Thesis Reference

Thesis

Modulation of peripheral T cell responses by unconventional antigen-presenting cells: role of plasmacytoid dendritic cells in anti-tumor immunity and lymphatic endothelial cells in autoimmunity

HUMBERT, Marion

Abstract

The aim of this thesis was to characterize the role of plasmacytoid dendritic cells (pDCs) and lymphatic endothelial cells (LECs) in the modulation of peripheral T cell responses, with a particular focus on MHC class II-restricted antigen presentation. Depending on the immunological context, pDC MHCII-mediated antigen-presenting functions can be either tolerogenic or immunogenic. For instance, pDCs are maintained in a tolerogenic state by the tumor microenvironment. In the first part of this thesis, we asked the question whether tumor-associated pDCs could undergo a tolerogenic-to-immunogenic reprogramming following the intratumoral administration of CpG-B, a TLR9 ligand, along with a model MHC-II-restricted tumor antigenic peptide. LECs from lymph nodes (LN-LECs) were shown to impact peripheral CD8+ T cell responses, as antigen-presenting cells (APCs). Emerging evidence is in favor of a role for LN-LECs in CD4+ T cell tolerance to MHC-II-restricted antigens, although this phenomenon is still a matter of debate. In the second part of this thesis, we sought to determine the contribution of LN-LECs as MHC-II-restricted APCs [...]

Reference

HUMBERT, Marion. Modulation of peripheral T cell responses by unconventional antigen-presenting cells: role of plasmacytoid dendritic cells in anti-tumor immunity and lymphatic endothelial cells in autoimmunity. Thèse de doctorat : Univ. Genève, 2019, no. Sc. 5308

DOI : 10.13097/archive-ouverte/unige:115554 URN : urn:nbn:ch:unige-1155544

Available at: http://archive-ouverte.unige.ch/unige:115554

Disclaimer: layout of this document may differ from the published version.

1 / 1

UNIVERSITÉ DE GENÈVE

Département de Biologie Cellulaire FACULTÉ DES SCIENCES Professeur Jean-Claude Martinou

Département de Pathologie et Immunologie FACULTÉ DE MÉDECINE Professeure Stéphanie Hugues

Modulation of peripheral T cell responses by unconventional antigen-presenting cells

Role of plasmacytoid dendritic cells in anti-tumor immunity and lymphatic endothelial cells in autoimmunity

THÈSE présentée à la Faculté des Sciences de l’Université de Genève pour obtenir le grade de Docteur ès Sciences, mention Biologie

par Marion HUMBERT de Échirolles (France)

Thèse n° 5308

Atelier d’impression ReproMail Genève 2019

À Valentina

À mes parents et à mon frère

« Toutes les grandes personnes ont d’abord été des enfants, mais peu d’entre elles s’en souviennent. »

“All grown-ups were once children, but only few of them remember it.” Antoine de St Exupéry

---

« Tout le monde est un génie. Mais si vous jugez un poisson sur ses capacités à grimper à un arbre, il passera sa vie à croire qu’il est stupide. »

“Everybody is a genius. But if you judge a fish by its ability to climb a tree, it will live its whole life believing that it is stupid.”

Albert Einstein (attributed to)

---

« La seule voie qui offre quelque espoir d'un avenir meilleur pour toute l'humanité est celle de la coopération et du partenariat. »

“The only path that provides hope for a better future for all humanity is cooperation and partnership.”

Kofi Annan

ABSTRACT

Plasmacytoid DCs are professional producers of type I interferon and also possess the ability to present antigen to T cells. These cells are extremely plastic and, depending on the immunological context, their MHCII-mediated antigen-presenting functions can be either tolerogenic or immunogenic. For instance, pDCs are maintained in a tolerogenic state by the tumor microenvironment (TME). However, the power of pDCs that are not immersed in the TME can be harnessed to mount potent anti-tumor responses. In tumor-bearing mice, distal lymph node (LN) pDCs can be activated by a contralateral vaccination with the TLR9 ligand CpG-B, along with a model MHC-II-restricted tumor antigenic peptide, enhancing their MHC-II-mediated antigen-presenting functions and leading to anti-tumor immunity through Th17 cell priming. We asked the question whether tumor-associated pDCs could undergo a tolerogenic-to-immunogenic reprogramming following the intratumoral administration of the above-mentioned treatment. Although it led to tumor growth control, this local treatment did not reverse the tolerogenic phenotype of pDCs. These cells remained refractory to the treatment and therefore did not contribute to its efficacy. On the contrary, the working model we propose rather involves cooperation between neutrophils, conventional DCs and T cells that leads to tumor growth control.

LECs, a subset of LN stromal cells (LNSCs), long thought to function as simple scaffolds, were recently shown to indirectly affect T cell responses in many ways. In addition, LN-LECs have the ability to directly impact peripheral T cell responses as unconventional APCs. LN-LECs present MHC-I-restricted endogenously-expressed peripheral tissue-restricted antigens (PTAs) to CD8+ T cells, inducing their elimination by clonal deletion. Emerging evidence is in

favor of a role for LNSCs in CD4+ T cell tolerance to MHC-II-restricted PTAs, although this phenomenon is still a matter of debate. We recently showed that elderly naïve mice, in which MHC-II expression was abrogated in LNSCs, presented signs of spontaneous autoimmunity. Here, I report our recent findings regarding the contribution of LECs as MHC-II-restricted APCs to autoreactive CD4+ T cell responses in experimental autoimmune encephalomyelitis (EAE), a mouse model for multiple sclerosis. The selective genetic ablation of MHC-II in LECs exacerbated disease severity. In contrast, the genetic enforcement of MHC-II-restricted myelin-derived peptide presentation in LECs led to delayed onset and dampened EAE severity, accompanied by a drastic decrease in the spinal cord CD4+ T cell infiltrate and a modulation of these cells towards a tolerogenic phenotype. The CD4+ T cell numbers and phenotype in the LNs draining the site of EAE immunization were unaffected, suggesting that this pathway may inhibit effector T cell responses, rather than the priming of naïve T cells in LNs. Our results highlight a role for LECs as unconventional tolerogenic APCs in EAE.

Altogether, our studies highlight important contributions of unconventional APCs in shaping peripheral T cell responses. Therefore, these unusual pathways need to be considered in the development of therapies aiming at modulating immune responses in disease development.

RÉSUMÉ

Afin de générer des réponses immunitaires efficaces tout en évitant des effets secondaires tels que l’auto-immunité, un équilibre entre activation et tolérance doit être finement régulé. Les réponses médiées par les lymphocytes T, qui jouent un rôle central dans l’immunité adaptative, sont modulées par les cellules présentatrices d’antigènes (CPA). En plus des CPA professionnelles, plusieurs types de cellules ont la capacité de cross-présenter des antigènes par le complexe d’histocompatibilité majeur de classe I (CMH-I) aux lymphocytes T CD8+ et/ou de présenter des antigènes par le CMH-II aux lymphocytes T CD4+, en tant que CPA non-conventionnelles. Selon le contexte et le type de CPA, les conséquences sur les réponses lymphocytaires T sont variables. L’objectif de cette thèse était de caractériser le rôle des cellules plasmacytoïdes dendritiques (pDC) et des cellules endothéliales lymphatiques (LEC), en tant que CPA non-conventionnelles, dans la modulation des réponses lymphocytaires T périphériques, avec un intérêt particulier pour la présentation d’antigènes restreinte par les molécules du CMH-II.

Les pDC sont des cellules sécrétrices d’interféron de type I professionnelles, qui ont également la capacité de présenter des antigènes aux lymphocytes T. Ces cellules ont des propriétés très plastiques et leurs fonctions de présentation d’antigènes restreinte par les molécules du CMH-II peuvent conduire à des effets tolérogènes ou immunogènes, selon le contexte. Les pDCs associées aux tumeurs (TA-pDC) sont maintenues dans un état tolérogène par le micro-environnement tumoral (MET). Cependant, le potentiel immunogène des pDC qui ne sont pas immergées dans le MET peut être optimisé afin de générer des réponses immunitaires anti- tumorales efficaces. Dans des souris porteuses de tumeurs, les pDC des ganglions lymphatiques distaux peuvent être activées par une vaccination contralatérale avec du CpG-B, un ligand du TLR9, avec un peptide dérivé d’un antigène tumoral modèle présenté par les molécules du CMH- II. Cette vaccination augmente la capacité de présentation d’antigènes restreinte par les molécules du CMH-II par les pDC, conduisant à une réponse immunitaire anti-tumorale médiée par les lymphocytes Th17. Dans la première partie de cette thèse, nous avons posé la question de savoir si les TA-pDC tolérogènes pouvaient subir une reprogrammation et adopter un phénotype immunogène après administration intra-tumorale du traitement mentionné ci-dessus. Bien qu’il induise une réduction de la croissance tumorale, ce traitement n’a pas donné lieu à une reprogrammation des TA-pDC. Ces cellules restent réfractaires au traitement et ne contribuent pas à son efficacité. Au contraire, le modèle que nous proposons implique une coopération entre les neutrophiles, cellules

vii

dendritiques conventionnelles et lymphocytes T, conduisant à une inhibition de la croissance tumorale.

Il a été récemment démontré que les LECs, un sous-type de cellules stromales des ganglions lymphatiques (LNSCs), considérées pendant longtemps comme de simples éléments structurels, affectent les réponses lymphocytaires T périphériques de plusieurs manières. Les LECs des ganglions lymphatiques ont notamment un impact direct sur ces réponses, en tant que CPA non- conventionnelles. Les LEC des ganglions lymphatiques présentent des antigènes restreints aux tissus périphériques (PTA), exprimés de manière endogène, aux lymphocytes T CD8+ par le CMH-I, induisant la délétion clonale de ces derniers. De plus en plus de travaux sont en faveur d’un rôle des LNSC dans la tolérance périphérique envers les lymphocytes T CD4+ spécifiques de PTA restreints au CMH-II, même si ce sujet reste matière à débat. Notre groupe a récemment montré que les souris naïves âgées dans lesquelles l’expression du CMH-II est abrogée dans les LEC présentent des signes d’auto-immunité spontanée. Dans la seconde partie de cette thèse, je décris nos résultats concernant la contribution des LEC en tant que CPA pour la présentation d’antigènes restreints au CMH-II dans la modulation des réponses médiées par les lymphocytes T CD4+ auto-réactifs dans l’encéphalomyélite autoimmune expérimentale (EAE), un modèle murin de sclérose en plaque. L’ablation génétique du CMH-II dans les LECs conduit à une exacerbation de la sévérité de l’EAE. Au contraire, lorsque l’on force, par modification génétique, la présentation d’un péptide dérivé de la myéline et restreint au CMH-II par les LECs, l’apparition de la maladie est retardée et son développement moins sévère. Ceci est accompagné par une réduction importante de l’infiltrat de lymphocytes T CD4+ dans la moëlle épinière et par une modulation de ces lymphocytes, conduisant à un phénotype tolérogène. Les comptes et le phénotype des lymphocytes T CD4+ ne sont, cependant, pas modifiés dans les ganglions lymphatiques drainant le site d’immunisation, suggérant que cette voie pourrait inhiber les réponses lymphocytaires T effectrices, plutôt que d’affecter l’activation des lymphocytes T naïfs dans les ganglions. Ces résultats démontrent un rôle pour les LEC en tant que CPA non-conventionnelles tolérogènes dans l’EAE.

Globalement, cette thèse souligne une contribution importante des CPA non-conventionnelles dans la modulation des réponses lymphocytaires T périphériques. Par conséquent, ces voies inhabituelles doivent être prises en compte lors du développement de thérapies ayant pour but la modulation des réponses immunitaires.

viii

ACKNOWLEDGEMENTS

I would like to express my deepest gratitude to Prof. Cornelia Halin (Zurich, Switzerland) and Prof. Vassili Soumelis (Paris, France) for accepting to evaluate my thesis. It is a great honor for me that you take time to be part of my jury. I also wish to thank Prof. Jean-Claude Martinou for accepting to co-supervise and evaluate this thesis.

Of course, I would like to warmly thank Prof. Stéphanie Hugues. Steph, thank you for giving me the opportunity to work in your lab, for supervising and guiding me. Thanks as well for all the corrections of this long thesis! Thanks for trusting me for the writing of manuscripts and for giving me autonomy with experiments. Thank you as well for giving me the opportunity to write in reviews and to collaborate with other labs. I was always impressed by your scientific qualities, by your energy, and all the projects (scientific and others) that you take over. Finally, thanks for the lab dinners at your place or elsewhere, and for being understanding in tough moments.

I wish to sincerely thank Dr. Nathalie Bendriss-Vermare (Lyon, France) for our scientific discussions and her precious insights regarding the project on tumor-associated pDCs. She has been of great help on several points (pDCs, but also neutrophils, cDCs, etc…). I also would like to sincerely thank Prof. Fabienne Tacchini-Cottier (Lausanne, Switzerland) for giving me technical advises that allowed me to optimize the experimental design for the in vitro study of neutrophils.

I wish to thank Prof. Taija Mäkinen (Uppsala, Sweden) and Prof. Ari Waisman (Mainz, Germany) for the Prox1-creERT2 and IiMOGfl/fl strains.

I would like to deeply thank Prof. Carlo Chizzolini and Prof. Monique Gannagé, my thesis godparents, as well as Prof. Dominique Soldati-Favre and Prof. Patrick Linder. Thank you for supporting me and helping me when I needed it most.

I also would like to thank Prof. Camilla Bellone and Prof. Doron Merkler for accepting to evaluate my “hors-thèse” manuscript.

I am thankful to Prof. Claire-Anne Siegrist, and to her whole lab, for their precious advises given during our joint lab meetings. In particular, a great thanks to Beatris, Elodie, Floriane, David and Maria, for their scientific and technical advises, but also for helping me in my numerous reagent and antibody hunts!

I also wish to thank Prof. Walter Reith, as well as his whole lab, for our joint lab meetings and their help. A big thanks to my dear catalan friend, Adrià, for your friendship, your kindness and your support. I am thankful to and have very special thoughts for Isabelle. I did not know you very well, but I remember you as warm and always smiling, but also as very brave, with an impressive strength of character. I remember you as an admirable person… I will never forget this sentence you chose: “Life is not about waiting for the storm to pass, it is about learning to dance in the rain”.

I would like to thank Dr. Yalin Eimre, Dr. Adama Sidibe, Dr. Sylvain Lemeille and Dr. Walter Ferlin for helping me in the achievement of the project on tumor-associated pDCs. Yalin, thank you for giving me advises regarding neutrophil experiments. Adama, thanks a lot for helping me in the experimental design of the in vitro study of neutrophils, as well as for your friendliness and support. Sylvain, thank you for the RNA-Seq computational analysis, for your rapidity, and for your patience with us going back and forth in your office, with our numerous questions. Walter, many thanks for critically reading and giving your opinion on the manuscript, and thank you for your warm welcome for our lab dinners!

I wish to thank the people from the iGE3 platform, for the RNA-seq, and people from the animal facility for their help, especially Jenny and the two Anthony.

I would like to warmly thank the schtroumpf team Jean-Pierre, Cécile and Grégory from the flow cytometry platform. I learned a tremendous amount of technical tricks regarding FACS with you, and you have been of great help for many things (fixing problems with the machines, or with analyses, helping to optimize antibody panels, spending hours sorting cells, etc…). I also really appreciated the nice atmosphere of the platform. Thanks a lot, you are amazing (even Greg’s weird jokes)!

Of course, I want to thank the Hugues’s lab. Collectively, many thanks fo all the help, especially with big experiments, for the “lab parties” and for the “chikichiki”. To the former lab. Fernanda, although our “cohabitation” in the lab was very brief because you left just after I arrived, thank you for your advises. Leslie, many thanks for initiating the project on tumor-associated pDCs. Thank you for teaching techniques under the hood, at the bench and for analyzes, as well as for all the help. Thank you also for your kindness. A special thanks to Anjie. Thanks for your welcome on my first day in the lab, for your help, especially for teaching me techniques regarding EAE and genotyping. Most importantly, thank you for your moral support, for your friendship, and for the shared moments. Thanks for your sensitivity, for your dark humor, thank you as well for making me discover another Geneva, and also for providing me a roof. However, I will not thank you for going so far away, because I have missed you! Juan, who also left to faraway lands, thanks for your advises and for answering my theoretical and technical questions regarding a bit of everything, as well as for making me discover this delicious Spanish soup with ice cream and champagne. Carla, thank for replying my numerous questions and for all the help. Thanks for our political debates, for (real) lab parties you initiated, at your place or elsewhere, and thank you for your generosity. To the « heart of the lab ». Dalito/Dalicious, a huge thanks to you for all the technical help. Thank you as well for our musical discussions under the hood, for your mini-concerts just for the lab, your (very) dark humor (cf. Anjie), for your politeness, your diplomacy and for being so cool. It was a real pleasure working with you, don’t ever change! To the new lab. Olga, thanks for your help, for your little greek dishes and your amazing cooking skills, for your generosity and your funny expression (mais elle est énorme !), with a specific acknowledgement for your tee-shirts with the French-speaking sentences. Camcam/Tranquillou, thanks for being the “guinea pig” allowing me to test my teaching skills! Thank you, of course, for our collaboration and all the help, as well as for your refreshing honesty and spontaneity. Guillaume, thanks for your help and your support. Thank you for reading and giving your opinion on my paper, for reading my “hors-thèse” manuscript as a Neuroimmunology and stress “grand manitou”, and for the Nancy macarons! Laure, thank you for all the help, for replying all my questions, for reading my paper and giving your opinion to x

ameliorate it. Thank you as well for your singing southern accent, and for organizing lab activities or dinners. I also would like to thank the students that were in the lab for various length of stay; thanks to Carla G. for your help with “lab parties”, thanks to Antoine for your delicious desserts, and thanks as well to Julien, Clara, Hugo et Mathilde. Finally, thanks to Déborah, Romane and Sonia, it was a pleasure supervising you for your internships.

I wish to thank the people working (scientists and administrative staff) in the PATIM department. Thanks for advises and ideas during progress reports, for protocol exchanges and technical advises. It was also nice exchanging with you, during department retreats, Christmas parties, Immunology day, PATIM happy hours, or in the CMU corridors. I want to thank some people in particular. Thanks to Jen, JF, Filippo and Aurélie, it was a pleasure sharing the little office with you, and particularly to Mahdia. Thank you for your support, your smile and your warmth, you will stay the mum of the office, even if you are on the other side of the wall now. Thanks to Stéphane as well, for your kindness and your weird humor.

I would like to thank some of the people I met on the other side of rue Lombard; Laura, Lyssia et Natacha. Petit moineau, it was a pleasure being your lab neighbor. Lyssia, thank you for everything you taught me in the lab, and for your support and our discussions. Chacha, what would have I done without you? Thank you for your scientific and technical advises, as well as for the thesis writing. Thanks for being my running buddy, for our political and philosophical discussions, for your moral support, for always being there and for following me in CMU (I am sure all of this was planned!). Thanks as well to Pifou, for your joie de vivre, and for our debates about everything and anything.

I wish to thank my colleagues and friends from CMU. My thoughts first go to Valentina. Vale, you will forever have a special place in my heart, and I am deeply thankful for the time we spent together. I want to thank the PhD students I met for the selection rounds of the Biology-medicine program. A big thanks to Nico, Caroline, Claire, Vanessa, Amy, Lisa et Sunil. Time’s up, freezing cold, youth hostel, etc… So many things happened since then. We have laughed and cried together. I am really happy for sharing this “doctoral” period of my life with you. A particular thanks to Sunil and Amy. Thanks to both of you, for all the moments we shared together, and for always being there. There are so many things I want to thank you for that I don’t even know what to say! Thank you as well to Lingzi, Aleksandra and Ebru. Finally, I also want to thank Manon, Doro, Julie, Nico H., Damian, Loïc, Dani, Ines, Fatma, Marta, Soner, Hugo, Salva, Ronke, Aurélia, Anne-Laure, Alex, Piango, Tanja et Alex-le-grand, for the nice moments spent together in CMU, in café de la pointe, for TGIFs, pic-nics in perle du lac, etc.. I am thankful that I have met you all, and for being in this international atmosphere that enriched my life, opened my mind, widened my horizons and taught me to put things into perspective. ,Dankjewel , ش كرا ,Thank you all: धꅍयवाद, merci, dziękuję, danke, gracias, o ṣeun, grazie 謝謝, obrigado, σας ευχαριστώ, sağol, tualumba ! It is going to be really hard for me to say good bye to you guys.

I also would like to thank Clara, Karin, Julia, Tiphaine et Mirjam, who allowed me escaping from the thesis and the world of Biology PhD students from time to time.

