innate (1) and adaptive immunity (29), and can exhibit either a tolerogenic or immunogenic phenotype, depending on the immunologic context (10). Although the role of pDC in antitu-mor immunity is still debated, several studies have shown that tumor-associated pDC (TA-pDC), i.e.,in the tumor and tumor-draining lymph nodes (TdLN), exhibit a tolerogenic phenotype.

Indeed, TA-pDC are characterized by low type I interferon (IFN-I) secretion and costimulatory molecule expression and promote regulatory T-cell (Treg) induction (1114). Therefore, the

pre-sence of TA-pDC has been correlated to a bad prognosis, with little evidence showing their potential as efcient antigen-presenting cells (APC).

However, pDC can be activated at a site distal from the tumor and used as immunogenic APCs. Indeed, we previously showed that contralateral vaccination of mice with CpG oligonucleotides-B (CpG-oligonucleotides-B), a TLR9 agonist (15), together with a MHCII-restricted tumor antigen, leads to the activation of distal pDC that induce tumor-specic Th17 cell differentiation. Immune cells are conse-quently recruited to the tumor, and cytotoxic T cells (CTL) subsequently eliminate the tumor (16).

Aiming at reversing the tolerogenic phenotype of TA-pDC and other immune cells, by vaccinating directly at the tumor site, might be a promising approach. Indeed, several studies have been carried out using intratumoral (i.t.) delivery of different adju-vants, including CpG (17). Studies have used different classes of CpG administered intratumorally (i.t.), peritumorally, and intra-cranially (glioma/glioblastoma; refs. 1821). In addition, CpG has been injected systemically, vehicle in or in conjunction with molecules targeting the tumor, such as microparticles (22) or liposomes (23). Most of these studies have observed control of tumor growth, inltration of CD8þT cells, and decreased Treg numbers in the tumor (22, 2426). However, the mechanisms accounting for these observations have not been fully understood.

More importantly, CpG administration was, for the most part, combined with other treatments, such as cryosurgery, immune checkpoint blockade, adoptive transfer of tumor-specic T cells, or cytokine injection. Therefore, the intrinsic effects of i.t. CpG itself have not been deciphered.

Department of Pathology and Immunology, University of Geneva Medical School, Geneva, Switzerland.

Note:Supplementary data for this article are available at Cancer Research Online (http://cancerres.aacrjournals.org/).

Corresponding Author:S. Hugues, Geneva Medical School, Rue Michel-Servet 1, 1211 Geneva, Switzerland. Phone: 41-22-379-58; Fax: 41-22-379-57-46; E-mail:

Stephanie.Hugues@unige.ch

on June 18, 2018. © 2018 American Association for Cancer Research.

cancerres.aacrjournals.org Downloaded from

Here, we investigated whether i.t. administration of CpG-B alone or combined with a MHCII-restricted tumor antigen could affect tumor growth. We show that i.t. CpG-B injection in mice leads to signicant tumor growth inhibition that relies on T-cell responses, including an increased CTL inltration and decreased Treg numbers. The CpG-B effect was enhanced by the addition of a tumor peptide. In both cases, pDC were not reactivated and were not involved in the control of tumor growth, demonstrating that tolerogenic TA-pDC cannot be reprogrammed toward an immunogenic phenotype. In con-trast, this treatment led to robust TA-conventional DCs (cDC) activation. Moreover, the i.t. CpG-B delivery induced a massive i.t. inltration of activated neutrophils. These cells are therst responders to sites of acute tissue damage and infection. In settings of chronic inammation, neutrophils persist in tissues, and this has been associated with cancer progression (27).

However, the role of neutrophils in the tumor microenviron-ment remains controversial, with evidence for both pro- and antitumor roles (28). Our results show that i.t. neutrophil recruitment signicantly contributes to TA-cDC activation and antitumor T-cell responses following i.t. CpG-B administration, resulting in the control of tumor growth. Taken together, we suggest that intratumoral delivery of vaccines that include TLR9 agonists would benet current immunotherapy anticancer strategies by disrupting the tolerogenic tumor microenviron-ment and enhancing innate and adaptive tumor-associated immune cell responses.