I wish, finally, to thank my relatives and old friends, who contributed greatly to this thesis through their support. Thanks a lot for many things, including your understanding for my chronic lack of availability. I will be very brief because I don’t really know how to express my gratitude to you.

To my dear old friends, many thanks for your friendship throughout all these years and your great support during my very long studies. I could write tones of pages about you, but I don’t think I need a thesis acknowledgement section for you to know how important you are to me. A huge thanks to Laëtitia, Gaëlle, Marie, Lise, Cécile et Julie-Anne.

To my extended family, thank you a lot for your support during all these years. We don’t see each other very often but it does not prevent me from thinking about you and appreciating the time spent together.

Last but not least, I want to dedicate this thesis to my parents and my brother. Thanks to Sylvain. And, thanks to my parents, for your support, your love and everything you have done for me since I am a child.

xii

REMERCIEMENTS

Je souhaite exprimer ma profonde gratitude aux Prof. Cornelia Halin (Zurich, Suisse) et Prof. Vassili Soumelis (Paris, France) d’avoir accepté d’évaluer ma thèse. C’est un honneur pour moi que vous preniez de votre temps pour faire partie de mon jury. Je remercie également Prof. Jean-Claude Martinou d’avoir accepté de codiriger et d’évaluer cette thèse. Bien entendu, je remercie très sincèrement Prof. Stéphanie Hugues. Steph, merci de m’avoir accueillie dans ton labo, de m’avoir encadrée et guidée, et d’avoir toujours été là pour la moindre question. Merci également pour toutes les corrections de cette longue thèse ! Merci de m’avoir fait confiance pour la rédaction des manuscrits et de m’avoir laissé de l’autonomie pour les expériences. Merci de m’avoir donné l’opportunité d’être impliquée dans la rédaction de revues, et également dans des collaborations avec d’autres labos. J’ai toujours été impressionnée par tes qualités en tant que scientifique, par ton énergie et ton dynamisme, et par la quantité de projets, scientifiques et autres, que tu mènes de front. Finalement, merci pour les soirées labo chez toi ou ailleurs, et pour ta compréhension dans les moments difficiles.

Je voudrais vivement remercier Dr. Nathalie Bendriss-Vermare (Lyon, France), pour nos discussions et ses précieux conseils concernant le projet pDC associées aux tumeurs. Ils m’ont été d’une grande aide, sur beaucoup de points (pDC, mais aussi neutrophiles, cDC, etc…). Je souhaite également remercier Prof. Fabienne Tacchini-Cottier (Lausanne, Suisse), pour ses conseils techniques qui m’ont permis d’optimiser le design des expériences pour l’étude des neutrophiles in vitro. Je remercie aussi Prof. Taija Mäkinen (Uppsala, Suède) et Prof. Ari Waisman (Mayence, Allemagne) pour les souches Prox1-creERT2 et IiMOGfl/fl. Je souhaite remercier Prof. Carlo Chizzolini et Prof. Monique Gannagé, qui ont parrainé ma thèse, ainsi que Prof. Dominique Soldati-Favre et Prof. Patrick Linder. Merci de m’avoir aidée et soutenue lorsque j’en ai eu besoin. Je voudrais également remercier Prof. Camilla Bellone et Prof. Doron Merkler d’avoir accepté d’évaluer mon travail de hors-thèse. Je remercie Prof. Claire-Anne Siegrist, et toute son équipe, pour leurs précieux conseils apportés lors de nos réunions communes. Un grand merci en particulier à Beatris, Elodie, Floriane, David et Maria, pour leurs conseils scientifiques et techniques, mais aussi de m’avoir aidée dans mes nombreuses chasses aux anticorps et autres réactifs ! Je remercie aussi Prof. Walter Reith, ainsi que toute son équipe, pour nos réunions en commun et pour leur aide. Un merci particulier à Adrià, mon cher ami catalan, pour ton amitié, ta gentillesse et ton soutien. Je remercie et j’ai une pensée très particulière pour Isabelle. Je ne te connaissais pas très bien mais je garde de toi le souvenir d’une personne souriante et très chaleureuse, mais aussi très courageuse, avec une impressionnante force de vie. Une personne admirable… Je me rappellerai toujours de cette phrase que tu as choisie : « La vie, ce n’est pas d’attendre que l’orage passe, c’est d’apprendre à danser sous la pluie ». Je souhaite remercier Dr. Yalin Eimre, Dr. Adama Sidibe, Dr. Sylvain Lemeille et Dr. Walter Ferlin de m’avoir aidée dans l’aboutissement du projet pDC associées aux tumeurs. Yalin, merci pour tes conseils concernant les neutrophiles. Adama, merci de m’avoir aidée dans le design des expériences pour l’étude des neutrophiles in vitro, et également pour tes encouragements et ta xiii

sympathie. Sylvain, merci pour l’analyse informatique des données obtenues par RNA-Seq, pour ta rapidité, et ta patience avec nos allées et venues dans ton bureau et nos multiples questions. Walter, merci d’avoir lu et donné ton avis sur le manuscrit afin de l’améliorer, et merci pour ton accueil chaleureux lors de nos soirées de labo ! Je voudrais remercier le personnel de la plateforme iGE3, pour le RNA-seq, ainsi que le personnel de la zootechnie, en particulier Jenny et les deux Anthony, pour leur aide. Je remercie très chaleureusement l’équipe des schtroumpfs Jean-Pierre, Cécile et Grégory de la plateforme de cytométrie en flux. J’ai énormément appris à vos côté au niveau technique, et vous avez été d’une immense aide que ce soit pour résoudre des problèmes avec les machines, pour optimiser des panels, pour les tris, pour des problèmes d’analyse, etc… J’ai aussi beaucoup apprécié l’ambiance bon enfant très agréable qui règne sur la plateforme. Merci mille fois, vous êtes géniaux (même les blagues douteuses de Greg) ! Bien entendu, je souhaite remercier le labo Hugues. D’une manière générale, merci pour l’aide apportée, surtout lors des grosses expériences, les « lab parties » et « chickichicki ». A l’ ancienne équipe. Fernanda, même si notre « cohabitation » dans le labo a été très brève puisque tu es partie juste après mon arrivée, merci pour tes conseils. Leslie, merci d’avoir initié le projet pDC associées aux tumeurs, et de m’avoir formée sous la hotte, à la paillasse et pour les analyses, ainsi que pour toute l’aide que tu m’as apportée. Et merci également pour ta gentillesse. Un merci très particulier pour Anjie. Merci de m’avoir accueillie lors de mon premier jour au labo et pour l’aide que tu m’as apportée, notamment pour me former aux techniques en rapport avec l’EAE et les génotypages. Mais surtout, merci pour ton amitié et ton soutien moral. Merci pour ta sensibilité, ton engagement et ton humour noir, merci de m’avoir fait découvrir ta ville en tant que bonne Genevoise que tu es, et également de m’avoir fourni un toit. Pour autant, je ne te remercie pas d’être partie si loin, car tu m’as beaucoup manquée ! Juan, toi aussi parti vers des contrées lointaines, merci pour tes conseils et d’avoir répondu à mes questions théoriques et techniques concernant un peu tout et n’importe quoi. Merci de m’avoir fait découvrir cette fameuse soupe espagnole à la glace et au champagne ! Carla, merci d’avoir répondu à mes nombreuses questions et pour toute l’aide que tu m’as apportée. Merci pour nos débats politiques enflammés, les soirées que tu as initiées, chez toi ou ailleurs, et surtout, merci pour ta générosité. To the « heart of the lab ». Dalito/Dalicious, un énorme merci à toi pour toute l’aide technique que tu m’as apportée. Merci également pour nos discussions musicales pendant les manips sous la hotte, pour tes mini-concerts rien que pour le labo, pour ton humour (très) noir (cf. Anjie), pour ta politesse, ta diplomatie, et ta décontraction. Ce fut un réel plaisir de travailler avec toi, ne change surtout pas ! A la nouvelle équipe. Olga, ma voisine de bureau pendant quelques temps, merci pour ton aide, pour tes petits plats grecques et tes grands talents culinaires, ta générosité, et tes expressions rigolotes (mais elle est énorme !), avec une mention spéciale pour tes tee-shirts francophones. Camcam/Tranquillou, merci d’avoir été le cobaye m’ayant permis de « tester » mes compétences pédagogiques. Merci, bien sûr, pour notre collaboration et toute l’aide que tu m’as apportée, et pour ton honnêteté et ta spontanéité rafraîchissantes. Guillaume, merci pour ton aide et pour tes encouragements. Merci beaucoup d’avoir lu et donné ton avis sur mon papier, d’avoir passé du temps à lire mon manuscrit de hors-thèse, en tant que grand manitou de la Neuroimmunologie et du stress. Merci pour ton calme, tu es un peu la force tranquille du labo. Serait-ce l’âge ? (Attention, il s’agit d’une blague). Et merci pour les macarons de Nancy ! Laure, merci pour l’aide que tu m’as apportée, pour les réponses à mes questions, d’avoir lu mon papier et de m’avoir donné ton opinion afin de l’améliorer. Merci également pour ton accent ensoleillé du sud-ouest, et pour l’organisation de sorties de labo. Je remercie également les étudiants de passage au labo pour de plus ou moins longues périodes, avec qui j’ai eu des échanges agréables ; merci à Carla G., pour ton aide lors des « lab parties », merci à Antoine, pour tes délicieux desserts et tes récits d’ailleurs, et merci également à

xiv

Julien, Clara, Hugo et Mathilde. Enfin, merci à Déborah, et à Romane et Sonia, ce fut un plaisir de vous encadrer lors de vos stages.

Je souhaite remercier l’ensemble du personnel (scientifique et administratif) du département de pathologie et immunologie. Merci pour les avis et idées qu’on a pu me donner lors des progress report, pour les dépannages de réactifs, les échanges de protocoles, les conseils techniques. J’ai eu également le plaisir d’échanger avec beaucoup d’entre vous, lors des retraites de département, soirées de Noël, Immunology day, PATIM happy hours ou, tout simplement, dans les couloirs et au « coin micro-onde ». Je souhaite remercier certaines personnes en particulier. Merci à Jen, JF, Filippo et Aurélie, ce fut un plaisir de partager ce petit bureau avec vous, et merci tout particulièrement à Mahdia. Je te remercie pour ton soutien, qui m’a fait beaucoup de bien. Toujours souriante, chaleureuse, à l’écoute et réconfortante, tu resteras la Maman du bureau, même si tu es passée de l’autre côté du mur. Merci à Stéphane pour ton dynamisme, ta sympathie et ton humour un peu spécial.

Je voudrais remercier profondément certaines personnes rencontrées de l’autre côté de la rue Lombard, à savoir Laura, Lyssia et Natacha. Petit moineau, ça a été un plaisir d’être ta voisine de labo. Lyssia, merci pour tout ce que tu m’as appris au niveau technique, et surtout, merci pour ton soutien, nos discussions et ton ouverture d’esprit. Chacha, mais qu’aurais-je été sans toi ? Merci pour tous les conseils que tu m’as donnés tant aux niveaux scientifique et technique, que pour la rédaction de la thèse. Merci d’avoir été ma comparse de course, pour nos discussions philosophiques et politiques, pour ton soutien moral, d’avoir toujours été présente, et de m’avoir suivie au CMU (je suis sûre que tout cela était prémédité !). Merci aussi à Pifou, pour ta joie de vivre et nos débats sans queue ni tête, à propos de plus ou moins tout et n’importe quoi.

Je tiens beaucoup à remercier les collègues et amis du CMU. Mes premières pensées vont à Valentina. Vale, les mots me manquent… Tu garderas toujours une place spéciale dans mon cœur et je suis infiniment reconnaissante pour le temps précieux passé avec toi. Je souhaite remercier les doctorants rencontrés lors des entretiens du programme Biologie- Médecine. Un grand merci à Nico, Caroline, Claire, Vanessa, Amy, Lisa et Sunil. Souvenirs souvenirs : time’s up, froid de canard, et auberge de jeunesse… Tant de choses ce sont passées depuis. On a ri et pleuré ensemble. Je suis contente d’avoir partagé cette période « doctorale » de ma vie avec vous. Un merci tout particulier à Sunil et à Amy. A vous deux, merci pour tous les moments partagés, et merci infiniment d’être là pour moi, j’ai tellement de choses pour lesquelles vous remercier que je ne sais plus quoi dire ! Merci beaucoup également à Lingzi, Aleksandra et Ebru. Enfin, je remercie aussi, en espérant oublier personne, Manon, Doro, Julie, Nico H., Dani, Damian, Loïc, Ines, Fatma, Marta, Soner, Hugo, Salva, Ronke, Aurélia, Anne-Laure, Alex, Piango, Tanja et Alex-le-grand, pour les bons moments passés ensembles, au CMU, pendant les soirées au café de la pointe ou ailleurs, les TGIF, les pique-niques à la perle du lac, etc... Je suis reconnaissante de vous avoir tous rencontrés et d’avoir été plongée dans cette ambiance internationale qui m’a beaucoup apporté, a enrichit ma vie, ouvert mon esprit, élargit mes horizons, appris à relativiser et à prendre du recul. ,Dankjewel , ش كرا ,Merci à tous : धꅍयवाद, dziękuję, danke, gracias, o ṣeun, grazie 謝謝, obrigado, σας ευχαριστώ, sağol, tualumba ! Ça va être dur de vous dire au revoir.

Je voudrais également remercier Clara, Karin, Julia, Tiphaine et Mirjam, qui m’ont permis de me sortir la tête du guidon et du monde des doctorants en Biologie, de temps en temps.

Je souhaite, enfin, remercier mes proches, qui ont largement contribué à l’aboutissement de cette thèse, par leur soutien. Merci pour beaucoup, beaucoup de choses, notamment votre compréhension pour mon manque de disponibilité chronique. Je serai très brève car j’ai du mal à trouver mes mots pour vous exprimer ce que je ressens et vous remercier correctement.

A mes vieilles branches, merci infiniment pour votre amitié et pour votre soutien sans faille pendant mes longues études. Je pourrais écrire des pages et des pages sur vous ! Je vais donc être très brève pour ne pas m’embourber, je pense de toute manière ne pas avoir besoin de cette section pour que vous sachiez à quel point vous êtes importantes à mes yeux. Un énorme merci à, Laëtitia, Gaëlle, Marie, Lise, Cécile et Julie-Anne.

A ma famille élargie, merci beaucoup pour vos encouragements et pour votre soutien lors de toutes ces années. On ne se voit pas très souvent mais cela ne m’empêche pas de penser à vous, et d’apprécier les moments passés ensemble.

Enfin, je souhaite dédier cette thèse à mes parents et à mon frère. Merci à Sylvain, mon frérot. Et enfin, merci à mes parents, merci pour tout, pour votre soutien, votre amour, et tout ce que vous avez fait pour moi depuis que je suis petite.

J’espère sincèrement que je n’oublie personne dans cette section « remerciements », qui revêt un caractère de moins en moins professionnel au fil des lignes, et dont la rédaction aura eu de véritables vertus thérapeutiques à mi-chemin, et dans la dernière ligne droite, dans l’écriture de cette thèse.

xvi

LIST OF ABBREVIATIONS

A2AR adenosine A2A receptor Aire autoimmune regulator APC antigen-presenting cell

BAD-LAMP brain and DC-associated lysosome-associated membrane protein BBB blood-brain barrier BCG bacillus Calmette-Guérin BCR B cell receptor BDCA blood dendritic cell antigen BEC blood endothelial cell BLMB Blood-leptomeningeal barrier BM bone marrow BST2 bone marrow stromal cell antigen 2 (also called PDCA-1)

CCL CC-chemokine Ligand CCR CC-chemokine receptor CD cluster of differentiation cDC conventional dendritic cell CDP common dendritic cell progenitor CFA complete Freund's adjuvant CIITA class II major histocompatibility complex transactivator CLEC2 C-type lectin receptor 2 CLIP class II-associated Ii peptide CLP common lymphoid progenitor CNS central nervous system CpG-B class B CpG oligodeoxynucleotides CpG-ODN CpG oligodeoxynucleotides CSF cerebrospinal fluid cTEC cortical thymic epithelial cell CTL cytotoxic T lymphocyte CTLA-4 cytotoxic T-lymphocyte-associated protein 4 CXCL CXC-chemokine ligand CXCR CXC-chemokine receptor CyTOF cytometry by time-of-flight

DAMP danger-associated molecular pattern DC dendritic cell DCIR dendritic cell immunoreceptor DC-SIGN DC-specific intercellular adhesion molecule-3-grabbing non-integrin Deaf1 deformed epidermal autoregulatory factor 1 DN double negative DNC double negative cell DP double positive DPP-4 dipeptyl peptidase-4 xvii

DTR diphteria toxin receptor

EAE experimental autoimmune encephalomyelitis EBV Epstein-Barr virus EGC enteric glial cell Eif4g3 eukaryotic translation initiation factor 4 gamma 3 ER endoplasmic reticulum eTAC extrathymic Aire-expressing cell

FACS fluoresence-activated cell sorting FAS-L first apoptosis signal ligand FcγR Fcγ receptor Fezf2 fez family zinc finger 2 FLT3L FMS-related tyrosine kinase 3 ligand FRC fibroblastic reticular cell

GA glatiramer acetate GFAP glial fibrillary acidic protein GFP green fluorescent protein GLOBOCAN global cancer incidence, mortality and prevalence GM-CSF granulocyte-macrophage colony-stimulating factor gp38 glycoprotein 38 (also called podoplanin or PDPN) GvHD graft versus host disease GWAS genome-wide association studies

HA Hemagglutinin HEV high endothelial venule HIV human immunodeficiency virus HLA human leukocyte antigen HVEM herpes virus entry mediator

ICAM-1 intercellular adhesion molecule 1 ICOS inducible T cell co-stimulator ICOS-L inducible T cell co-stimulator ligand Id2 inhibitor of DNA binding 2 IDO indoleamine 2,3-dioxygenase IEC intestinal epithelial cell iFABP intestinal fatty acid-binding protein IFNAR interferon-α/β receptor IFN-I (or III) type I (or III) interferon IFN-γ interferon-γ IFN-γR interferon-γ receptor Ig Immunoglobulin Ii invariant chain IL interleukin ILC innate lymphoid cell

xviii

IRF interferon-regulatory factor

K14 keratin 14

LAG-3 lymphocyte activation gene 3 LEC lymphatic endothelial cell LFA-1 lymphocyte function-associated antigen 1 LN lymph node LNSC lymph node stromal cell LPS lipopolysaccharide LSEC liver sinusoidal endothelial cell Ltβr lymphotoxin β receptor LV lymphatic vessel Lyve-1 lymphatic vessel endothelial hyaluronan receptor 1 mAb monoclonal antibody Mac-1 macrophage-1 antigen MadCAM-1 mucosal vascular addressin cell adhesion molecule 1 MBP myelin basic protein MDSC myeloid-derived suppressor cell MHC major histocompatibility complex MIIC MHC class II compartment MIP-1β macrophage inflammatory protein 1β MM multiple myeloma moDC monocyte-derived dendritic cell MOG myelin oligodendrocyte protein MRI magnetic resonance imaging MS multiple sclerosis mTEC medullary thymic epithelial cell Mtg16 myeloid translocation gene 16 MyD88 myeloid differentiation primary response gene 88

NET neutrophil extracellular trap NF-κB nuclear factor-κB NK natural killer NSCLC non-small cell lung carcinoma

OVA ovalbumin p promoter (of CIITA) PAMP pathogen-associated molecular pattern PBMC peripheral blood mononuclear cell PCAM-1 platelet endothelial cell adhesion molecule 1 (also called CD31) PD-1 programmed cell death 1 pDC plasmacytoid dendritic cell PDCA-1 plasmacytoid dendritic cell antigen 1 (also called BST2)

xix

PD-L1 (or 2) programmed death 1 ligand 1 (or 2) PGE2 prostaglandin E2 PI3K phosphoinositide 3-kinase PLP proteolipid protein PP Peyer's Patches PPMS primary progressive multiple sclerosis Ppy pancreatic polypeptide Prox-1 prospero homeobox protein 1 PRR pattern recognition receptor PTA peripheral tissue-restricted antigen PTPRS (or PTPRF) protein-tyrosine phosphatase receptor type S (or F) pTreg peripherally-induced regulatory T cell PTX pertussis toxin

RANTES regulated on activation, normal T cell expressed and secreted ROS reactive oxygen species RRMS relapsing-remitting multiple sclerosis

S1P sphingosine-1-phosphate S1PR1 sphingosine-1-phosphate receptor 1 SAND Sp100, AIRE-1, NucP41/75, DEAF1 Siglec-H sialic acid binding Ig-like lectin H SLO secondary lymphoid organ SNP single nucleotide polymorphism SP simple positive SPMS secondary progressive multiple sclerosis

TA tumor-associated TAN tumor-associated neutrophils TAP transporter associated with antigen processing TCF4 transcription factor 4 (also called E2-2) TCM tumor-conditioned medium TCR T cell receptor TdLN tumor-draining lymph node TEC thymic epithelial cell

TFH T follicular helper TGF-β transforming growth factor-β TGF-βR transforming growth factor-β receptor Th T helper TIL tumor-infiltrating lymphocyte TLR toll-like receptor TLS tertiary lymphoid structure TME tumor microenvironment TNF tumor necrosis factor tOVA truncated ovalbumin Tr1 type 1 regulatory T cell xx

TRAIL tumor-necrosis-factor related apoptosis inducing ligand Treg regulatory T cell TSLP thymic stromal lymphopoietin tTreg thymus-derived regulatory T cell Tyr tyrosinase

UV ultra violet

VCAM-1 vascular cell adhesion molecule 1 VEGF-C vacular endothelial growth factor C VEGFR-3 vascular endothelial growth factor receptor 3 VLA-4 very late antigen 4

xxi

TABLE OF CONTENTS

I. INTRODUCTION ...... 1 I.1. Preamble ...... 1 I.2. The balance of the immune system ...... 2 I.3. Antigen presentation...... 3 I.3.a. Antigen presentation: mechanisms implicated ...... 3 I.3.b. Professional antigen-presenting cells (APCs) ...... 6 I.4. Thymic selection and central T cell tolerance ...... 9 I.4.a. T cell receptor gene rearrangement and positive selection ...... 9 I.4.b.Negative selection ...... 11 I.5. Peripheral T cell responses ...... 14 I.5.a. T cells: polarization and roles ...... 14 I.5.b. Peripheral T cell tolerance ...... 17 I.6. Unconventional antigen-presenting cells ...... 20 I.6.a. An overview of unconventional and semi-professional APCs ...... 20 I.6.b. Plasmacytoid dendritic cells (pDCs) ...... 23 I.6.b.i. Plasmacytoid dendritic cells: generalities ...... 23 I.6.b.ii. Ontogeny and heterogeneity of pDCs ...... 26 I.6.b.iii. Plasmacytoid DC functions independent of antigen presentation ...... 27 I.6.b.iv. Antigen-presenting functions of pDCs ...... 30 I.6.c. Lymphatic endothelial cells (LECs) ...... 38 I.6.c.i. Different subsets of lymph node stromal cells (LNSCs) ...... 38 I.6.c.ii. Ontogeny and development of LECs ...... 40 I.6.c.iii. Major types of LECs ...... 42 I.6.c.iv. Regulation of LN-LEC proliferation and survival ...... 44 I.6.c.v. LECs impact peripheral T cell responses through mechanisms independent of antigen presentation ...... 44 I.6.c.vi. Antigen-presenting abilities of LECs: uptake of exogenous antigens and presentation to cells...... 48

II. THESIS AIM ...... 53

xxii

III. CHAPTER A ...... 54 III.1. Introduction (Chapter A) ...... 54 III.1.a. Tumor immunity ...... 54 III.1.a.i. Cancer and tumor immunity ...... 54 III.1.a.ii. Cancer immunotherapies ...... 57 III.1.b. Plasmacytoid DCs in tumor immunity ...... 60 III.1.b.i. pDCs in tumor immunity: Generalities ...... 60 III.1.b.ii. pDC innate functions in tumor immunity ...... 61 III.1.b.ii. pDC antigen-presenting functions in tumor immunity ...... 61 III.1.c. Specific aim (Chapter A) ...... 67 III.2. Results (Chapter A) ...... 69 III.3. Discussion (Chapter A) ...... 96

IV. CHAPTER B ...... 112 IV.1. Introduction (Chapter B)...... 112 IV.1.a. Multiple sclerosis and its animal model, EAE ...... 112 IV.1.a.i. Multiple sclerosis ...... 112 IV.1.a.ii. Causes ...... 112 IV.1.a.iii. EAE model ...... 115 IV.1.a.iv. Immunopathophysiology of MS and EAE ...... 115 IV.1.a.v. MS therapies targeting the immune system ...... 115 IV.1.b. Role of self-antigen presentation by LECs in autoimmunity ...... 123 IV.1.b.i. Expression of peripheral tissue-restricted antigens by LECs ...... 123 IV.1.b.ii. Impact of LEC presentation of endogenously-expressed PTAs on T cell responses ...... 126 IV.1.b.iii Molecular pathways implicated in peripheral T cell tolerance mediated by LECs ... 129 IV.1.c. Specific aim (Chapter B) ...... 130 IV.2. Results (Chapter B) ...... 132 IV.3. Discussion (Chapter B) ...... 153

V. CONCLUDING REMARKS ...... 162

VI. APPENDICES ...... 166 VI.1. Appendix 1 (Humbert, Dubrot* & Hugues*, Front Immunol, 2016) ...... 166 VI.2. Appendix 2 (Humbert & Hugues, Oncoimmunology, 2018) ...... 180

VII. REFERENCES ...... 184

xxiii

LIST OF FIGURES

Figure 1. Classical pathway for the presentation of exogenous antigens loaded onto MHC-II to CD4+ T cells………………………………………………………………………………….....5

Figure 2. Professional antigen-presenting cells, based on the attribute definition………………………………………………………………………………………..8

Figure 3. Interaction of developing thymocytes with stromal cells in the thymus……………...10

Figure 4. Thymic selection…………………………………………………………………….11

Figure 5. Negative selection and central T cell tolerance………………………………….…...13

Figure 6. CD4+ T cell polarization………………………………………………………….…16

Figure 7. Co-stimulatory and co-inhibitory signals…………………………………………….17

Figure 8. Peripheral T cell tolerance……………………………………………………….….19

Figure 9. The multiple functions of plasmacytoid dendritic cells………………………………25

Figure 10. Specific location of the lymph node stromal cell subsets…………………………....39

Figure 11. Antigen acquisition and presentation by lymphatic endothelial cells……….………..50

Figure 12. Anti-tumor immunity cycle………………………………………………………....57

Figure 13. A wide variety of cancer immunotherapies…………………………………………58

Figure 14. Signaling pathways of type A and B CpG oligodeoxynucleotides in early and late endosomes in plasmacytoid dendritic cells……………………………………………………..65

Figure 15. Priming of Th17 cells by activated antigen-presenting plasmacytoid dendritic cells leads to tumor growth control: working model………………………………………………...66

Figure 16. Can intratumoral administration of CpG-B along with tumor antigenic peptide reverse the tolerogenic phenotype of tumor-associated plasmacytoid dendritic cells?...... 68

Figure 17. Intratumoral administration of CpG-B and tumor antigenic peptide induces tumor cell death: working model………………………………………………………………….…107

Figure 18. Role of antigen-presentation by lymph node stromal cells in peripheral T cell tolerance……………………………………………………………………………………...125

Figure 19. Antigen presentation-dependent role of lymph node stromal cells in peripheral T cell responses………………………………………………………………………………….….126

xxiv

I. INTRODUCTION

I.1. Preamble

The present thesis aims at understanding the role of unconventional antigen presenting cells (APCs) in the modulation of peripheral T cell responses, in particular the role of two different cell types studied in distinct immunological contexts, therefore necessitating two chapters. Chapter A analyzes the role of plasmacytoid dendritic cells (pDCs), in anti-tumor immunity, while Chapter B investigates the role of lymphatic endothelial cells (LECs), in the context of autoimmunity. Both chapters are focused on antigen-presenting functions, in particular restricted to major histocompatibility complex class II (MHC-II), and its impact on CD4+ T cell responses. A general introduction first summarizes our current knowledge on antigen presentation and T cells responses, and provides an overview of unconventional APCs, with a description of pDCs and LECs. This general section is followed by two specific introductions; Chapter A (pDCs in tumors) and Chapter B (LECs in autoimmunity).

I.2. The balance of the immune system

From Janeways’s Immunobiology (Murphy et al., 2008)

The immune system is constituted of two branches, namely, innate and adaptive immunity, and aims at protecting the organism against infection and other harms, including tumors and damaged cells. In addition, this highly evolved system self-regulates, in order to prevent adverse effects, such as autoimmunity. Therefore, a tightly regulated balance between immune activation and tolerance is required. An immune response typically starts with the activation of the innate immune system, which proceeds to a first discrimination between self and non-self. If the innate immune response is not sufficient to eradicate a pathogen, a - highly specific - adaptive immune response can be initiated.

Innate immunity APCs, such as dendritic cells (DCs) and macrophages, and some other immune cells, express pattern recognition receptors (PRRs) that sense pathogen- or damage-associated molecular patterns (PAMPs or DAMPs). DCs are considered as a bridge linking innate and adaptive immunity. If an adaptive immune response is initiated, it is “polarized”, depending on the type of PAMPs/DAMPs that have been recognized by the innate immune system, in order to induce an adequate response.

Adaptive immunity The adaptive immune system comprises two arms; a T cell-mediated immunity, which will be described in the next sections, and a humoral immunity, which involves B lymphocytes. Extremely briefly, when activated, B cells differentiate into plasma cells that produce antibodies, constituting the humoral immune response, which protects the extracellular space.

I.3. Antigen presentation

I.3.a. Antigen presentation: mechanims implicated

T cell development, described in the next section, occurs in the thymus and generates a vast repertoire of naïve CD4+ and CD8+ T cells (Klein et al., 2014; Takaba and Takayanagi, 2017). When fully mature, these cells exit the thymus, enter the circulation and can reach the secondary lymphoid organs (SLOs), among which lymph nodes (LNs), where they can be primed (Fu et al., 2016). The specificity of T cells is encoded by their T cell receptor (TCR). In order to be activated, the TCR must recognize its cognate antigenic peptide bound to MHC molecules; MHC-I for CD8+ T cells and MHC-II for CD4+ T cells, allowing a tight control of T cell responses (Benvenuti, 2016; Murphy et al., 2008). These peptide/MHC complexes are presented by APCs to naïve T cells, in the SLOs, structures that favour the encounter of naïve T cells with APCs having acquired antigens (Murphy et al., 2008). Naïve T cell priming takes place in the presence of three integrated signals provided by APCs, the interface between the APC and the T cell being called the “immunological synapse” (Benvenuti, 2016). Signal 1 corresponds to the engagement of the TCR with the above-mentionned peptide/MHC complex, ensuring the specificity of the activation. Signal 2 is the ligation of co- stimulatory molecules, such as CD80/CD86 and CD40, expressed by APCs, with CD28 and CD40-L expressed by T cells, respectively. Signal 3 is the pattern of cytokines produced by APCs, which polarizes the T cell response. Signals 2 and 3 translate the state of activation of APCs by DAMPs or PAMPs, leading to the polarization of the T cell response, which is described later in this introduction (Benvenuti, 2016; Carbo et al., 2014). While signal 1 is strictly required, signal 1 in the absence of signals 2 and 3 leads to peripheral T cell tolerance (Baldwin and Hogquist, 2007; Iberg et al., 2017). Once naïve T cells have been primed in the SLOs, they can migrate to the tissue from which their cognate antigen has been acquired (Fu et al., 2016).

Presentation of antigens loaded onto MHC-I to CD8+ T cells With the exception of erythrocytes, all cell types express MHC-I in order to present endogenous antigens, including intracellular pathogen and tumor antigens, enabling the recognition of these cells by CD8+ T cells in tissues, and their subsequent killing (Neefjes et al., 2011). Briefly, the classical antigen-processing pathway for the presentation of endogenous antigens to CD8+ T cells encompasses the degradation of antigens by the proteasome in the cytosol, and the translocation

of generated peptides into the endoplasmic reticulum (ER) through transporter associated with antigen processing (TAP), where they are loaded onto MHC-I (Neefjes et al., 2011). Alternative antigen-processing pathways for MHC-I also exist but will not be described here (Oliveira and van Hall, 2015). Nonetheless, APCs have the ability to internalize, process and present exogenous antigens loaded onto MHC-I to CD8+ T cell, a mechanism called “cross-presentation”, for which multiple pathways of antigen-processing have been described (Adiko et al., 2015; Joffre et al., 2012). In the TAP-dependent cytosolic pathways, exogenous antigens, after phagocytosis, are either exported to the cytosol, degraded by the proteasome and loaded onto MHC-I in the ER, or they are re- imported into the phagosome and loaded onto MHC-I (Gutierrez-Martinez et al., 2015; Joffre et al., 2012). Fusion between the ER and the phagosome with subsequent loading into the mix compartment has also been described (Guermonprez et al., 2003; Gutierrez-Martinez et al., 2015). The vacuolar pathway (TAP-independent) involves the degradation of antigens and the loading on MHC-I directly in the phagosome (Joffre et al., 2012).

Presentation of antigens loaded onto MHC-II to CD4+ T cells In contrary to MHC-I, MHC-II expression, which is regulated by the master regulator Class II MHC complex transactivator (CIITA), is restricted to APCs (Kambayashi and Laufer, 2014; Reith et al., 2005). The classical pathway for the presentation of exogenous antigens to CD4+ T cells through MHC-II, depicted in Fig. 1, involves the assembly of MHC-II molecules in the ER, where they form a complex with the invariant chain (Ii). The complex is then transported to the MHC class II compartment (MIIC), where Ii and the internalized exogenous antigens are degraded by MIIC-resident proteases (Neefjes et al., 2011; Roche and Furuta, 2015). The MHC- II peptide-binding groove is normally hidden by Class II-associated Ii peptide (CLIP), a fragment from Ii. With the help of a chaperone, called human leukocyte antigen (HLA)-DM in human and H2-M in mouse, CLIP is exchanged with the antigenic peptide, which is finally loaded onto MHC-II. MHC-II molecules are subsequently transported to the plasma membrane, where they can present the peptide to CD4+ T cells (Neefjes et al., 2011; Roche and Furuta, 2015). Non-classical pathways of MHC-II loading for the presentation of endogenous antigens to CD4+ T cells also have been described, involving autophagy as well as non-autophagic pathways (Leung, 2015; Munz, 2015; Roche and Furuta, 2015). It is for example the case for intracellular viral antigens in infected APCs, or for the presentation of endogenously-expressed self-antigens by thymic epithelial cells (TECs), which use macroautophagy (Anderson and Takahama, 2012; Klein et al., 2009; Veerappan Ganesan and Eisenlohr, 2017).

Figure 1. Classical pathway for the presentation of exogenous antigens loaded onto MHC-II to CD4+ T cells. Major histocompatibility complex class II (MHC-II) is assembled in the endoplasmic reticulum (ER), where it forms a complex with Ii, the invariant chain. This complex is transported via the Golgi to the MHC-II compartment (MIIC), through the plasma membrane or directly. Ii and the exogenous antigens are degraded in MIIC by proteases. Class II-associated Ii peptide (CLIP), a fragment from Ii, hides the MHC-II peptide-binding groove, until it is exchanged with an antigenic peptide by the chaperone HLA- DM, in human (or H2-M in mice). Subsequently, MHC-II molecules are transported to the plasma membrane where they can present the peptide to CD4+ T cells. APC, antigen-presenting cell; TCR, T cell receptor. Adapted from Neefjes et al., Nat Rev Immunol, 2011 (Neefjes et al., 2011).

I.3.b. Professional antigen-presenting cells

The definition of professional APCs is not a consensus among the scientific community. Two main definitions co-exist in the literature. The first one is based on attributes (related to antigen uptake, processing and presentation) harbored by a given cell type (Kambayashi and Laufer, 2014; Kashem et al., 2017; Mann and Li, 2014; Schuijs et al., 2019). According to this definition, professional APCs include conventional DCs (cDCs), macrophages and B cells. Their hallmark is the constitutive expression of MHC-II and of the exogenous antigen-processing machinery, as well as the expression of co-stimulatory molecules upon activation (Kambayashi and Laufer, 2014) (Fig. 2). B cells internalize exogenous antigens through the B cell receptor (BCR). T cell activation by B cells is crucial for the T cell- induced activation of B cells (Yuseff and Lennon-Dumenil, 2015). Macrophage and cDC phagocytic activity allows the internalization of exogenous antigens (Fig. 2) (Kambayashi and Laufer, 2014). This description also relies on the ability of cDCs, macrophages and B cells to activate naïve T cells in vitro. However, this definition may not be appropriate in vivo, as the ability of B cells and macrophages to present antigens and to be optimally localized in order to prime naïve T cells (in the LN paracortex) is not as efficient as that of cDCs. Therefore, B cells and macrophages may not be considered “professional” APCs, to the same extent as cDCs. The second definition of professional APCs is related to the function of the cell type, rather than the attributes. According to this definition, cDCs are professional APCs because their main function is to present antigens, which is not the case for macrophages and B cells. The term “professional” is defined as “engaged in a specified activity as one’s main paid occupation rather than amateur” (Oxford dictionary) and is used in this thesis to refer to APCs whose main function is to initiate T cell responses, or in other words, to prime naïve T cells. We therefore considered cDCs as professional APCs in this thesis, their main function being indeed to initiate T cell responses. The main function of B cells is to produce antibodies, and they act as APCs in order to get help from T cells. Finally, the main function of macrophages is to phagocyte pathogens and to be implicated in inflammation.

DCs comprise cDCs, monocyte-derived inflammatory DCs and pDCs, the later cell type, being the object of this thesis, is described later in the introduction (Anderson et al., 2018; Dalod et al., 2014; Guilliams et al., 2014; Merad et al., 2013). Two distinct subsets of cDCs (CD11chi MHC-II+ in mouse) have been described, cDC1 and cDC2 (Anderson et al., 2018; Dalod et al., 2014; Guilliams et al., 2014; Merad et al., 2013). Although the markers used to define these subtypes

differ between human and mice, and there is an inter-tissue heterogeneity, equivalent populations have been found across species and tissues (Guilliams et al., 2016). Recent investigations from human peripheral blood, however, suggest that cDCs might comprise more than two subtypes (See et al., 2017; Villani et al., 2017). In mice, cDC1 are defined as CD103+ and/or CD8+ and are particularly efficient at cross- presenting exogenous antigens via MHC-I to CD8+ T cells. Conventional DC2 are defined as CD11b+ and are specialized in the presentation of exogenous antigens to CD4+ T cells via MHC- II (Anderson et al., 2018; Dalod et al., 2014; Guilliams et al., 2014; Gutierrez-Martinez et al., 2015; Merad et al., 2013). The use of the term “conventional” for cDCs, as opposed to pDCs and monocyte-derived DCs, may be due to the chronology of scientific discoveries, as cDCs were described first, in 1973 before pDCs and monocyte-derived DCs (Steinman and Cohn, 1973). Nowadays, the term “conventional” for this DC subtype may not be very accurate, with “cDCs” including cDC1 and cDC2, which might be further sub-divided, each subsets having specific features and abilities regarding antigen presentation to CD4+ to CD8+ T cells. The use of the term “conventional” for cDCs may also be due to the fact that cDCs have been extensively studied, compared with the two other cell types.

Besides professional APCs, other cell types can behave as APCs, referred to as unconventional APCs in this thesis, and cross-present exogenous antigens via MHC-I and/or present antigens through MHC-II (Kambayashi and Laufer, 2014). These cell types and the conditions in which they can act as APCs are described later in this introduction, unconventional APCs representing the object of this thesis.

Figure 2. Professional antigen-presenting cells, based on the attribute definition Conventional dendritic cells (cDCs), macrophages and B cells, constitutively express major histocompatibility complex class II (MHC-II) and the machinery for antigen processing. APC; antigen-presenting cells; BCRs, B cell receptors; DAMPs, damage-associated molecular patterns; PAMPs, Pathogen-associated molecular patterns Adapted from Kambayashi and Laufer, Nat Rev Immunol, 2014 (Kambayashi and Laufer, 2014).