Materials and Methods

Mice

Mice were of pure C57BL/6 background and were bred and maintained under specic pathogen-free conditions at Geneva Medical School animal facility or under specic and opportu-nistic pathogen-free conditions at Charles River, France or Italy.

Rag2/(29), BDCA2-DTR (30), and IFNAR/(31) mice have been described elsewhere. All animal husbandry and experi-ments were approved by and performed in accordance with the guidelines of the animal research committee of Geneva canton, Switzerland.

Tumor experiments

Ovalbumin-expressing tumor cell lines and EG7-OVA thymo-ma cells were obtained from S. Amigorena laboratory in 2011 and used at below passage 20. Cells were not authenticated and were tested negative for Mycoplasma.EG7-OVA cells were cultured at 37C and 5% CO2in complete RPMI medium [RPMI 1640 GlutaMAX with 10% heat-inactivated FBS, 1% penicillin/strep-tomycin (10,000 U/mL penicillin and 10 mg streppenicillin/strep-tomycin/mL), 1% sodium pyruvate 100 mmol/L, 0.1% 2-b-mercaptoethanol 50 mmol/L] supplemented with G418 (Mediatech; 0.4 mg/mL) for the selection of OVA-expressing cells. G418 was removed from the medium 1 day prior to tumor cell implantation. Tumor cells (1106in PBS) were injected s.c. into the leftank.

OVAII peptide (ISQAVHAAHAEINEAGR) was purchased from Polypeptide and CpG-B 1668 from Invivogen. Mice were injected i.t. once tumors were established, at indicated times (D7 to D12 after tumor implantation), with CpG-B (30mg), in the presence or not of OVAII(10mg), in PBS (40mL). Tumor size was measured with a caliper [L (length)l (width)] every 1 to 2 days over the indicated periods of time.

In vitroneutrophil migration assay

Tumor-conditioned media (TCM) were prepared as followed:

solid tumors were injected i.t. with PBS or CpG-B (30mg) 11 days after tumor implantation. Twenty-four hours later, tumors were excised, cut in small pieces (25 mg/mL), and incubated in com-plete RPMI at 37C and 5% CO2. Twenty-four hours later, tumor pieces were removed by centrifugation, and the supernatant, i.e., TCM, wasltered (0.22mm). In parallel, bone marrow cells were harvested from nave mice, followed by magnetic isolation of untouched neutrophils (Miltenyi Biotech). Bone marrow derived neutrophils were seeded in complete RPMI in the upper compartment (80% conuence on polycarbonate membrane with 8mm pores) of transwell chambers (Corning). The lower compartment contained either complete RPMI with indicated CpG-B concentrations or medium conditioned from EG7 tumors (TCM). After 24 hours of incubation at 37C and 5% CO2, cells were counted in the lower compartment.

pDC and neutrophil depletion

For the depletion of pDCin vivo, diphtheria toxin (DT; Sigma;

1mg/mL in 100mL of PBS) was injected i.p. in BDCA2-DTR mice, 1 day prior to i.t. CpG-BOVAII administration and every 3 to 4 days over the indicated period. Plasmacytoid DC depletion was analyzed in the tumor on the day of i.t. CpG-BOVAII

administration byow cytometry.

Monoclonal anti-Ly6G (1A8) antibody and Rat IgG2a iso-type control (2A3) were purchased from BioXcell and were injected i.v. (1 mg/mL in 100mL of PBS) and i.t. (2.5 mg/mL in 40mL of PBS), 1 day prior to i.t. CpG-B injection. Neutrophil depletion was analyzed byow cytometry in the blood and tumors 2 days after depleting antibody injection.

Preparation of single-cell suspensions

Tumors and TdLNs were excised from mice at indicated times.

Cell isolation from tumors was performed as follows: tumors were digested chemically with collagenase D (1 mg/mL; Roche) and DNAse I (10 mg/mL; Roche) and mechanically. Dead cells were eliminated using lympholyte M (Cedarlane Laboratory). For DC and neutrophil analyses and ow-cytometry cell sorting, TdLNs were subjected to the same treatment. For intracellular cytokine analyses, cells were restimulated for 4 hours at 37C and 5% CO2, in complete RPMI medium in the presence of phorbol 12-myristate 13-acetate (Sigma; 100 ng/mL) and ionomycin (Sigma; 1mg/mL). GolgiPlug solution (BD Biosciences; 1mL/mL) was added to the culture medium for the last 2.5 hours.