I.4. Thymic selection and central T cell tolerance

I.4.a. TCR gene rearrangement and positive selection

Lymphocytes precursors migrate from the bone marrow (BM), where the hematopoiesis takes place, to the thymus (Klein et al., 2014). In the cortex, thymocytes undergo TCR gene rearrangement, which generates a very diverse T cell repertoire, and a process called “positive selection”, in order to get rid of the “useless” thymocytes (Fig. 3) [reviewed in (Klein et al., 2014; Takaba and Takayanagi, 2017)]. This mechanism ensures that, once they reach the periphery, selected thymocytes will have the ability to recognize potential pathogens. Developing thymocytes progressively commit to T cell lineage (vs B cells), to αβ T cells (vs γδ T cells), and finally to CD8 or CD4 lineage, depending on the affinity of their TCR for self- peptide/MHC-I or MHC-II, respectively, therefore becoming simple positive (SP) cells (Klein et al., 2014; Murphy et al., 2008; Takaba and Takayanagi, 2017). Thymocytes harboring a TCR that reacts strongly enough to self-peptide/MHC complexes presented by cortical thymic epithelial cells (cTECs) receive survival signals, whereas the other ones die by neglect (Fig. 4) (Anderson and Takahama, 2012; Klein et al., 2009; Mandl et al., 2013; Stefanova et al., 2002).

Figure 3. Interaction of developing thymocytes with stromal cells in the thymus. T cell progenitors migrate from the bone marrow and reach the thymus through blood vessels lining the cortico-medullary junction. Developing double negative (DN) thymocytes, which undergo T cell receptor (TCR) gene rearrangement, progressively migrate to the sub-capsular zone. Double positive (DP) thymocytes, which express both CD4 and CD8 molecules at their surface, scan cortical thymic epithelial cell (cTEC) surface for self-peptide loaded onto major histocompatibility complex (MHC). Cortical TECs provide survival signals to DP if they express a TCR with a sufficient affinity for self-peptide/MHC complexes. Other DP die by neglect. This process is called positive selection. In the meantime, DP commit to either CD4 or CD8 lineage, and the newly generated simple positive (SP) thymocytes subsequently migrate to the medulla, where they go through the negative selection process. Briefly, dendritic cells (DCs) and medullary TECs (mTECs) present tissue-restricted self-peptide/MHC to SP thymocytes. SP thymocytes harboring a TCR with a too high affinity for self-peptide/MHC are eliminated by clonal deletion, while the other thymocytes enter the circulation. Adapted from Klein et al., Nat Rev Immunol, 2009 (Klein et al., 2009).

Figure 4. Thymic selection. T cell receptor (TCR) affinity for self-peptide/major histocompatibility complex (MHC) determines the fate of developing thymocytes. During the positive selection, only the double positive thymocytes harboring a TCR with a sufficient affinity for self-peptide/MHC receive survival signals, whereas the ones expressing a TCR with no or a too low affinity die by neglect. During the negative selection, simple positive thymocytes harboring a TCR with a too high affinity for self-peptide/MHC undergo apoptosis. A narrow affinity threshold determines whether the interaction of the TCR with self-peptide/MHC induces a positive or negative selection Adapted from Klein et al., Nat Rev Immunol, 2009 (Klein et al., 2009).

I.4.b. Negative selection

After positive selection, SP thymocytes undergo the process of negative selection, which takes place in the thymus medulla (Fig. 3). This mechanism is part of the central tolerance that aims at preventing autoimmunity, by deleting autoreactive T cell clones before exiting the thymus and reaching the periphery [reviewed in Ref. (Klein et al., 2014; Xing and Hogquist, 2012)]. Medullary TECs express peripheral tissue-restricted antigens (PTAs), which are normally exclusively found in the periphery (Derbinski et al., 2001; Kyewski et al., 2000) (Fig. 5). PTA expression by mTECs is, for a large majority, regulated by transcription factors, such as the autoimmune regulator (Aire) and the recently identified fez family zinc finger 2 (Fezf2), which are differentially regulated (Derbinski et al., 2005; Takaba et al., 2015; Takaba and Takayanagi, 2017). Of note, mutations in Aire are involved in severe autoimmune disorders (Anderson and Su, 2016; Anderson et al., 2002; Liston et al., 2004). PTAs can be presented to SP thymocytes by different ways; PTAs can be directly presented by mTECs, acquired from mTECs and presented by thymus-resident cDCs, or acquired in

peripheral tissues by migrating cDCs/pDCs and presented to the SP thymocytes (Fig. 5) (Audiger et al., 2017; Bonasio et al., 2006; Gallegos and Bevan, 2004; Hadeiba et al., 2012; Koble and Kyewski, 2009). SP thymocytes harboring a TCR with a too strong affinity for self- antigen/MHC complexes are eliminated by clonal deletion (Fig. 4 and 5) (Baldwin and Hogquist, 2007; Bonasio et al., 2006; Gallegos and Bevan, 2004; Koble and Kyewski, 2009). A small part of the CD4+ SP thymocytes expressing a TCR with a high affinity for self- antigen/MHC complexes differentiates into thymus-derived regulatory T cells (tTregs), previously called natural Tregs, in which the expression of the transcription factor Foxp3 is induced and whose development depends on transforming growth factor (TGF)-β (Fig. 5) (Abbas et al., 2013; Hsieh et al., 2012; Liu et al., 2008b; Lu et al., 2017). Of note, tTregs induced by cDCs or pDCs are functionally different from one another, with a distinct pattern of cytokine production (Martin-Gayo et al., 2010). Although less frequent than CD4+ tTregs, CD8+ tTregs have also been characterized (Cosmi et al., 2003; Menager-Marcq et al., 2006; Pomie et al., 2008; Vuddamalay et al., 2016). Despite the mechanism of negative selection, some (non-Treg) autoreactive T cells escape central tolerance and reach the periphery (Anderson et al., 2000; Lohse et al., 1996). This results either from an absence of presentation of specific antigens in the thymus, or from a lack of clonal deletion of thymocytes expressing a TCR with an affinity for self-antigen/MHC complexes that does not reach the negative selection threshold (Fig. 5) (Enouz et al., 2012). In addition, for some PTAs, central T cell tolerance relies exclusively on the differentiation of tTregs and not at all on clonal deletion (Legoux et al., 2015). Inefficiency/failure in central tolerance is strongly linked with autoimmunity [reviewed in (Cheng and Anderson, 2018; Theofilopoulos et al., 2017)].

Figure 5. Negative selection and central T cell tolerance. Following positive selection - not illustrated in this scheme -, simple positive (SP) thymocytes go through the negative selection process in the thymic medulla. This mechanism involves conventional dendritic cells (cDCs) that reside in the thymus, medullary thymic epithelial cells (mTECs), which express peripheral tissue-restricted antigens (PTAs) (green), as well as plasmacytoid DCs (pDCs) and cDCs that acquire exogenous antigen (yellow) in the periphery before migrating to the thymus. These four cell types present self-antigen peptide/major histocompatibility complex (MHC) to SP thymocytes. The thymocytes that express a T cell receptor (TCR) with a too strong affinity for self (dark colors) are eliminated by clonal deletion, while the ones harboring a TCR with an intermediate affinity (medium colors) differentiate into thymus-derived regulatory T cells (tTregs). SP thymocytes expressing a low-affinity TCR (light colors) exit the thymus and reach the periphery to constitute the naïve T cell pool. However, some self-reactive T cells (dark colors) escape the mechanisms of central T cell tolerance and reach the periphery. Antigen transfers and cell migration are respectively depicted in dashed and dotted arrows. Ags, antigens; exo Ags, exogenous antigens; migr. cDC, migratory cDC; thym. cDC, thymus-resident cDC. Adapted from Humbert, Hugues* and Dubrot*, Front Immunol, 2016 [Appendix 1 (Humbert et al., 2016)].

I.5. Peripheral T cell responses

I.5.a. T cells: polarization and roles

Mature naïve CD8+ and CD4+ T cells exit the thymus through the bloodstream and reach SLOs, where they can be activated by APCs (Fu et al., 2016). As mentioned earlier, the polarization of naïve T cells depends on co-stimulatory molecule expression (signal 2) and cytokines secreted by APCs (signal 3), which themselves are a consequence of the type of PAMPs or DAMPs that have been detected (Fig. 6) (Benvenuti, 2016). Cytotoxic T lymphocytes (CTLs) differentiate from activated CD8+ T cells. These cells are crucial for the elimination of infected or tumor cells, although their functions are often hampered in cancer and chronic viral infections (Halle et al., 2017; Wang et al., 2017). CTLs produce interferon (IFN)-γ, have direct killing activity through the production of granzyme B and perforin and via the expression of molecules such as tumor-necrosis-factor related apoptosis inducing ligand (TRAIL) and first apoptosis signal ligand (FAS-L) (Halle et al., 2017). CD4+ T cell activation gives rise to different subsets of effector T cells, called T helper (Th) cells. Th cell polarization is orchestrated by transcription factor, or master regulators, and characterized depending on the pattern of cytokines they secrete (Fig. 6). The first Th subsets discovered were Th1 and Th2 cells (Mosmann and Coffman, 1989). Th1 cells, which produce IFN-γ, are implicated in intracellular pathogen clearance, anti-tumor immunity and can also play pathogenic roles in autoimmunity (Ivashkiv, 2018; Mosmann and Coffman, 1989). Th2 cells secrete interleukin (IL)-4, IL-5 and IL-13, and are involved in extracellular pathogen clearance and parasites (Mosmann and Coffman, 1989). Th cells also include Th17, Th9 and Th22 cells, which have been discovered later. Th17 cells produce IL-17, can also produce other cytokines, and play roles in immune responses against fungal infections and extracellular pathogens (Korn et al., 2009). Increasing evidence has shown an important implication of Th17 cells in autoimmune diseases and in anti-tumor responses, although their functions in this context are still unclear (Guery and Hugues, 2015a; Hirahara and Nakayama, 2016; Stockinger and Omenetti, 2017). Th9 and Th22 have been less studied and their roles have not been fully elucidated. Th9, which produce IL-9, have been implicated in inflammation, allergy and autoimmunity (Hirahara and Nakayama, 2016; Kaplan et al., 2015). IL-22-producing Th22 are involved in skin homeostasis and appear to play immunogenic or tolerogenic functions, depending on the immunological context (Jia and Wu, 2014). Finally, T follicular helper cells (TFH cells), which secrete IL-4 and IL- 21, are implicated in the activation of B cells (Crotty, 2011).

In addition to Th cell subsets, naïve CD4+ T cells can differentiate into peripherally-induced Tregs (pTregs), previously called induced Tregs, as part of the peripheral tolerance, described in the next section (Fig. 6) (Abbas et al., 2013; Kanamori et al., 2016). The differentiation of pTregs (Foxp3+ CD25+) occurs under tolerogenic conditions and relies on IL-2 and TGF-β (Chen et al., 2011; Chen et al., 2003; Kanamori et al., 2016; Lu et al., 2017). These cells have an important role in preventing autoimmunity (Sakaguchi et al., 1995). Peripherally-induced Tregs are much less stable than tTregs (Dominguez-Villar and Hafler, 2018; Kanamori et al., 2016; Li and Rudensky, 2016). Tregs target effector T cells but also DCs and natural killer (NK) cells (Vignali, 2008). For instance, they can inhibit DC maturation, prevent T cell priming by DCs, or induce a tolerogenic DC phenotype, for example in pDCs (Lippens et al., 2016; Vignali, 2008). Tregs inhibit effector T cell functions by diverse mechanisms. They are able to directly eliminate them by producing Granzyme B and perforin (Cao et al., 2007). They can induce the arrest of cell cycle through Galectin-1 secretion or surface expression, and also secrete anti-inflammatory cytokines including IL-10 and TGF-β (Garin et al., 2007). The expression of CD103 and inducible T cell co- stimulator (ICOS) by Tregs define a subpopulation with greater suppressive abilities (Barthlott et al., 2015). Other CD4+ T cell subsets with a regulatory phenotype co-exist, such as type 1 regulatory T (Tr1) cells (Foxp3-negative), which produce IL-10 and TGF-β and play important roles in the prevention of inflammation and autoimmunity (Bacchetta et al., 2005; Battaglia et al., 2006). Tr17 cells also have been described, with suppressive activity and implicated in the prevention of autoimmunity (Kim et al., 2017). In addition, CD8+ Tregs also have been characterized, although much less studied compared to CD4+ Tregs (Pomie et al., 2008). Finally, CD4+ Th cells and Tregs exhibit important plasticity and diversity (DuPage and Bluestone, 2016; Kunicki et al., 2018). For instance, Th17 and Tregs both exhibit reciprocal plasticity and instability, and require TGF-β for their differentiation, consequently Tregs can convert into Th17 and the other way round, a phenomenon with important consequences in autoimmunity and anti-tumor immunity (Dominguez-Villar and Hafler, 2018; Geng et al., 2017; Guery and Hugues, 2015b; Kleinewietfeld and Hafler, 2013; Li and Rudensky, 2016). In addition, Th17 cells can exhibit tolerogenic functions under certain conditions (Guery and Hugues, 2015b; Stockinger and Omenetti, 2017).

Figure 6. CD4+ T cell polarization. Naïve CD4+ T cell differentiate into distinct type of effector or regulatory cells, depending on the cytokines secreted by antigen-presenting cells and other innate immune cells. Polarizing cytokines (next to the arrows), transcription factors defining the subtypes (under the cells) and the main cytokine produced by each subset are depicted. Adapted from Carbo et al., Front Cell Dev Biol, 2014 (Carbo et al., 2014).

I.5.a. Peripheral T cell tolerance

Figure 7. Co-stimulatory and co-inhibitory signals. This figure depicts ligand-receptor interactions that are mentioned in this thesis, between antigen- presenting cells (APCs), on the left, and T cells, on the right. Signal 1 corresponds to the engagement of the T cell receptor (TCR) by the peptide/MHC complex. Signal 2 corresponds to the engagement of receptor on T cells by ligands on APCs. These ligands are co-stimulatory or co-inhibitory molecules. The outcomes of these interactions are depicted: activation (plus green) or inhibition (minus red). The secretion of polarizing cytokines by APCs is depicted as “Signal 3”. ICOS, inducible T cell co-stimulator, KIR, killer cell immunoglobulinlike receptor; LAG3, lymphocyte activation gene 3; PD1, programmed cell death protein 1; PDL1, programmed cell death protein 1 ligand. Modified from Pardoll, Nat Rev Cancer, 2012 (Pardoll, 2012).

As mentioned above, the process of negative selection, which takes place in the thymus, aims at eliminating autoreactive thymocytes before they exit the thymus. However, some (non-Treg) autoreactive T cells escape thymic central T cell tolerance and enter the periphery (Anderson et al., 2000; Legoux et al., 2015; Lohse et al., 1996; Richards et al., 2016). Therefore, there are additional mechanisms that take place in the periphery, the so-called peripheral tolerance [reviewed in (Walker and Abbas, 2002; Xing and Hogquist, 2012)]. The induction of cross- tolerance by peripheral DCs has been extensively investigated in the past (Steinman et al., 2000). Antigens are acquired by immature DCs via apoptotic cell phagocytosis in peripheral tissues and are presented to T cells in the SLOs (Adler et al., 1998; Heath et al., 1998; Kurts, 2000).

In the absence of any co-stimulatory signal (signal 2), the presentation of antigens induces CD4+ and CD8+ T cell deletion or anergy, which correspond to a physical elimination (apoptosis) or a functional inactivation, respectively (Baldwin and Hogquist, 2007; Iberg et al., 2017). Ligation of cytotoxic T-lymphocyte-associated protein 4 (CTLA-4) or programmed cell death 1 (PD-1), expressed by T cells, with CD80/CD86 or programmed death 1 ligand 1 or 2 (PD-L1 or 2), respectively, expressed by APCs, can also induce a state of unresponsiveness in T cells (Fig. 7) (Freeman et al., 2000; Iberg et al., 2017; Pardoll, 2012). In addition, antigen-presentation in the absence of co-stimulation can lead to the induction of pTregs, in the presence of anti- inflammatory factors and/or co-inhibitory signals (Fig. 7) (Abbas et al., 2013; Blander and Medzhitov, 2006; Steinman et al., 2003; Wilson et al., 2003). The functions of pTregs are similar to tTregs and are described in the previous section. As above-mentionned, other T cells with a regulatory phenotype can be induced, such as Tr1 or Tr17 cells (Bacchetta et al., 2005; Battaglia et al., 2006; Kim et al., 2017). These processes take places in the LNs and involve both resident and migratory DCs, including pDCs (Fig. 8) (Allan et al., 2003; Allan et al., 2006; Audiger et al., 2017; Guery and Hugues, 2013; Iberg et al., 2017; Lukacs-Kornek et al., 2008; Osorio et al., 2015). However, peripheral tolerance does not rely exclusively on DCs. Emerging evidence demonstrates that lymph node stromal cells (LNSCs), among LECs, also play an important role in peripheral tolerance [our review in Appendix 1 (Humbert et al., 2016)]. The involvement of LNSCs, including LECs, and pDCs in peripheral T cell tolerance will be described later.

Figure 8. Peripheral T cell tolerance. Following negative selection in the thymus, SP thymocytes that successfully passed the selection process and thymus-derived regulatory T cells (tTregs) exit from the thymus and reach the periphery. Some self- reactive T cells manage to escape the mechanism of central tolerance and also enter the periphery. Hence, self antigen-specific T cell tolerance is maintained in the periphery, in the LNs. Exogenous antigens are acquired from peripheral tissues (yellow) by conventional dendritic cells (cDCs) and plasmacytoid DCs (pDCs) that subsequently migrate to the LNs where they present self-antigen peptides to autoreactive T cells. CD4+ and CD8+ T cell outcomes after antigen presentation by cDCs and pDCs are indicated in the figure. Antigen transfers and cell migration are respectively depicted in dashed and dotted arrows. Ags, Antigens; exo Ags, exogenous antigens; pTreg, peripherally-induced Treg; thym. cDC, thymus- resident cDC; tTregs, thymus-derived regulatory T cells. Adapted from Humbert, Hugues* and Dubrot*, Front Immunol, 2016 [Appendix 1 (Humbert et al., 2016)].

I.6. Unconventional antigen-presenting cells

I.6.a. An overview of unconventional and semi-professional APCs

As mentioned previously, cDCs can be considered as professional APCs. They are able to internalize exogenous antigens, constitutively express the antigen processing machinery and MHC-II, and express co-stimulatory molecules upon activation, and their main function is to initiate T cell responses (Kambayashi and Laufer, 2014). Nonetheless, several other APC type have been described as being able to cross-present via MHC-I and/or present via MHC-II antigens to CD8+ and/or CD4+ T cells, as well as to present endogenously-expressed self-antigens through MHC-I or MHC-II, including cells from the hematopoietic and non-hematopoietic compartments (Kambayashi and Laufer, 2014). We refer to these cells as “unconventional” APCs in this manuscript. A major difference with professional APCs is that they do not constitutively express MHC-II but its expression is upregulated upon maturation under certain conditions.

The term “unconventional” has been used by others to describe certain types of APCs (Crispe, 2011; Reynoso and Turley, 2009; Uto et al., 2018; Valitutti and Espinosa, 2010). Other terms used in the literature include “atypical” (Kambayashi and Laufer, 2014), “non-classical” (Costantino et al., 2012), “non-traditional” (LaSalle et al., 1992), “non-conventional” (Horst et al., 2016) and “alternative” (Moser and Brandes, 2006). Other terms, in opposition/reference to “professional” have been also used, such as “non-professional” (Lopes Pinheiro et al., 2016; Mehrfeld et al., 2018; Montealegre and van Endert, 2018; Pisapia et al., 2015; Wosen et al., 2018), “amateur” (Royer et al., 2018; Schuijs et al., 2019; Sprent, 1995), and “semi-professional” (Razakandrainibe et al., 2012). The term “unconventional” is defined as “not based on or conforming to what is generally done or believed” (Oxford dictionary) and is used in this thesis to describe APCs that are not professional APCs and for which the most studied function in the literature is not related to antigen presentation. “Unconventional” is a very subjective, and somehow arbitrary term, which, in the context of research highly depends on the chronology of scientific discoveries, the proportion of scientists working on a given topic compared to another (e.g. study about innate vs adaptive functions of pDCs), and the dogma in place at a given moment in time. In addition, it could be related to the ontogeny of cell types.

In their review, Kambayashi and Laufer asked whether any cell could replace professional APCs (Kambayashi and Laufer, 2014). Some studies have shown that DCs were necessary and sufficient to activate naïve T cells, while other studies showed that, under certain conditions, DCs were not required for the development of CD4+ T cell responses (Kambayashi and Laufer, 2014). Depending on the cell type, the location, and the immunological context, unconventional APCs play various roles in immunity and tolerance (Duraes et al., 2013; Kambayashi and Laufer, 2014; Mehrfeld et al., 2018; Reynoso and Turley, 2009).

Hematopoietic cells Mast cells, eosinophils and basophils have been described to express MHC-II under certain conditions and behave as APCs, although it is debated in the literature (Kambayashi and Laufer, 2014; Mehrfeld et al., 2018). Most of the studies were performed in vitro and there is no evidence that these cells can present in vivo. In addition, neutrophils have been reported to perform MHC- I-mediated cross-presentation or MHC-II-mediated presentation, inducing Th1 or Th17 polarization (Abi Abdallah et al., 2011; Beauvillain et al., 2007; Kambayashi and Laufer, 2014; Mehrfeld et al., 2018; Vono et al., 2017). Results are controversial and here also, most of the studies were carried out in vitro. In addition to their ability to differentiate into macrophages or inflammatory DCs, monocytes can migrate from the blood to LNs, acquire the expression of MHC-II, and present antigens as APCs per se (Jakubzick et al., 2017). Furthermore, innate lymphoid cells (ILCs) have been implicated in MHC-II-mediated antigen presentation to CD4+ T cells and drive Th2 cell responses (Kambayashi and Laufer, 2014; Mirchandani et al., 2014; Oliphant et al., 2014). Human CD4+ T cells themselves express MHC-II upon activation and present self-antigens to other CD4+ T cells (Costantino et al., 2012; Kambayashi and Laufer, 2014; LaSalle et al., 1992). Finally, megakaryocytes, the hematopoietic cells responsible for the maintenance of platelets, have been shown to uptake exogenous antigens, process and cross-present them to CD8+ T cells (Zufferey et al., 2017).

Non-hematopoietic cells Several cells from the non-hematopoietic compartment have been shown to perform MHC-I- mediated cross-presentation or MHC-II-mediated antigen presentation; hepatocytes, epithelial and endothelial cells, originating from different organs such as liver, skin, LNs and parenchymal

tissues (Duraes et al., 2013; Kambayashi and Laufer, 2014; Mehrfeld et al., 2018; Reynoso and Turley, 2009). Epithelial cells, including cells from the gastrointestinal tract, respiratory tract and cornea have been shown to present antigens through MHC-II (Kambayashi and Laufer, 2014; Royer et al., 2018; Wosen et al., 2018). These cells also include the APC type TECs, described in the “central tolerance” section (Anderson and Takahama, 2012; Klein et al., 2009; Roche and Furuta, 2015). Endothelial cells from many tissues and organs have also been implicated in antigen presentation, with consequences in various pathologies, such as transplantation and parasite infection (Al- Soudi et al., 2017; Kambayashi and Laufer, 2014; Pober et al., 2017; Razakandrainibe et al., 2012; Taflin et al., 2011). Antigen-presenting non-hematopoietic cells also include LNSCs, among which LECs, described later in this introduction (Duraes et al., 2013; Reynoso and Turley, 2009; Turley et al., 2010). Finally, aberrant acquisition of constitutive MHC-II expression has been reported in certain non- hematopoietic tumor cells, such as melanoma cells, and subsequent antigen presentation to CD4+ T cells (Donia et al., 2015).

Cross-dressing In addition to the presentation of antigens internalized or endogenously-expressed, cells (APCs or not) can acquire MHC/peptide complexes from other cells, without the need to internalize and process antigens, a phenomenon called cross-dressing (Nakayama, 2014). Transfers occur via direct cell-cell contact, called trogocytosis, or through secretion of membrane vesicles, such as exosomes, by the donor cell (Nakayama, 2014). Donor cells can be APCs or not, including infected and tumor cells, and recipient cells also include APCs, unconventional APCs and non- APCs (de Heusch et al., 2007; Dubrot et al., 2014; Kedl et al., 2017; Koble and Kyewski, 2009; Nakayama, 2014; Rouhani et al., 2015; Tamburini et al., 2014; Wakim and Bevan, 2011; Zhang et al., 2008). This phenomenon, reviewed by Nakayama, give recipient cells the ability to present MHC-I- or MHC-II-restricted antigens, while not necessarily able to internalize and/or process those antigens, or which do not have the intrinsic ability to express MHC-II (Nakayama, 2014).

Plasmacytoid DCs and LECs as unconventional APCs In this thesis, we consider pDCs and LECs as unconventional APCs, although these two distinct cell types cannot be considered on the same level regarding their properties to function as bona

fide APCs. For instance, in contrast with pDCs, LECs originate from the non-haematopietic compartment, they do not possess the ability to migrate, and they do not have the ability to upregulate the expression of co-stimulatory molecules, to cite a few examples (Humbert et al, 2016). Therefore, whether a cell type could be considered as “unconventional” APC, as opposed to “professional” APC, might depend on a spectrum rather than on a “all or nothing” concept. This spectrum could: - start from cells such as LECs, which arise from the non-haematopoietic compartment (ontogeny), are well known to play structural roles (main function), whose role as APCs was recently discovered (chronology of discoveries), and which are not able to express molecules considered important for professional APCs (attributes), such as co-stimulatory molecules - include pDCs, which, under certain circumstances (activation with specific stimuli) exhibit features that are not far from that of professional APCs (cDCs), regarding their antigen- presenting capacities. - and end with cDCs, the “professional” APCs, whose main function is to initiate T cell responses. Whether pDCs can be considered as “unconventional” APCs, as it is not a consensus, is discussed at the end of the description of antigen-presenting functions of pDCs, in the next section.

I.6.a. Plasmacytoid dendritic cells

I.6.b.i. Plasmacytoid dendritic cells: generalities

Human pDCs have been first described in the late 50´s by Lennert and Remmel, and further characterized by the groups of Liu and Colonna in the late 90´s (Grouard et al., 1997; Lennert and Remmele, 1958; Liu, 2005; Siegal et al., 1999). Murine pDCs were discovered few years leater (Asselin-Paturel et al., 2001). Since then, a lot of effort has been made by the scientific community in order to better understand the ontogeny, biology and functions of these cells, for which different names have been used, including interferon-producing cells, plasmacytoid T cells, plasmacytoid pre-DCs and plasmacytoid DCs (Alculumbre et al., 2018a; Cella et al., 2000; Grouard et al., 1997; Liu, 2005; Reizis et al., 2011b; Soumelis and Liu, 2006). “Plasmacytoid” and “dendritic” are mutually exclusive terms, in the sense that a cell cannot be both dendritic and oval-shaped like a plasma cell at the same time (Reizis et al., 2011b; Soumelis and Liu, 2006).

These words rather describe two sequentially distinct states of a same cell, which confer different functions (Jaehn et al., 2008; Reizis et al., 2011b; Soumelis and Liu, 2006). Plasmacytoid DCs link the innate and adaptive immunity. They are implicated in immune responses in several different ways, for example, by producing IFN-I and by performing MHC-I-mediated cross-presentation and MHC-II-restricted antigen presentation (Jaehn et al., 2008; Reizis et al., 2011a; Swiecki and Colonna, 2015) (Fig. 9). Depending on the immunological context, pDCs can be either tolerogenic or immunogenic (Guery and Hugues, 2013; Matta et al., 2010). In the following sections, the origins and roles of pDCs will be summarized, with an emphasis on functions indirectly and directly affecting T cell responses, especially antigen presentation.

Figure 9. The multiple functions of plasmacytoid dendritic cells. Plasmacytoid dendritic cells (pDCs) are implicated in both innate and adaptive immunity. Innate immune responses (upper panel). Upon viral infections, they produce type I interferon (IFN-I), leading to an anti-viral state that induces the expression of IFN-stimulated genes by infected cells, and their subsequent apoptosis. Furthermore, IFN-I, pDC-derived-IL-12 and IL-18 promote natural killer (NK) cell activation, and their effector functions, including the secretion of IFN-γ and target cell lysis. Adaptive immune responses (lower panel). The expression of the co-stimulatory molecules CD40, CD80 and CD86 by pDCs, in addition to major histocompatibility complex class I (MHC-I) and MHC-II expression, allow these cells to cross-prime CD8+ T cells, and to present antigens to CD4+ T cells. IFN-I and IL-12 secretion by pDCs promotes CD8+ T cell proliferation and effector functions and CD4+ T cell polarization into Th1 cells. Moreover, indoleamine 2,3-dioxygenase (IDO) and inducible T cell co- stimulator ligand (ICOS-L) expression by pDCs, as well as their secretion of transforming growth factor-β (TGFβ) and IL-6 support the commitment of regulatory T cells (Tregs) and Th17 cells, respectively. Interactions between pDCs and invariant NKT (iNKT) cells through OX40-OX40L and programmed cell death protein 1 (PD1)–PD1 ligand 1 (PDL1) dampen antiviral adaptive immune responses. In addition, pDCs impact B cell activation, the generation of plasma cells and the secretion of antibodies, via the secretion of IL-6 and IFN-I, and the expression of B cell-activating factor (BAFF) and proliferation- inducing ligand (APRIL). Plasmacytoid DCs also harbor TNF-related apoptosis-inducing ligand (TRAIL) and granzyme B, which leads to infected CD4+ T cell apoptosis, inhibition of T cell proliferation and tumor cell killing. Finally, pDCs produce chemokines that recruit immune cells to inflammation sites, such as CC-chemokine ligand 3 (CCL3), CCL4, CXC-chemokine ligand 8 (CXCL8) and CXCL10. Adapted from Swiecki and Colonna, Nat Rev Immunol, 2015 (Swiecki and Colonna, 2015).

I.6.b.ii. Ontogeny and heterogeneity of pDCs

Along with cDCs, pDCs constitute the heterogeneous DC family (Merad et al., 2013). Plasmacytoid DCs develop in the BM from early committed hematopoietic progenitors; pDCs can originate from the common lymphoid progenitor (CLP) or from the common DC progenitor (CDP), which also gives rise to cDCs (Anderson et al., 2018; Guilliams et al., 2014; Liu et al., 2009; Merad et al., 2013). The generation of the precursors for both pDCs and cDCs, i.e. pre- pDCs and pre-cDCs, requires FMS-related tyrosinase kinase ligand (FLT3L) (Karsunky et al., 2003). The transcription factor E2-2, selectively expressed by pDCs and also called transcription factor 4 (TCF4), is essential for pDC differentiation and maintenance, E2-2 deletion in mature pDCs leading to a switch towards cDC commitment (Cisse et al., 2008; Ghosh et al., 2010; Grajkowska et al., 2017). Other important transcription factors for pDCs include, Zeb2, which regulates pDC and cDC2 development, Irf8, shared with the cDC1 lineage, which is not essential for pDC development but whose deletion leads to an alteration of pDC phenotype and gene- expression profile, and myeloid translocation gene 16 (Mtg16), which represses the cDC- promoting transcription factor inhibitor of DNA binding 2 (Id2) (Ghosh et al., 2014; Scott et al., 2016; Sichien et al., 2016). Fully developed pDCs travel through the blood stream, from which they can reach the SLOs (Merad et al., 2013).

In mouse, pDCs express CD45, and the T cell and B cell markers CD4 and B220 (Asselin-Paturel et al., 2001). In addition, they express bone marrow stromal cell antigen 2 (BST2), also called plasmacytoid dendritic cell antigen 1 (PDCA-1), which is not fully pDC-specific as it can be upregulated by other immune cells upon inflammation, as well as the molecule sialic acid binding Ig-like lectin H (Siglec-H) (Blasius et al., 2006; Zhang et al., 2006). Human pDCs express CD123 (IL-3Rα), CD4 and CD45RA (Reizis et al., 2011a). In addition, they express blood dendritic cell antigen 2 (BDCA-2), also called CD303, although the expression of this marker is downregulated upon activation, and BDCA-4 (CD304), is, on the contrary, upregulated upon stimulation (Dzionek et al., 2000; Dzionek et al., 2001). However, recent studies using single-cell RNA-seq and cytometry by time-of-flight (CyTOF), have shown that BDCA-2, BDCA-4 and CD123 are also expressed by other DC subtypes (Alcantara-Hernandez et al., 2017; Alculumbre et al., 2018a; See et al., 2017; Villani et al., 2017). These new studies and the confusion they generated in the scientific community show the difficulty to define cell subsets uniquely based on cell surface marker expression, especially for cells like pDCs that are extremely plastic and whose phenotype highly depend on the immunological context (tissue, stimulus, inflammatory conditions). This is

why a classification depending on cell functions and morphology might be a stronger approach (Alculumbre et al., 2018a).

Plasmacytoid DCs form a heterogeneous population due to reasons including tissue specialization and developmental origins. In addition, interindividual heterogeneity has been observed in DC subsets, including pDCs. Indeed, the set of markers expressed by a given DC subtype varies greatly depending on the individuals (Alcantara-Hernandez et al., 2017). There is also heterogeneity across tissues (Alcantara-Hernandez et al., 2017). In addition, Zhang and colleagues characterized a subpopulation of human pDCs (CD2hi CD5+ CD81+), for which there is an equivalent population in mice, expressing classical pDC markers, producing little to no IFN- I and having increased ability to activate B and T cells in response to stimulation (Zhang et al., 2017). Furthermore, the group of Tussiwand demonstrated that pDCs arising from CDPs or CLPs exhibited functional differences in human (Rodrigues et al., 2018). Both subsets were able to produce type I IFN (IFN-I) but only CDP-derived pDCs could process and present antigens. Finally, Alculumbre et al. observed heterogeneity among human pDCs after stimulation, they described three distinct populations, based on the expression of PD-L1 and CD80, with functional differences (Alculumbre et al., 2018b). These populations were stable after secondary stimuli and did not depend on pre-existing variability before stimulation (Alculumbre et al., 2018b).

I.6.b.iii. Plasmacytoid DC functions independent of antigen presentation

Activation of pDCs Plasmacytoid DCs express toll-like receptor (TLR)7 and TLR9, whose expression in human is restricted to B cells and pDCs, while in mice, these receptors are also expressed by other cell types (Gilliet et al., 2008; Rothenfusser et al., 2002). TLR7 and TLR9 recognize nucleic acids in early endosomes; viral ssRNA for TLR7 and bacterial unmethylated CpG DNA for TLR9, both TLR7 and TLR9 also have been shown to sense patterns of the parasite Plasmodium (Bauer and Wagner, 2002; Demaria et al., 2014; Diebold et al., 2004; Gilliet et al., 2008; Lund et al., 2004; Rothenfusser et al., 2002; Yu et al., 2016). Sensing via TLR7 and TLR9 triggers the MyD88-PI3K- IRF7 signaling pathway (myeloid differentiation primary response gene 88; phosphoinositide 3- kinase; interferon-related factor 7), leading to IFN-I production (Diebold et al., 2004; Gilliet et al., 2008; Guiducci et al., 2008; Lund et al., 2004). It also can lead to maturation of pDCs, which relies on nuclear factor (NF)-κB pathway, with upregulation of MHC-II and co-stimulatory molecules, and secretion of pro-inflammatory cytokines (Gilliet et al., 2008). Synthetic analogs of

TLR7 and TLR9 ligands have been designed, such as Imiquimod and CpG olygodeoxynucleotides (CpG-ODN), respectively (Gilliet et al., 2008; Hemmi et al., 2002). Different classes of CpG exist, leading to different outcomes. Class A CpG-ODN (CpG-A) leads to the production of IFN-I by pDCs, while CpG-B induces the secretion of cytokines, including IL-12 and IL-6, and pDC maturation (Gilliet et al., 2008; Honda et al., 2005; Vollmer and Krieg, 2009). CpG-C promotes both IFN-I production and pDC maturation (Vollmer and Krieg, 2009). TLR7 and TLR9 normally do not sense self-nucleic acids (Demaria et al., 2014; Gilliet et al., 2008). Nonetheless, in certain autoimmune diseases, such as type 1 diabetes, psoriasis and systemic lupus erythematosus, host endogenous factors, such as self-nucleic acid-specific immunoglobulins or antimicrobial peptides, can break innate immune tolerance and enable the entry of self-nucleic acids into endosomes where they can trigger TLR7 or TLR9 signaling (Demaria et al., 2014; Diana et al., 2013; Gilliet et al., 2008; Henault et al., 2016; Lande et al., 2007; Panda et al., 2017; Sakata et al., 2018). In addition, TLR7 and TLR9 can recognize DAMPs. For instance, TLR9 can sense oxidized mitochondrial DNA, which contains unmethylated CpG repeats (Caielli et al., 2016). Conversely, impaired TLR7 or TLR9 signaling is implicated in persistent viral infections and cancer (Hirsch et al., 2010). In addition to nucleic acid sensing through TLR7 and TLR9, pDCs can also be activated by cytokines, such as IL-3 and granulocyte-macrophage colony-stimulating factor (GM-CSF) (Ghirelli et al., 2010; Grouard et al., 1997). Depending on the cytokine used to activate pDCs, the pattern of pDC cytokine production differs (Ghirelli et al., 2010; Grouard et al., 1997). Finally, pDCs can also be activated by CD40-L, a molecule expressed by activated T cells (Grouard et al., 1997).

Innate pDC functions having no direct link with T cell responses Plasmacytoid DCs have direct innate antiviral and antitumoral roles, and are also implicated in the activation/regulation of innate and adaptive immune cells (Fig. 9). Through their secretion of chemokines, pDCs can induce the recruitment of innate immune cells to sites of inflammation, including neutrophils, monocytes and macrophages (Swiecki and Colonna, 2015). As above-mentioned, pDCs are professional IFN-I producers (Reizis et al., 2011a). They also produce IFN-III (IFN-λ), although the roles and targets of this family of cytokines [reviewed in (Wack et al., 2015)] will not be detailed here. A crucial faculty of pDCs is to induce an anti-viral state through its production of IFN-I, leading to the expression of IFN-stimulated genes in infected cells and their subsequent apoptosis (Reizis

et al., 2011a; Swiecki and Colonna, 2015). IFN-I production, as well as that of pro-inflammatory cytokines, by pDCs is also implicated in NK cell activation (Dalod et al., 2003; Hervas-Stubbs et al., 2011; Swiecki and Colonna, 2015). In addition, IFN-I and IL-6 secretion, along with the expression of specific surface markers, by pDCs play a role in B cell activation, plasma cell generation and the secretion of antibodies (Menon et al., 2016; Swiecki and Colonna, 2015). Due to TRAIL and Granzyme B expression, pDCs have the ability to directly kill tumor cells (Drobits et al., 2012; Lombardi et al., 2015; Tel et al., 2014; Tel et al., 2012a).

Innate pDC functions having an impact on T cell responses Plasmacytoid DC innate roles can affect cDCs, as well as other APCs, and T cell recruitment and functions, therefore indirectly impacting T cell responses. On the one hand, by producing chemokines, pDCs can influence the migration of cDCs and T cells (Swiecki and Colonna, 2015). On the other hand, cDC and T cell functions can be modulated by the innate roles of pDCs. Plasmacytoid DCs have been found to regulate several cDC functions via the secretion of IFN-I, leading to different outcomes, which are context-dependent (Hervas-Stubbs et al., 2011; Swiecki and Colonna, 2015). For example, pDC IFN-I production modulates cDC maturation and their secretion of cytokines and enhances tolerance mediated by cDCs, in a mouse model of autoimmunity (Dalod et al., 2003; Prinz et al., 2008). Antigen uptake abilities of cDCs can also be regulated by pDCs (Kastenmuller et al., 2011). In addition, pDCs can promote cDC cross- presentation (Brewitz et al., 2017; Rogers et al., 2017; Schiavoni et al., 2013). Innate pDC functions can also affect T cell directly. For example, the expression of Granzyme B and TRAIL by pDCs contributes to the suppression of T cell expansion, and to the induction of the apoptosis of infected CD4+ T cell (Jahrsdorfer et al., 2010; Swiecki and Colonna, 2015). IFN- I production by pDCs has been shown to be important for the survival of CD8+ T cells, in a mouse model of viral infection (Swiecki et al., 2010). In addition the production of IFN-I and IL- 12 by pDCs drive CD8+ T cell activation and the polarization of Th1 cells, while the secretion of TGF-β and IL-6 are involved in Treg and Th17 cell commitment (Bonnefoy et al., 2011; Cella et al., 2000; Gautreau et al., 2011; Hervas-Stubbs et al., 2011; Swiecki and Colonna, 2015). Therefore, depending on the context, innate pDC functions lead to various outcomes on T cell responses, such as suppression, apoptosis or Treg polarization on the one hand, and activation, survival, proliferation and effector T cell polarization, on the other hand.