Blockade of T-cell egress from the LNs

Mice were injected i.p. with the sphingosine-1-phosphate receptor-1 antagonist FTY720 (Sigma; 100mg/mL in 200mL of PBS) 1 day prior to i.t. CpG-BOVAIIadministration, and every day within the time period indicated.

Flow cytometry

Anti-CD62L (MEL-14), anti-CD69 (H1.2F3), anti-FasL (MFL3), antiICAM-1 (YN1/1.7.4), anti-TCRb(H57-597), anti-CD4 (RM4-5), anti-CD8a(53-6.7), FOXP3 (FJK-16s), anti-ICOS (15F9), anti-CD25 (PC61.5), anti-CD103 (2E7), anti-IL17 (eBio17B7), anti-IFNg (XMG1.2), anti-CD11b (M1/70), anti Siglec-H (eBio927), antiPDCA-1 (eBio440c), anti-B220 (RA3-6B2), anti-Ly6C (HK1.4), and anti-F4/80 (BM8) monoclonal antibodies were purchased from Thermo Fisher Scientic, and Intratumoral CpG-B Enhances Immune Cell Functions

www.aacrjournals.org Cancer Res; 78(12) June 15, 2018 3281

on June 18, 2018. © 2018 American Association for Cancer Research.

cancerres.aacrjournals.org Downloaded from

anti-CD45 (30F11), anti-CD86 (PO3), antiI-Ab (AF6-120.1), anti-CD11c (N418), anti-CD40 (3/23), and anti-Ly6G (1A8) monoclonal antibodies and DRAQ7 were purchased from Biolegend.

Cells were stained with Fixable viability dye eFluor 780 (Thermo Fisher Scientic) for 30 minutes, at 4C, for cell analyses, or with DRAQ7 immediately before cell sorting.

Pentamer and intracellular staining were performed on sepa-rate single-cell suspensions due to technical incompatibilities.

H-2kb SIINFEKL pentamer (OVAI; Proimmune) staining was performed in accordance with the manufacturer protocol.

Single-cell suspensions were incubated with anti-CD16/32 Fcg RII-RIII, a Fc receptor monoclonal blocking antibody (Thermo Fisher Scientic), for 10 minutes, at 4C, before staining with antibodies.

Intracellular stainings (cytokines and Foxp3) were performed using the Intracellular Fixation and Permeabilization buffer set (Thermo Fisher Scientic).

Data were acquired using Gallios (Beckman Coulter) and analyzed using FlowJo Software (FlowJo, LLC). Flow cytometry cell sorting of tumor-associated cDC, pDC, and neutrophils was performed using MoFlo Astrios (Beckman Coulter). Doublets were excluded by gating on the cells along the diagonal in a FSC height/FSC area dot plot. Dead cells were excluded by gating on the live cells that have an intermediate uorescence intensity for the viability dye eFluor 780 (cell analyses) or by excluding the DRAQ7þcells (cell sorting). cDC were dened as CD45hi CD11chiF4/80and pDC as CD45hiCD11cintB220þSiglec-Hþ PDCA-1þ. Neutrophils were dened as CD45þCD11bþLy6Cint Ly6Ghi. In the experiments in which neutrophils were depleted using the anti-Ly6G antibody (1A8; BioXcell), neutrophils were dened as CD45þCD11bþLy6Cintas the use of the anti-Ly6G (1A8, same clone used forin vivodepletion) antibody forow cytometry would not have allowed a valid analysis of the depletion efciency.

RNA sequencing

Library preparation, sequencing, and read mapping to the reference genome. Flow cytometryisolated TA-pDC and TA-cDC (see above) were collected in RNAprotect Cell Reagent (Qiagen). RNA was isolated using an RNeasy Plus Micro Kit (Qiagen), and three to four replicates per condition were used. RNA integrity and quantity were assessed with a Bioanalyzer (Agilent Technologies).

cDNA libraries were constructed by the Genomic platform of the University of Geneva as follows: the SMARTer Ultra Low RNA Kit from Clontech was used for the reverse transcription and cDNA amplication according to the manufacturer's specications, starting with 1 ng of total RNA. cDNA (200 pg) was used for library preparation using the Nextera XT Kit from Illumina.