I.6.b.iv. Antigen-presenting functions of pDCs

Uptake of exogenous antigens Whether pDCs can uptake exogenous antigen to cross-present them via MHC-I to CD8+ T cells and present them through MHC-II to CD4+ T cells is still debated. In mice, Siglec-H and BST2 (PDCA-1) have been used as endocytic receptors allowing the internalization of exogenous antigens conjugated with anti-Siglec-H or anti-BST2 antibodies (Loschko et al., 2011a; Loschko et al., 2011b; Moffat et al., 2013; Zhang et al., 2006). In human, different pDC receptors have been used to deliver exogenous antigens to pDCs, such as CD32 (Fcγ receptor II), DEC205, DC immunoreceptor (DCIR) and BDCA-2, leading to successful antigen internalization (Benitez-Ribas et al., 2006; Meyer-Wentrup et al., 2008; Tel et al., 2011; Tel et al., 2013b). In addition, the uptake of exogenous antigen, e.g. encapsulated in microparticles, solubles, or from virus-infected cell material has also been observed in human cells (Di Pucchio et al., 2008; Lui et al., 2009; Ruben et al., 2018; Segura et al., 2013a; Tel et al., 2010; Tel et al., 2012a). Of note, pDCs were found by some groups to have the capacity to internalize exogenous antigens, although their endocytic abilities were lower than that of cDCs (Lui et al., 2009; Tel et al., 2012a). On the contrary, some studies concluded that pDCs could not acquire exogenous antigens (Bonaccorsi et al., 2014; Salio et al., 2004). Using pDCs from human tonsils, Segura and colleagues, however, observed that internalization of soluble exogenous antigens by pDCs was as efficient as for cDCs (Segura et al., 2013a). The observation that pDCs could not internalize exogenous antigens might therefore be due to the use of pDCs from peripheral blood in other investigations, which could have different properties from that of SLOs, which are the ones that present antigens in vivo (Bonaccorsi et al., 2014; Salio et al., 2004; Segura et al., 2013a). For instance, pDCs were found to be the most abundant subset of DCs in LNs, and to have the capacity to migrate from the periphery to LNs via the afferent lymphatics (Kohli et al., 2016; Segura et al., 2013a). Finally, human pDCs were shown to have the ability to process synthetic long peptides, which contain both MHC-I-restricted and MHC-II-restricted epitopes (Aspord et al., 2014b). Antigen processing was increased in the presence of TLR7 or TLR9 ligands.

MHC-I-restricted cross-presentation of exogenous antigens to CD8+ T cells With the exception of erythrocytes, all cell types in mouse and human express MHC-I molecules and can present endogenous antigens to CD8+ T cells (Neefjes et al., 2011). However, the presentation of exogenous antigens to CD8+ T cells, called cross-presentation, is a selective feature of certain types of APCs (Joffre et al., 2012; Kambayashi and Laufer, 2014). Cross-

presentation of exogenous antigens to CD8+ T cells by pDCs has been observed in human and mice (Nierkens et al., 2013; Villadangos and Young, 2008) (Fig. 9). Nonetheless, in contrary to cDCs that can cross-present exogenous antigens in inflammatory situations but also in the steady-state, pDCs acquire cross-presentation ability only upon activation. In human, the cross-presentation of exogenous antigens, either soluble, opsonized, encapsulated, cell-associated or from virus-infected cells, has been observed in several studies in vitro (Benitez- Ribas et al., 2006; Di Pucchio et al., 2008; Hoeffel et al., 2007; Lui et al., 2009; Segura et al., 2013a; Tel et al., 2013b; Tel et al., 2012a). Similar observations have been made in different mouse models, in vitro and in vivo, in which antigens were soluble, in particle, or delivered to pDCs via BST2 (Kool et al., 2011; Loschko et al., 2011b; Moffat et al., 2013; Mouries et al., 2008; Oberkampf et al., 2018). However, in other investigations, the cross-presentation of exogenous antigens by pDCs to CD8+ T cells could not be observed (Salio et al., 2004; Sapoznikov et al., 2007). For instance, antigen-targeting to pDCs through BST2 did not lead to CD8+ T cell response in vitro (Sapoznikov et al., 2007). Moreover, pDCs could elicit a CD8+ T cell response against exogenous peptides but not exogenous antigens, suggesting an issue in exogenous antigen internalization or processing, which was not elucidated (Salio et al., 2004). Interestingly, Gilliet and colleagues observed an induction of IL-10-producing CD8+ Tregs by culturing in vitro naïve human CD8+ T cells with allogeneic pDCs activated with CD40-L (Gilliet and Liu, 2002). These CD8+ T cells were anergic and had decreased cytotoxic activity. This suggested a potential implication for pDCs in CD8+ T cell tolerance. Discrepancies between the conclusions of distinct analyzes, on whether pDC can cross-present or not, might be due to different pDC maturation state (different stimulators used) or due to the method used for antigen targeting.

The intracellular pathways involved in antigen cross-presentation to CD8+ T cells are still unclear. The processing of exogenous antigens dedicated to cross-presentation is different from the mechanism of endogenous antigen-processing that relies on the proteasome and for instance, multiple cross-presentation pathways exist in cDCs (Joffre et al., 2012; Neefjes et al., 2011). The processing of exogenous antigens in pDCs did not require the transit of antigens to the cytosol, as it was observed to occur in endocytic organelles (Di Pucchio et al., 2008; Hoeffel et al., 2007). In addition, the processing machinery is dependent on the activation of pDCs through TLR ligands (Kool et al., 2011; Mouries et al., 2008). Finally, an implication of mitochondrial ROS production was recently demonstrated to be important for the ability of pDCs to cross-present (Oberkampf et al., 2018).

MHC-II-restricted antigen presentation to CD4+ T cells Plasmacytoid DCs express the promoter III (pIII) of CIITA, the master regulator of MHC-II expression, while cDCs express CIITA pI, hence, their pattern of MHC-II expression are differentially regulated (LeibundGut-Landmann et al., 2004; Reith et al., 2005). Unlike cDCs that express MHC-II constitutively, pDC MHC-II expression is upregulated upon activation (Kool et al., 2011; Sadaka et al., 2009; Young et al., 2008). In addition, in contrary to cDCs, the synthesis of MHC-II by pDCs continues long after stimulation and MHC-II molecules accumulate in antigen-loading compartments, therefore making pDC good candidates for prolonged MHC-II- restricted antigen presentation in chronic pathologies (Sadaka et al., 2009; Young et al., 2008). In addition, the endocytosis of MHC-II and antigen internalization also continues to occur after pDC maturation. Therefore, pDCs have the capacity to constantely renew peptide-MHC-II complexes exposed at their surface (Sadaka et al., 2009). Finally, the expression of MHC-II by pDCs is differentially regulated from their function of IFN-I-production; MHC-II expression is independent of IRF7, but relies on NFκB signaling (Sadaka et al., 2009). Several studies, in human and mouse, have shown that pDCs could present antigens to CD4+ T cells via MHC-II molecules (Benitez-Ribas et al., 2006; Kool et al., 2011; LeibundGut-Landmann et al., 2004; Moffat et al., 2013; Sapoznikov et al., 2007; Tel et al., 2010; Tel et al., 2012a) (Fig. 9). Depending on the immunological context, MHC-II-restricted antigen-presenting functions of pDCs can be either tolerogenic or immunogenic (Guery and Hugues, 2013). On the one hand, pDCs can induce anergy or deletion of CD4+ T cells, as well as pTreg differentiation or expansion of Tregs already differentiated. On the other hand, pDCs can induce the differentiation of naïve T cells into effector CD4+ T cells, or the polarization of antigen- experienced CD4+ T cells, such as Th1 or Th17 cells (Guery and Hugues, 2013).

In addition, as mentioned previously, heterogeneity has recently been observed among human pDCs and different groups have characterized pDC subpopulations with increased ability to induce CD4+ T cell expansion, compared with other pDC subpopulations (Alculumbre et al., 2018b; Rodrigues et al., 2018; Zhang et al., 2017).

In human, activation of pDCs with influenza virus along with CD40L led to the upregulation of MHC-II and co-stimulatory molecule expression by pDCs, and led to the stimulation of CD4+ T cells in vitro, with a Th1 polarization due to IL-12 and IFN-I production (Cella et al., 2000). In mice, Krug and colleagues found that pDCs could promote the expansion of antigen- experienced – non-polarized – CD4+ T cells and their polarization towards Th1 cells (Krug et al.,

2003). However, pDCs were unable to induce naïve T cell proliferation (Krug et al., 2003). Another study in mice, using in vitro-cultured pDCs or ex vivo splenic pDCs, showed Th1 or Th2 polarization of T cells cultured with these pDCs, depending on TLR stimulation and antigen dose; high or low antigen dose led to Th1 or Th2 polarization, respectively (Boonstra et al., 2003). In addition, in human, Alculumbre et al. characterized three pDC populations, based on the expression of PD-L1 and CD80, after stimulation with influenza virus (Alculumbre et al., 2018b). The population P3 (CD80+ PD-L1-) had increased capacity to induce the expansion of CD4+ T cells and led to the differentiation of naïve T cells into Th2 cells. In a mouse model of graft vs host disease (GvHD), adoptive transfer of antigen-loaded MHC-II- sufficient pDCs into MHC-II-deficient recipient mice led to the priming of IFN-γ+ CD8+ T cells and to GvHD induction, MHC-II-deficient mice being resistant to the disease (Koyama et al., 2009). Furthermore, antigen-targeting to pDCs through BST2 in mice led to increased IFN-γ production by CD4+ T cells, and subsequent antiviral or antitumoral effects (Loschko et al., 2011b). Finally, Mallat’s group, in collaboration with our group, demonstrated a pathogenic role of MHC-II-mediated antigen presentation by pDCs in a mouse model of atherosclerosis (Sage et al., 2014). Indeed, in the absence of MHC-II expression by pDCs, IFN-γ production by proatherogenic T cells was decreased, as well as the infiltration of T cells in lesions, showing an immunogenic role of pDC MHC-II-mediated antigen presentation in this context, with a Th1 polarization (Sage et al., 2014). Immunogenic pDCs can also induce the differentiation or promote the expansion of Th17 cells. In GvHD patients, Th17 cells and pDC counts were found to be increased in the intestinal mucosa, compared with healthy donors (Bossard et al., 2012). In addition, the number of pDCs in mucosal intestina correlated with clinical score. Although this analysis did not demonstrate that pDCs were directly implicated in the differentiation or expansion of Th17 cells, it suggests a potential link between those cells and a pathogenic role of pDCs in this context (Bossard et al., 2012). Interestingly, pDCs from the peripheral blood of healthy donors induced Th17 cell differentiation in vitro upon TLR7 ligation (Yu et al., 2010). In experimental autoimmune encephalomyelitis (EAE), a mouse model for multiple sclerosis, pDC depletion before immunization led to decreased severity and decreased Th17 cell counts, although a direct link between pDCs and Th17 cells was not demonstrated (Isaksson et al., 2009). Therefore, pDCs were immunogenic in the priming phase of EAE. On the contrary, pDC depletion a week after immunization increased the severity of the disease (Isaksson et al., 2009). Furthermore, TGF-β-treated pDCs induced the polarization of Th17 cells in vitro, due to increased IL-6 production by TGF-β-treated pDCs compared with non-treated pDCs,

subsequently leading to CD4+ T cells Th17 cells commitment over Tregs (Bonnefoy et al., 2011). Adoptive transfer of TGF-β-treated pDCs in a mouse model of rheumatoid arthritis increased the clinical score (Bonnefoy et al., 2011). Of note, a study using mice and rats showed pDCs were implicated in the modulation of the Treg/Th17 balance; pDCs can induce the conversion of Tregs into Th17 cells, induced by IL-6 production (Gautreau et al., 2011). Finally, our group showed that subcutaneous injection of the TLR9 ligand CpG-B, along a MHC-II-restricted antigenic peptide, in mice, induced Th17 differentiation in LNs draining the injection site (Guery et al., 2014). Th17 differentiation was abrogated in mice in which pDCs lacked MHC-II expression, showing that MHC-II-antigen presenting functions of pDCs were directly implicated in this mechanism. Altogether, these mouse and human studies, although not all demonstrating a direct link between MHC-II-restricted antigen presentation by pDCs and effector T cell differentiation or expansion, showed that in certain immunological contexts, or in the presence of specific molecules, pDCs antigen-presenting functions can be immunogenic and be involved in Th1 or Th17 cell polarization, with implications in antiviral and antitumoral immunity, as well as transplantation and autoimmunity.

On the contrary, several human and mouse investigations have observed a tolerogenic role of antigen-presenting pDCs, either by suppressing effector CD4+ T cells, or by inducing the differentiation or expansion of Tregs. Kolhi et al. suggested that intralymphatic injection of ovalbumin (OVA)-loaded pDCs led to tolerance in LNs, by inducing the abortive proliferation of OT-II cells. However, tolerance induction was not direct but mediated by endogenous APCs, since CD4+ T cells abortive proliferation was abrogated when antigen-loaded MHC-II-sufficient pDCs were transferred into MHC-II-deficient mice (Kohli et al., 2016). Nonetheless, many studies proposed a direct role for pDCs in the induction of tolerance via their ability of MHC-II-mediated antigen presentation. Of note, Kohli and colleagues used resting pDCs, in steady-state conditions. Plasmacytoid DCs have been observed to play a tolerogenic role in multiple mouse models, such as EAE, asthma, allergy to food antigen and organ transplant, in which pDCs were involved in the suppression of CD4+ T cell responses, by inducing the anergy or deletion of CD4+ T cells, by preventing the priming of these cells, or by mechanisms not specified (Chappell et al., 2014; de Heer et al., 2004; Goubier et al., 2008; Liu et al., 2011; Loschko et al., 2011a; Takagi et al., 2011). For instance, antigen-targeting to pDCs through Siglec-H dampened EAE severity, by decreasing

CD4+ T cell expansion and Th1/Th17 cell polarization, however, not converting these cells into Tregs or other T cells with a regulatory phenotype (Loschko et al., 2011a). In addition, many studies have shown an implication of pDCs in tolerance induction by promoting the differentiation or expansion of Tregs. In human, activation of pDCs with CpG-B led to a cell-cell contact-dependent differentiation of naïve T cells into Foxp3+ Tregs that inhibited the proliferation of allogeneic CD4+ T cells, in vitro (Moseman et al., 2004). Zhang and colleagues characterized a subpopulation of human pDCs (CD2hi CD5+ CD81+), with an equivalent population in mice, which has an enhanced capacity to induce the expansion of CD4+ T cells, in absence of stimulus, leading to the differentiation of Tregs from naïve T cells (Zhang et al., 2017).

In mouse, pDCs activated with IL-3 and CD40-L, or with thymic stromal lymphopoietin (TSLP) have been shown to generate tTregs in the thymus. Thymus-derived Tregs primed by pDCs had a different phenotype from that induced by cDCs, with an increased IL-10- and decreased TGF- β-production by pDC-primed tTregs, suggesting a non-redundant role for cDCs and pDCs in tolerance induction in the thymus (Hanabuchi et al., 2010; Martin-Gayo et al., 2010). In a mouse model of GvHD, pDCs migrated to the gut, promoting Treg functions and subsequent decreased disease severity (Hadeiba et al., 2008). Moreover, pDCs were shown to induce oral tolerance by supporting Tregs (Dubois et al., 2009). Finally, our group showed that MHC-II-restricted antigen-presenting functions of pDCs mediate tolerance in EAE development, by promoting Tregs (Duraes et al., 2016; Irla et al., 2010; Lippens et al., 2016). Mice in which pDCs were depleted or lacked MHC-II expression exhibited exacerbated EAE scores. Two main mechanisms, relying on the expression of inducible T cell co-stimulator ligand (ICOS- L) or indoleamine 2,3-dioxygenase (IDO) by pDCs, have been described for the induction of Tregs by pDCs and will be decribed in the following paragraph. In human, the expression of ICOS-L by pDCs, a molecule that is not expressed by cDCs, was shown to induce the differentiation of naïve CD4+ T cells into IL-10-producing Tregs (Ito et al., 2007). Human pDCs were found to express high levels of IDO (Chen et al., 2008). Activation of pDCs with TLR9L led to upregulation of MHC-II, costimulatory molecules and IDO, leading to a subsequent generation of pTregs. Adding IDO inhibitor in the pDC-CD4+ T cell in vitro co- culture abrogated Treg generation by pDCs (Chen et al., 2008). Moreover, high levels of IDO expression has been observed in rheumatoid arthritis patients responding positively to therapy, compared with non-responding ones (Kavousanaki et al., 2010). In addition, responding patients

had increased numbers of circulating pDCs. Responding patient pDCs led to ex vivo differentiation of allogeneic CD4+ T cells into Tregs, which had the ability to suppress autologous naïve CD4+ T cell proliferation (Kavousanaki et al., 2010). Furthermore, using pDCs from human immunodeficiency virus (HIV)-infected patients or pDCs infected with the virus in vitro, HIV was shown to induce the expression of IDO by pDCs, leading to Treg induction and inhibition of CD4+ T cell proliferation (Boasso et al., 2007; Manches et al., 2008). In mice, Fallarino and colleagues showed increased immunosuppressive properties of pDCs upon CD200R engagement on these cells, leading to increased pDC IDO expression and subsequent suppression of antigen-specific CD4+ T cell responses in vivo (Fallarino et al., 2004). More recently, a fungal infection was shown to induce pDC infiltration into the lungs and pDC maturation, as well as increasing the number of IDO-expressing pDCs. Depletion of pDCs led to decreased Treg frequency and decreased disease severity (Araujo et al., 2016). The immunogenicity of pDCs was enhanced in IDO-deficient mice. Finally, our group showed a cross-talk between pDCs and Tregs, in the context of autoimmunity (Lippens et al., 2016). During EAE development, Treg generation and suppressive functions relied on IDO and MHC- II expression by pDCs, as demonstrated using IDO-deficient mice and mice in which pDCs lacked MHC-II expression. Conversely, the expression of IDO by pDCs was found to be dependent upon cell-cell interaction with Tregs (Lippens et al., 2016). Altogether, these investigations show that pDCs, in certain immunological context, behave as tolerogenic APCs by directly inhibiting effector CD4+ T cell responses or by promoting Tregs.

Overall, whether MHC-II-mediated antigen-presenting functions of pDCs are immunogenic or tolerogenic tightly depends on the immunological context.

Plasmacytoid DC functions in the tumor context will be described in chapter A (introduction section).

Plasmacytoid DCs as unconventional APCs? As mentioned previously, whether pDCs can be considered as “unconventional” APCs, is not a consensus in the scientific community. Plasmacytoid DCs have been referred to as professional APCs in different publications (Chan et al., 2012; Loschko and Krug, 2012; Tel et al., 2012b; Yanofsky et al., 2013). Whether they can be considered as “professional” APCs may also depend on the definition employed. As above-mentioned, the ability of pDCs to uptake and process

exogenous antigens, and to (cross-)present them to CD8+ and CD4+ T cells is debated in the literature. In addition, cDCs and pDCs exhibit distinct antigen uptake and processing machinery. One major difference between cDCs and pDCs, is that pDC acquire the ability to cross-present MHC-I restricted antigens upon inflammation, i.e. they require a stimulus, while cDCs exhibit this function in the steady-state (Kool et al., 2011; Mouries et al., 2008). Regarding MHC-II restricted antigen presentation, MHC-II is constitutively expressed in high levels in human pDCs. In contrast, in mouse, the expression of MHC-II is upregulated by pDCs upon activation. Furthermore, the expression of MHC-II is differentially regulated between cDCs and pDCs (LeibundGut-Landmann et al., 2004; Reith et al., 2005). Regarding the functions of pDCs, their ability to prime naïve T cells has been debated in the literature (Cella et al., 2000; Krug et al., 2003), and still nowadays, the APC functions of pDCs are not fully accepted by the whole scientific community (See et al., 2017; Villani et al., 2017). In addition, the particularity of pDCs is that they exist in distinct functional states (“plasmacytoid” and “dendritic”). The plasmacytoid state is considered as professional IFN-I producer and this activity is primordial in antiviral innate immunity (Liu, 2005; Reizis, 2019). Only the dendritic state is capable of antigen presentation to T cells, and depends on cell maturation due to activatory signals (Liu, 2005; Reizis, 2019). Once in the dendritic state, pDCs were shown to uptake antigens as efficiently as cDCs, to be the most abundant DC subset in LNs (Kohli et al., 2016; Segura et al., 2013a), and they may exhibit antigen-presenting capacities really close to that of cDCs. Nonetheless, the fact that pDCs can exist in these distinct functional states, one of which is not able to function as APC, could itself be considered as unconventional. Finally, in vivo, the relative contributions of cDCs and pDCs to T cell responses are not similar. During acute inflammation, with high antigen dose and severe inflammation, the contribution of pDCs to T cell responses may be hidden by that of cDCs. Indeed, cDCs may present antigens and be activated more efficiently compared to pDCs. In addition, the relative contribution of pDCs may increase in chronic settings, such as in anti-tumor immunity and autoimmunity, when the antigen source persists, a context in which cDCs may acquire an exhausted phenotype regarding their antigen-presenting functions. In this context, in contrast, pDC continuously renew peptide/MHC-II complexes at their surface.

I.6.a. Lymphatic endothelial cells

I.6.c.i. Different subsets of lymph node stromal cells

Lymph node stromal cells (LNSCs) LNs are SLOs that drain the lymph flow and orchestrate adaptive immunity. LNSCs are essential to the structure and functions of LNs. These cells arise from non-hematopoietic origin (CD45- negative) and are radioresistant (D'Rozario et al., 2018). They comprise four main subsets, which can be distinguished by the expression of CD31 (also known as platelet endothelial cell adhesion molecule 1; PCAM-1) and glycoprotein (gp)38 (also known as podoplanin or PDPN), as reported after a transcriptomic analysis of murine LNSCs subsets in steady-state and inflammatory conditions (Malhotra et al., 2012). Subsets include fibroblastic reticular cells (FRCs: CD31- gp38+), blood endothelial cells (BECs: CD31+ gp38-), lymphatic endothelial cells (LECs: CD31+ gp38+) and double negative cells (DNCs: CD31- gp38-) (Fig. 10) (D'Rozario et al., 2018; Malhotra et al., 2012; Turley et al., 2010). LNSC subsets are heterogeneous and new approaches should allow a better characterization of these populations (Gentek and Bajenoff, 2017; Rodda et al., 2018). For instance, the group of Cyster recently performed an analysis of murine non-endothelial LNSCs, thus excluding BECs and LECs, by single-cell RNA-seq, which allowed the identification of nine clusters, among which different types of T-zone reticular cells, marginal reticular cells, follicular dendritic cells and perivascular cells (Rodda et al., 2018). Nonetheless, in this introduction, only FRCs, BECs and LECs will be mentioned, the other LNSC subsets [reviewed in (D'Rozario et al., 2018; Gentek and Bajenoff, 2017)] will not be addressed. FRCs, LECs and BECs have long been thought to function as a simple scaffold. However, these cells are involved in T cell responses in many different ways (Mueller and Ahmed, 2008; Turley et al., 2010).

Figure 10. Specific location of the lymph node stromal cell subsets. Lower left panel. Fibroblastic reticular cells (FRCs) are located in T cell zones, in the lymph node (LN) paracortex, where dendritic cells (DCs) and T cells encounter. Upper right panel. Efferent and afferent lymphatic vessels are composed by lymphatic endothelial cells (LECs). Leukocytes and lymph flow into the LNs through these vessels. Lower right panel. Blood endothelial cells (BECs) line blood vessels. They form high endothelial venules (HEVs), highly specialized structures that are surrounded by perivascular sheath and basal lamina. Adapted from Turley*, Fletcher* and Elpek*, Nat Rev Immunol, 2010 (Turley et al., 2010).

Fibroblastic reticular cells (FRCs) FRCs reside in the T cell zone, in the cortex of LNs, where T cells and DCs interact (Fig. 10) (D'Rozario et al., 2018; Gentek and Bajenoff, 2017; Turley et al., 2010). Besides their role as simple scaffold, they are affecting T cell responses in many ways (Brown and Turley, 2015; D'Rozario et al., 2018; Siegert and Luther, 2012). Of note, they ectopically express PTAs and can present them directly to T cells, a feature that will be described in the introduction of Chapter B (Turley et al., 2010). A new subset of LN-FRCs, called medullary FRCs, has recently been described. This population, observed by dynamic imaging approaches, was located in the LN medullary cords and implicated in the homeostasis of plasma cells (Huang et al., 2018). The ‘classical’ FRC population to which we refer to later in this thesis is characterized as CD31- gp38+ (D'Rozario et al., 2018; Gentek and Bajenoff, 2017; Malhotra et al., 2012; Turley et al., 2010). A very recent study performed by the group of Fletcher analyzed the role of human FRCs in T cell functions (Knoblich et al., 2018). Using in vitro T cell differentiation and proliferation assays and tissue (tonsils and LNs)-based in situ assay, in the presence or not of molecules inhibiting

specific pathways, they showed that FRCs inhibit CD4+ and CD8+ T cell proliferation and differentiation into effector cells. This inhibition was due to four distinct mechanisms occurring simultaneously: IDO, adenosine A2A receptor (A2AR), prostaglandin E2 (PGE2) and TGF-β receptor (TGF-βR).

Blood endothelial cells (BECs) BECs line the blood vessels. They form high endothelial venules (HEVs), which are highly specialized structures surrounded by perivascular sheath and basal lamina (Fig. 10) (Turley et al., 2010). They are involved in the transport of soluble factors, nutrients and blood-borne cells, they secrete chemokines and HEVs are implicated in the regulation of lymphocyte influx (D'Rozario et al., 2018; Gentek and Bajenoff, 2017).

Lymphatic endothelial cells (LECs) LECs compose the afferent and efferent lymphatic vessels (LVs), which themselves, along with lymphoid tissues, constitute the lymphatic system (Fig. 10) (Turley et al., 2010). These vessels drain the extracellular compartment of almost all tissues. It conveys lymph fluid, which comprises proteins and immune cells drained from interstitial tissues, and helps to get rid of toxins and undesired components from the body. When migrating to infection sites, lymphocytes follow the lymphatic system, facilitating immune responses directed towards potential harms. Lymphatics have often been underestimated by researchers, but its importance in the control of the immune system beyond leukocyte trafficking regulation is now recognized, thanks to recent discoveries. A brief history of initial discoveries related to the lymphatic system can be found in our review [Appendix 1 (Humbert et al., 2016)]. The understanding of the multiple functions of lymphatics has rapidly evolved, due to the characterization of LEC specific markers, including the transcription factor prospero homeobox protein 1 (Prox-1) or the surface protein lymphatic vessel endothelial hyaluronan receptor 1 (Lyve-1), which are not expressed by other endothelial cell types. Many studies have later shown that LECs affect immune responses in several ways [reviewed in (Randolph et al., 2017)]. In the following sections, we describe the current understanding of LEC immunoregulatory properties, and in particular, LEC ability to impact T cell responses.

I.6.c.ii. Ontogeny and development of LECs

It is currently well accepted that LECs differentiate in the developing veins, from specialized angioblasts, during embryogenesis (Choi et al., 2012; Nicenboim et al., 2015). Ink intravenous

administration in pig embryos unraveled the development of lymph sacs from embryonic vein budding and the identification of vascular endothelial growth factor receptor-3 (VEGFR-3) further confirmed that LECs and BECs arise from a common origin (Kaipainen et al., 1995). In the adult, VEGFR-3 is specifically expressed in LECs (Kaipainen et al., 1995; Oliver, 2004). Nonetheless, during embryonic development, it is also expressed by angioblasts and developing veins (Dumont et al., 1998; Kaipainen et al., 1995; Wigle et al., 2002). By modulating the expression of vascular endothelial growth factor C (VEGF-C), the main ligand of VEGFR-3, studies further demonstrated the importance of this molecule for the development of LVs, including that of meningeal LVs (Antila et al., 2017). Indeed, the overexpression of VEGF-C led to lymphatic sprouting and lymphangiogenesis (Jeltsch et al., 1997; Kukk et al., 1996; Saaristo et al., 2002). The discovery of Prox-1 gene supported the theory of a venous origin of LVs. Prox-1 deletion in mice induces a lack of LEC differentiation, and consequently, there is a complete absence of lymphatic system in Prox-1 knockout mice (Wigle et al., 2002; Wigle and Oliver, 1999). The selective expression of Prox-1 in particular cell types of the embryonic veins at E9.5 leads to lymphatic polarization and to LEC signature imprinting (Petrova et al., 2002; Wigle et al., 2002; Wigle and Oliver, 1999). Transcriptome analyzes demonstrated a close similarity between the gene expression profiles of LECs and BECs. Nonetheless, Prox-1 specifically regulates genes inversely regulated, in a cell-type specific manner (Hirakawa et al., 2003; Podgrabinska et al., 2002). For instance, all venous endothelial cells could possibly give rise to lymphatic or blood endothelium, as shown by the induction of reprogramming in BECs when Prox-1 was overexpressed in these cells (Petrova et al., 2002). Prox-1 is also required for the maintenance of lymphatic phenotype, even after development (Choi et al., 2012; Johnson et al., 2008; Nicenboim et al., 2015; Yaniv et al., 2006). However, it has been shown that the origin of the lymphatic vasculature, depending on the organ, can be slightly different. Using cell-fate mapping, a recent study suggested that the developing cardiac lymphatics would combine LECs derived from venous and non-venous origins (Klotz et al., 2015). Additional details on LEC development can be found in our review [Appendix 1 (Humbert et al., 2016)]. During development, LEC specification involves functional and structural differences between blood and lymphatic systems. The lymphatic system is a linear and blind-ended circuit, whereas blood vasculature is circular and closed. Lymphatic system capillaries drain interstitial fluids from tissues and organs in the periphery, due to LEC organization in the terminal lymphatics. The uptake of macromolecules, interstitial fluid and cells is allowed thanks to the thin-walled capillary

vessels, which are highly permeable, composed of a LEC single layer but not covered by smooth muscle cells or pericytes, and have no or a tiny basement membrane (Alitalo et al., 2005). Lymphatic capillaries harbor “button-like” or discontinuous junctions, in which the gaps between junctions function as sites where leukocytes enter the vessels (Baluk et al., 2007; Tammela et al., 2007). Anchoring filaments, which sense any change in interstitial pressure during inflammation, allow the linkage between terminal lymphatic capillaries and the surrounding extracellular matrix. This leads to the aperture of vessel lumen and junction, which facilitates the uptake of fluids derived from tissues. Deeper, LVs change from a morphology enabling fluid drainage to one of “collector”, specialized in lymph transport. Smooth muscle cells and pericytes surround the collecting lymphatics, which possess a basement membrane exhibiting continuous “zipper-like” junctions. The presence of valves enables lymph circulation and also prevents retrograde flow (Alitalo et al., 2005; Baluk et al., 2007).

I.6.c.iii. Major types of LECs

Afferent LVs are present in most of the vascularized organs, with some exceptions, such as the spleen, in which the presence of efferent lymphatics is debated, and the BM (Golub et al., 2018; Kellermayer et al., 2011; Mebius and Kraal, 2005; Petrova and Koh, 2018). There is an heterogeneity within the lymphatic system, depending on their anatomical locations, LECs have different functions and possible roles (Petrova and Koh, 2018; Ulvmar and Makinen, 2016). Lymphatic drainage has been suggested for decades to be implicated in local immune responses (Friedlaender et al., 1973). While the role of DCs in initiating and eliciting activation or tolerance of the immune system has drawn most of the attention, modulation of adaptive immunity by lymph drainage has remain unexplored for long.

Skin LECs Keratin 14 (K14)-VEGFR-3-Ig mice, in which soluble VEGFR3-3-Ig is expressed through the keratin promoter, lack skin LVs, and have a reduced fluid clearance, in a model of dermal vaccination (Thomas et al., 2012). In addition, local lymphatic drainage is critical in these mice for acquired tolerance and humoral immunity, whereas T cell responses are delayed but mainly unaffected.

Liver sinusoidal endothelial cells (LSECs) LSECs have a high ability to filter fluids, particles and solutes from hepatic circulation, occupy a vast area that is exposed to blood, which transports external food- and commensal bacteria-

derived antigens (Limmer et al., 2000; Wisse, 1970). LSECs are important players for the maintenance of liver immune homeostasis. Among other functions, they are able to modulate DC-mediated antigen presentation, and they can themselves present antigens to T cells (Lukacs- Kornek, 2016; Mehrfeld et al., 2018; Shetty et al., 2018; Wohlleber and Knolle, 2016).

LECs in the central nervous system (CNS) The dogma of the CNS as an immune privilege lacking a lymphatic system has long persisted, despite the knowledge that T cells are patrolling the brain in homeostasis, a phenomenon called immune surveillance (Ellwardt et al., 2016; Mohammad et al., 2014). Important advances has been made recently with the (re)-discovery of a lymphatic vasculature in rodent CNS in a specific area of the meninges that line the dural sinuses, allowing cerebrospinal fluid (CSF) drainage to the deep cervical LNs (Louveau et al., 2015b; Mezey and Palkovits, 2015). Markers specific for LECs, including Lyve-1, Prox-1 and gp38 were expressed by these meningeal LVs. This achievement was concomitant with the (re)-discovery of dural LVs, which drain CSF from the subarachnoid space and brain interstitial fluid through the glymphatic system, also important for the clearance of macromolecules from the CNS (Aspelund et al., 2015). The connection between the meningeal lymphatic and glymphatic systems has been reviewed by Louveau and colleagues (Louveau et al., 2017). Meningeal LVs have been visualized by magnetic resonance imaging (MRI) in human and non-human primates (Absinta et al., 2017). Overall, these findings have induced a reassessment of the immune privilege dogma and brought insights in the initiation and progression of neurological disorders (Bower and Hogan, 2018; Engelhardt et al., 2017; Louveau et al., 2015a). For instance, in a very recent study using a mouse model of Alzheimer’s disease, it was demonstrated that an alteration of meningeal lymphatics is linked with amyloid-β deposition in meninges and an aggravation of the disease, showing the importance of this CNS waste clearance system (Da Mesquita et al., 2018). In contrast, intracerebral delivery of VEGF-C in elderly mice led to an improvement of memory and learning performances (Da Mesquita et al., 2018; Wen et al., 2018). In addition, brain mural LECs, which express LEC-specific markers, have been recently identified. They surround meningeal blood vessels and are crucial for the vascularization of meninges (Bower et al., 2017).

Tumor-associated LECs Lymphangiogenesis, the development of lymphatics in the tumor microenvironment (TME), has been extensively studied, in particular the involvement of tumor lymphatics in disease spreading, called metastasis. For instance, most human carcinomas and melanomas metastasize via the

lymphatic system (Leong et al., 2011). Similarly to the presence of cancer-associated fibroblasts and BECs, the presence of LECs in the tumor is associated with bad clinical prognosis in many types of cancer (Turley et al., 2015; Ziani et al., 2018). Therapies that aim at blocking tumor lymphangiogenesis are nowadays considered as promising for the treatment of these malignancies (Dieterich and Detmar, 2016; Stacker et al., 2014). Increasing evidence highlight the impact of tumor-associated (TA)-LECs in dampening anti-tumor immunity (Swartz, 2014). In this regard, PD-L1 has been found to be upregulated in TA-LVs, likely due to increased IFN-γ levels in the TME (Dieterich et al., 2017). In vitro, PD-L1 blockade led to increased T cell activation by LECs. These data suggest that TA-LECs are directly implicated in T cell inhibition (Dieterich et al., 2017).

I.6.c.iv. Regulation of LN-LEC proliferation and survival

LN expansion and contraction are important processes of immune responses, allowing adaptation to the influx of immune cells into the LNs. The proliferation and survival of LN- LECs have been shown to be regulated by T cells, more specifically their IFN-γ production [reviewed in (Yeo and Angeli, 2017)]. Using interferon-α/β receptor (IFNAR)-/- and Pdl1-/- mice, a recent analysis showed that IFN-I induces PD-L1 expression by LECs, which in turn limits LEC division during a viral infection but promote their survival (Lucas et al., 2018). DCs also have been implicated in the architectural remodeling of the LN stromal compartment, among which LECs, for example by regulating the secretion of VEGF-C by other cell types, which is necessary for LEC growth (Acton and Reis e Sousa, 2016; Dasoveanu et al., 2016; Malhotra et al., 2013).

I.6.c.v. LECs impact peripheral T cell responses through mechanisms independent of antigen presentation

T cell immunity comprises the generation of pathogen-specific effector responses in order to protect against a vast range of invaders, without causing any undesired damage to self-tissue. Naïve T cells scan for their cognate antigen constantly. Nonetheless, this challenging task only takes place into well-organized SLOs, such as LNs, Peyer’s patches (PPs) and the spleen, due to the very low frequency of T cells that are specific for a given peptide/MHC complex (Moon et al., 2007; Obar et al., 2008). Blood-borne and tissue-derived antigens are both contained in SLOs, which facilitates the encounter of naïve T cells with their cognate antigens, and subsequently helps T cell activation and differentiation. The following paragraph describes the different

pathways by which T cell are impacted by LECs, a topic that has been reviewed by Card et al. (Card et al., 2014).

Delivery of antigens and DC migration to the LNs As mentioned previously, LNs are linked to lymphatics that drain peripheral fluids derived from tissues. Therefore, LECs facilitate the passive entry of tissue-derived antigens by connecting draining LNs and tissues, those antigens can subsequently be acquired, processed, and presented by LN-resident DCs to T cells, which enter the LNs via HEVs (Roozendaal et al., 2009; Sixt et al., 2005). LN-resident DCs immediately sample soluble antigens, while particles that carry antigens, including micro-vesicles, apoptotic bodies and exosomes, and that have not been uptaken by subcapsular sinus macrophages, flow to the medullary sinuses, where DCs can scan them (Gerner et al., 2015). In addition, LECs facilitate the migration of tissue-resident DCs into LNs (Russo et al., 2013; Teijeira et al., 2014). The migration of DCs from tissues to draining LN LVs is a major way for antigen presentation and naïve T cell activation. DCs enter afferent lymphatics independently of integrin-mediated adhesion, via preformed portals (Lammermann et al., 2008; Pflicke and Sixt, 2009). However, the expression of adhesion molecules in LECs can be upregulated upon inflammation, depending on the stimulus, which further favors the access of DCs to LVs (Johnson et al., 2006; Russo et al., 2013; Vigl et al., 2011). Moreover, CC-chemokine ligand (CCL)21 secreted by LECs in afferent lymphatics upon inflammation has been shown to be important, depending on the stimulus, for DC egress from tissue and entry into lymphatics (Russo et al., 2016; Teijeira et al., 2014; Vigl et al., 2011). Moreover, the expression of C-type lectin receptor 2 (CLEC2) by DCs supports their migration towards LNs through lymphatics, thanks to an interaction with its ligand gp38, expressed both by FRCs and LECs (Acton et al., 2012). Finally, it was recently shown that the entry of tissue-resident DCs into lymphatics and transit to the lumen, is dependent on a cell-cell contact enabled by the interaction between hyaluronan and Lyve-1, expressed by DCs and LECs, respectively (Johnson et al., 2017). This mechanism was demonstrated to be crucial for the resolution of inflammation, in a model of myocardial infarction (Vieira et al., 2018).

Regulation of DC functions Tissue-resident DCs that have captured peripheral antigens migrate via afferent lymphatics into LNs, by a mechanism depending on CC-chemokine receptor (CCR)7. However, not only the lymphatic system supports the migration of DCs from tissues to LNs, but it also induces

functional and phenotypic changes in DCs, due to close interactions between DCs and LECs (Malhotra et al., 2013). Contacts between DCs and tumor necrosis factor (TNF)-α-stimulated LECs induce reduced expression of co-stimulatory molecules by DCs in vitro, which therefore impairs the ability of DCs to induce T cell proliferation (Podgrabinska et al., 2009). In addition, the regulation of DC functions by LECs depends on interactions between intercellular adhesion molecule (ICAM)-1 on LECs and CD11b (Mac-1) on DCs (Podgrabinska et al., 2009). Of note, LECs can inhibit the function of lipopolysaccharide (LPS)-activated DCs, which further suggests a role for LECs in the regulation of the resolution phase of inflammation. A study recently showed that LECs act as reservoirs of PTAs, which are subsequently uptaken by DCs and presented to T cell, inducing their anergy, thus contributing to peripheral T cell tolerance (Rouhani et al., 2015).

T cell Homeostasis T cell migration in the LNs is conducted by FRC-secreted CCL21 and CCL19 (Luther et al., 2002). Nonetheless, the maintenance of naïve and memory T cells in the SLOs depends to a high degree on IL-7. LECs are an important source of IL-7 in vivo, along with FRCs, and therefore involved in the regulation of T cell homeostasis and their entry into SLOs (Link et al., 2007). Using IL-7-GFP (green fluorescent protein) knock-in mice, it was shown that IL-7 is highly expressed in LECs from both LNs and tissues, while it is moderately expressed in FRCs from LNs (Hara et al., 2012; Miller et al., 2013). In addition, LECs have been demonstrated to be an important source of IL-7, both in murine and human LNs (Onder et al., 2012). Moreover, in addition to the production of IL-7, LECs also express the IL-7Rα of the IL-7 receptor and CD132, which suggests a potential role for IL-7 as a lymphatic drainage mediator (autocrine). For instance, LECs stimulated with IL-7 in vitro induce lymphangiogenesis in mice cornea, while lymphatic drainage in IL-7Rα-/- mice is compromised (Iolyeva et al., 2013). In addition, the upregulation of IL-7 by LECs and FRCs is crucial for the reconstruction of LNs and their remodeling, after avascular transplantation or viral infection (Onder et al., 2012). These studies suggest that IL-7 secretion in the LNs after resolution of inflammation may be implicated in the homeostasis of memory T cells. In this regard, IL-7 supports the development, proliferation and survival of memory CD8+ T cells (Schluns et al., 2000; Schluns and Lefrancois, 2003).