Library molarity and quality were assessed with the Qubit and Tapestation using a DNA High sensitivity chip (Agilent Technol-ogies). Pools of 10 libraries were loaded for clustering on a single-read Illumina Flow cell. Reads of 50 bases were generated using the TruSeq SBS chemistry on an Illumina HiSeq 4000 sequencer.

FastQ reads were mapped to the ENSEMBL reference genome (GRCm38.89) using STAR version 2.4.0j (https://github.com/

alexdobin/STAR) with standard settings, except that any reads mapping to more than one location in the genome (ambiguous reads) were discarded (m¼1). Sequences data have been sub-mitted to the GEO database under the accession number GSE112206.

Unique gene model construction and gene coverage reporting. A unique gene model was used to quantify reads per gene. Briey, the model considers all annotated exons of all annotated protein coding isoforms of a gene to create a unique gene where the genomic region of all exons are considered coming from the same RNA molecule and merged together.

RNA-seq analysis.All reads overlapping the exons of each unique gene model were reported using featureCounts version 1.4.6- p1 (http://bioinf.wehi.edu.au/featureCounts/). Gene expressions were reported as raw counts and in parallel normalized in reads per kilobase million (RPKM) in order tolter out genes with low expression value (1 RPKM) before calling for differentially expressed genes. Library size normalizations and differential gene expression calculations were performed using the package edgeR (http://bioconductor.org/packages/release/bioc/html/edgeR.html) designed for the R software (http://www.R-project.org/). Only genes having a signicant fold change (BenjaminiHochberg cor-rectedPvalue<0.05) were considered for the rest of the RNA-seq analysis.

Gene ontology and/or KEGG analysis.Gene ontology (GO) term and Kyoto Encyclopedia of Genes and Genomes (KEGG) meta-bolic pathways enrichment were performed using homemade scripts for the R software.

GSEA: pathway enrichment.Gene sets were selected according to their known involvement in either "Antigen presentation,"

"Immunogenicity," or "DC migration" (see Supplementary Table S1). Genes were ranked by their calculated fold changes (decreas-ing rank(decreas-ing). A gene set analysis us(decreas-ing the gene set enrichment analysis (GSEA) package Version 2.2 (http://software.broadinsti tute.org/gsea/index.jsp; ref. 32) from the Broad Institute (MIT, Cambridge, MA) was used to analyze the pattern of differential gene expression between the two groups. Gene set permutations were performed 1,000 times for each analysis. The normalized enrichment score (NES) was calculated for each gene set.

RT-qPCR

Flow cytometrysorted tumor-associated neutrophils (TAN;

see above) were collected in RNAprotect Cell Reagent (Qiagen).

Total RNA was isolated using the RNeasy Plus Micro Kit (Qiagen), and RT-qPCR was performed as described (4). cDNA was synthesized with random hexamers and M-MLV Reverse transcriptase (Promega). PCR was performed with CFX Connect Real-time System Rad) and iQ SYBR green Super-mix (Bio-Rad). HPRT mRNA was used for normalization. Primer sequen-ces used were as follows: HPRT, forward, 50 -GAGGAGTCCTGTT-GATGTTGCCAG-30and reverse, 50 -GGCTGGCCTATAGGCTCA-TAGTGC-30; CCL3, forward, 50-TACAGCCGGAAGATTCCACG -30and reverse, 50-GTCTCTTTGGAGTCAGCGCA-30; CCL4,

Total RNA was isolated using the RNeasy Plus Micro Kit (Qiagen), and RT-qPCR was performed as described (4). cDNA was synthesized with random hexamers and M-MLV Reverse transcriptase (Promega). PCR was performed with CFX Connect Real-time System Rad) and iQ SYBR green Super-mix (Bio-Rad). HPRT mRNA was used for normalization. Primer sequen-ces used were as follows: HPRT, forward, 50 -GAGGAGTCCTGTT-GATGTTGCCAG-30and reverse, 50 -GGCTGGCCTATAGGCTCA-TAGTGC-30; CCL3, forward, 50-TACAGCCGGAAGATTCCACG -30and reverse, 50-GTCTCTTTGGAGTCAGCGCA-30; CCL4,