Egress of T cells from the LNs The specificities of the lymphatic vasculature that allow T cell trafficking from tissue to tumor- draining lymph nodes (TdLNs), through afferent lymphatics, as well as through efferent LVs to

return to the blood circulation from the LNs have been reviewed by the group of C. Halin (Hunter et al., 2016; Schineis et al., 2018). T cell egress from the LNs depends on sphingosine-1-phosphate (S1P) receptor 1 (S1PR1) expression. By using mice that selectively lack S1P in LECs, Cyster and colleagues have demonstrated that LECs are an important in vivo source of S1P in the LNs, which allows the egress of T cells from the LNs and PPs (Pham et al., 2010). The expression of S1PR1 is downregulated in lymphocytes circulating in the blood and upregulated in the LNs. Egress from the LNs is promoted by interactions between LECs that produce S1P and T cells that express S1P1R, which overcome CCR7-mediated retention signals (Grigorova et al., 2009; Pham et al., 2008). Despite the fact that LECs express low levels of S1P in the steady-state, S1P secretion is upregulated in LECs in the medullary sinus upon inflammation mediated by PAMPs or DAMPs. This suggests that LECs which highly express S1P can support the egress of T cells from the LNs in pathogenic contexts. On the contrary, in situation of sterile - non-infectious - inflammation, LECs that produce moderate levels of S1P might rather decrease T cell effector functions, by promoting the retention of T cells in the LNs

Migration of T cells in lymphatics In addition to their function of tumor cell transporter, emerging evidence is in favor of important roles played by TA-lymphatics in T cell migration. For instance, the modulation of TA-lymphatic expansion can affect both metastatic and primary tumor progression. In the case of solid tumors, lymph flow from tumors is intense, which drives elevated interstitial flow in the stroma of tumors and enhances the lymphatic drainage from the tumor to the TdLNs (Swartz, 2014; Swartz and Lund, 2012). Therefore, it is possible that, in combination with a suppressive cytokine environment, enhanced drainage of tumor antigens could facilitate tumor antigen-specific T cell dysfunction, such as apoptosis and anergy. Moreover, the lymph supports cells that migrate from tissues, particularly CCR7+ DCs, which is critical for the initiation of anti-tumor immune responses (Roberts et al., 2016). T cell infiltration in the tumor is a major step in anti-tumor immunity. The infiltration of CTLs is associated with a good prognosis, while Treg or naïve T cell infiltration correlates with a poor clinical outcome (Fridman et al., 2012; Galon et al., 2006). In this regard, CCL21 expression in tumors facilitates immune escape and tumor progression, which may partly be explained by an increase in the recruitment of naïve T cells (Shields et al., 2010). Although TCR-transgenic tumor-infiltrating naïve T cells could be activated in situ, it is unlikely that it would induce efficient effector T cell differentiation, due to the suppressive TME

(Thompson et al., 2010). Likewise, whether CCL21-secreting LECs participate in this effect and how they contribute to the TME tolerogenic properties remain to be determined. We have previously shown that VEGF-C, a lymphangiogenic growth factor, which is secreted in the tumor promotes tolerance to melanoma in mice, by inducing the deletion of tumor antigen-specific CD8+ T cells (Hirosue et al., 2014; Lund et al., 2012). These results are in favor of a new role for lymphatics played in tumor development, suggesting that tumor lymphatic endothelium could be a target for immunomodulation. A study supporting these hypotheses demonstrated that TdLN- FRCs, after tumor-derived factor exposure, are able to adapt in multiple ways to harbor characteristics that are associated with immunosuppression, including reduced IL-7, CCL19 and CCL21 secretion (Riedel et al., 2016). Whether it is also the case for LECs in the TdLNs is still an open question. On the other hand, using K14-VEGFR3-Ig mice, which lack dermal LVs, Swartz’s group showed that the absence of lymphatics within melanoma tumor leads to decrease leukocyte infiltration and enhanced tumor growth, with potential implications for immunotherapies (Lund et al., 2016). In this regard, lymphangiogenesis blockade using anti- VEGFR-3 antibodies inhibited response to immunotherapy, due to a lack of naïve T cell infiltration into the tumor (Fankhauser et al., 2017). The authors suggested that lymphangiogenesis is important for a favorable response to immunotherapy, by supporting tumor infiltration of naïve T cells, which are subsequently locally activated.

I.6.c.vi. Antigen-presenting abilities of LECs: uptake of exogenous antigens and presentation to T cells

By controlling antigen availability, the lymphatic system constitutes one of the primary immune response checkpoints (Hirosue and Dubrot, 2015; Randolph et al., 2017). Therefore, it is not surprising that LECs display various mechanisms for antigen acquisition and processing, knowing they have early access to any given antigen (Fig. 11). Indeed, recent investigations revealed that antigen trafficking can be observed at various levels, not only the classical concept of LECs as lymph carriers. Complex interactions between DCs and LECs are involved in bidirectional antigen exchanges, which may serve, ultimately, to modulate the immune response magnitude (Fig. 11) (Dubrot et al., 2014; Kedl et al., 2017; Kedl and Tamburini, 2015; Rouhani et al., 2015; Tamburini et al., 2014). Indeed, LECs have a function of antigen archiving; they are able to capture and archive antigens, which can later be acquired by hematopoietic cells, directly or through LEC apoptosis (Kedl et al., 2017; Kedl and Tamburini, 2015; Tamburini et al., 2014). DCs also acquire antigens endogenously-expressed by LECs (Rouhani et al., 2015). Conversely, LECs acquire peptide/MHC complexes from DCs (Dubrot et al., 2014).

Acquisition of exogenous antigens LECs exhibit an active endocytotic capacity (Fruhwurth et al., 2013; Sixt et al., 2005); they can capture exogenous molecules and, depending on their location, process antigens for cross- presentation and cross-priming of antigen-specific CD8+ T cells (Fig. 11) (Hirosue et al., 2014; Lund et al., 2012). Of note, LN-LECs loaded with antigens have been shown to be able to cross- prime antigen-specific CD8+ T cells, by mechanisms depending on TAP (Hirosue et al., 2014). Antigen-loaded LECs induce T cell apoptosis, the most probable explanation being the lack of co-stimulatory molecule expression. Indeed, LECs do not express the co-stimulatory molecules CD40, CD80 and CD86 after TLR ligation or in the presence of TNF-α or interferon (IFN)-γ (Dubrot et al., 2014; Tewalt et al., 2012). On the contrary, they upregulate MHC-II and the immune-stimulatory molecules CD48 and herpes virus entry mediator (HVEM) in the presence of these cytokines, as well as the expression of the co-inhibitory molecule PD-L1 (Dubrot et al., 2014; Fletcher et al., 2010; Norder et al., 2012; Tewalt et al., 2012). In this regard, antigen cross-presentation by LSECs leads to CD8+ T cell tolerance in the liver, using a mechanism involving the expression of PD-L1 (Diehl et al., 2008; Limmer et al., 2000; von Oppen et al., 2009). Interestingly, in the absence of inflammation, LSEC-educated T cells that survived had a phenotype of antigen-experienced memory-like cell, in the SLOs (Bottcher et al., 2013). In addition, LSEC-primed memory T cells could be reactivated in an antigen-specific manner in vitro and in vivo, and could contribute to a viral challenge (Bottcher et al., 2013). Studies have reported that LSECs are also able to present MHC-II-restricted antigens to CD4+ T cells, leading to different outcomes, including Treg differentiation (Kruse et al., 2009; Wittlich et al., 2017). The antigen-presenting ability of LSECs and its outcome on CD8+ and CD4+ T cell responses have recently been reviewed by different groups (Lukacs-Kornek, 2016; Mehrfeld et al., 2018; Shetty et al., 2018; Wohlleber and Knolle, 2016). In the case of LECs, whether they directly participate in anti-viral CD4+ T cell responses, by presenting MHC-II-restricted antigens, is still under debate. As described above, LECs act as antigen reservoir upon viral infections (Kedl et al., 2017; Kedl and Tamburini, 2015; Tamburini et al., 2014) (Fig. 11). However, the genetic ablation of MHC-II molecules in LN radioresistant stromal cells resulted in a longer maintenance of antigen-specific CD4+ T cells (Abe et al., 2014).

Figure 11. Antigen acquisition and presentation by lymphatic endothelial cells. There are various pathways by which lymphatic endothelial cells (LECs) acquire antigens (Ag) and by which those antigens are loaded onto major histocompatibility complex (MHC) molecules. Complex mechanisms allow the transfer of antigens between dendritic cells (DCs) and LECs, in both directions. DCs use LECs as antigen reservoirs and uptake those antigens. Conversely, LECs are able to acquire peptide/MHC-II complexes (DC-derived antigen in yellow) at the surface of DCs, in a cell-cell contact- dependent mechanism. Exosomes derived from DCs might also be involved. Peripheral tissue-restricted antigens (PTAs) (pink) that are expressed by LECs can be loaded onto MHC-I molecules. However, the intracellular pathways accounting for PTA degradation remains to be deciphered. In addition, whether PTAs can be incorporated in MHC-II compartments is still debated. LECs are also able to acquire exogenous antigens (lymph-borne and tumor-derived), which can be incorporated in MHC-I pathway, by a mechanism involving transporter associated with antigen processing 1 (TAP-1). Associated references are depicted: 1. (Tamburini et al., 2014); 2. (Rouhani et al., 2015); 3. (Kedl et al., 2017); 4. (Dubrot et al., 2014); 5. (Hirosue et al., 2014); 6. (Lund et al., 2012). Adapted from Humbert, Hugues* and Dubrot*, Front Immunol, 2016 [Appendix 1 (Humbert et al., 2016)].

Cellular antigen transfers As described previously, several cell types, from hematopoietic and non-hematopoietic origins, are able to express MHC-II and to interact with CD4+ T cells in the periphery (Duraes et al., 2013; Hirosue and Dubrot, 2015; Kambayashi and Laufer, 2014). LECs are a non-professional APC type expressing MHC-II, under the presence of IFN-γ. For instance, the expression of

MHC-II in LN-LECs has been observed, both at the transcriptional and protein levels (Dubrot et al., 2014; Fletcher et al., 2010; Malhotra et al., 2012). Using transgenic mouse models that lack different CIITA promoters, we have previously shown that the levels of MHC-II molecules on the surface of LECs and other LNSCs, at the steady state, represent a combination of basal activity, which is IFN-γ-inducible, and peptide/MHC-II complexes acquired from DCs (Dubrot et al., 2014). Captured MHC-II molecules were loaded with antigens derived from DCs, licensing LECs to induce anergy and increased antigen-specific CD4+ T cell apoptosis (Fig. 11). The absence of measurable efficient T cell responses has for long been a major difficulty that prevented the characterization of the effect of antigen-presentation by LECs on CD4+ T cell responses. Similarly to the case of CD8+ T cell responses, the lack of costimulatory molecules, including CD80 and CD86, and the constitutive PD-L1 expression by LECs prevent the priming of functional effector CD4+ T cells. In this respect, it has been shown in human in vitro that LN- LECs do not possess the ability to induce the proliferation of allogeneic CD4+ T cells (naïve or memory), even in the presence of IFN-γ (Norder et al., 2012). As mentioned previously, in the immune system, membrane exchange between cells are not rare (Davis, 2007). Peptide/MHC-I and peptide/MHC-II complexes can be transferred between tumor cells or infected cells and DCs, between mTECs and DCs, and also between DCs (de Heusch et al., 2007; Kyewski et al., 2000; Wakim and Bevan, 2011; Zhang et al., 2008). The transfer of antigens can be a peptide exchange on cell surfaces. Epitopes can directly bind to the cell surface, or to MHC molecules in early endosomes, where MHC-I and MHC-II are receptive to the binding of lymph-borne peptides (Griffin et al., 1997). This is of particular relevance in the context of tolerance to self. Indeed, recent investigations demonstrated that lymph peptidome in human contains mainly self- peptides, such as ones derived from protein extracellular processing (Clement and Santambrogio, 2013). Exosomes have also been shown to be involved in the transfer of peptide/MHC-II complexes from DCs to LNSCs, and the possibility that they could contribute to alternative antigen-trafficking cannot be excluded (Dubrot et al., 2014) (Fig. 11). However, as mentioned previously, the transfer of antigens between DCs and LECs is bidirectional; in addition to the transfer of antigens captured and archived by LECs, the transfer of LEC endogenously-expressed PTAs to hematopoietic cells has also been described (Kedl et al., 2017; Rouhani et al., 2015; Tamburini et al., 2014) (Fig. 11). Indeed, LECs endogenously-express PTAs that can be transferred to DCs or directly presented to CD8+ T cells, inducing their tolerization, the presentation to and impact on CD4+ T cells still being a matter of debate (Humbert et al., 2016; Rouhani et al., 2015). In the analysis performed by Rouhani and colleagues, neither cytoplasmic nor membrane-bound PTAs were presented by LECs in a direct manner to induce antigen-

specific CD4+ T cell responses. This has been attributed to the expression of H2-M in LECs, which is lower compared with professional APCs, H2-M being required for the binding of peptides onto the MHC-II groove. However, H2-M expression is upregulated upon inflammation (Dubrot et al., in press1). Peptides derived from LEC-expressed PTAs are instead loaded onto MHC-II in DCs (Rouhani et al., 2015). Although the mechanisms by which antigen transferred are enabled are still under examination, it has been reported that it does not depend on the recognition of apoptotic cells or the phagocytosis of DCs. Therefore, there is a close relationship and communication between LECs and professional APCs that allows MHC-II presentation.

The role of PTA endogenous expression and direct presentation to T cells by LECs in peripheral tolerance to self will be described in Chapter B (introduction section).

1 Dubrot et al., in press (Life Science Alliance)

II. THESIS AIM

A tightly regulated balance between immune activation and tolerance is required in order for the immune system to mount efficient immune responses without inducing adverse effects, such as autoimmunity. T cell responses, which play key roles in adaptive immunity, are modulated by APCs. As mentioned earlier, cDCs are professional APCs. They are able to internalize exogenous antigens, constitutively express MHC-II and the antigen-processing machinery, can express co- stimulatory molecules following activation, and their main function is to initiate T cell responses (Kambayashi and Laufer, 2014). Several other cell types are able to cross-present antigens via MHC-I to CD8+ T cells and/or present antigens through MHC-II to CD4+ T cells as unconventional APCs, including cells from the hematopoeitic, such as pDCs, and non-hematopoietic compartments, such as LNSCs, leading to various outcomes on T cell responses, depending on the type of unconventional APC and on the immunological context (Kambayashi and Laufer, 2014). Plasmacytoid DCs are professional IFN-I-producing cells, which also have the ability to (cross)- present antigens via MHC-I and MHC-II molecules (Alculumbre et al., 2018a; Reizis et al., 2011a; Reizis et al., 2011b; Swiecki and Colonna, 2015). These cells are extremely plastic and, depending on the immunological context, their antigen-presenting functions can be either tolerogenic or immunogenic (Guery and Hugues, 2013). LNSCs, among which LECs, were long thought to function as simple scaffolds (Turley et al., 2010). However, new roles have emerged for these cells in the regulation of T cell responses, including direct implication in antigen presentation (Humbert et al., 2016; Turley et al., 2010).

The aim of this thesis was to better characterize the role of these two cell types, pDCs and LECs, as unconventional APCs, in the modulation of peripheral T cell responses. The immunological contexts, i.e. anti-tumor immunity (lack of immune activation) and autoimmunity (lack of tolerance), as well as specific aims, are described in Chapter A and Chapter B, respectively.

III. Chapter A

III.1. Introduction

III.1.a. Tumor immunity

III.1.a.i. Cancer and tumor immunity

Definition, epidemiology and biology of cancer Cancer, also called malignant tumor, refers to a wide range of diseases affecting any part of the organism, and is defined by a rapid abnormal cell growth and a potential to spread to other parts of the body, or in other words, to metastasize2, 3. There are several different types of cancers, with lung, breast and colorectal cancers being the most common. It is the second leading cause of death worldwide, although its incidence and the cancer types vary greatly by country. The global cancer incidence, mortality and prevalence (GLOBOCAN) study estimated that 18 million people had been diagnosed with cancer and about 10 million persons died from it, in 2018 (The, 2018). It is currently estimated that, in their life, 1/5 men and 1/6 women will be diagnosed with cancer, and 1/8 men and 1/10 women will die from the disease (The, 2018). Moreover, the incidence of cancer is on the rise. Indeed, by 2030, 13 million people will die each year from cancer, with 3/4 of deaths in middle- and low-income countries (The, 2018). Cancer arises from a multi-stage process that leads to normal cell transformation into malignant cells resulting from genetic predispositions and environmental factors, including physical, chemical, and biological carcinogens (Danaei et al., 2005; Rudolph et al., 2016; The, 2018). Major risk factors include diet, lack of physical activity, and the consumption of tobacco and alcohol (The, 2018). Depending on the type/subtype of cancer, the selective contribution of genes and environment to the development of cancer differs. Regarding the biology of cancer, specific characteristics, acquired during tumor development, constitute the hallmarks of cancer (Hanahan and Weinberg, 2011). Malignant cells sustain a proliferative signaling, escape processes that suppress their growth and resist against cell death. In addition, they allow replicative immortality, promote angiogenesis, and support invasion and metastasis. Factors underlying these mechanisms are genome instability and tumor-promoting inflammation. Lastly, cellular metabolism reprogramming and immune escape come into play

2 https://www.cancer.gov/about-cancer/understanding/what-is-cancer 3 http://www.who.int/en/news-room/fact-sheets/detail/cancer

(Hanahan and Weinberg, 2011). This last hallmark, evading anti-tumor immunity, will be described later. Tumors are not constituted of simple masses of malignant cells but comprise many different type of other - non-malignant – cells (Balkwill et al., 2012; Binnewies et al., 2018; Hui and Chen, 2015; Mbeunkui and Johann, 2009). This includes infiltrating immune cells as well as recruited stromal cells, such as cancer-associated fibroblast, mesenchymal cells, or cells constituting blood and LVs (Balkwill et al., 2012; Turley et al., 2015). In addition, features including nutrient availability, pH and oxygen tension are modified compared with non-invaded organs (Lyssiotis and Kimmelman, 2017). Altogether, the interaction between tumor cells and recruited cells constitute the tumor microenvironment (TME) (Balkwill et al., 2012; Hanahan and Weinberg, 2011; Hui and Chen, 2015; Mbeunkui and Johann, 2009). The TME, which varies greatly depending on the type and stage of cancer, plays a crucial role in carcinogenesis initiation and progression (Balkwill et al., 2012; Binnewies et al., 2018; Hui and Chen, 2015; Mbeunkui and Johann, 2009).

Mouse models for cancer research There is a wide range of mouse models for cancer research. Tumor cells can be injected/implanted in mice, leading to orthotopic or heterotopic models (Zitvogel et al., 2016). In orthotopic models, tumor cells are injected/implanted into the organ from which originated the cancer or they reach this organ through specific mechanisms when injected in the bloodstream. In heterotopic models, the tumor cells grow in an organ that is different from the one from which the cancer originated. Genetically-engineered mice that develop tumors spontaneously or upon induction (tamoxifen Cre/LoxP system), and carcinogen-induced models also have been established (Cheon and Orsulic, 2011; Day et al., 2015; DuPage et al., 2009; Zitvogel et al., 2016). Finally, humanized mouse models have been developed (Landgraf et al., 2018; Zitvogel et al., 2016). Each model has its own advantages and drawbacks, such as recapitulating more or less accurately specific types of human cancer, or taking more or less time for the tumor to develop. In our study, we used heterotopic mouse models, in which tumor cells were injected subcutaneously.

Anti-tumor immunity Several innate and adaptive immune cells, as well as numerous molecules, are implicated in the recognition and destruction of cancer cells, a phenomenon called immunosurveillance (Zitvogel et al., 2006). The theoretical and simplified cycle of anti-tumor immunity encompasses seven main steps (Fig. 12) (Chen and Mellman, 2013, 2017). Dying tumor cells release antigens that are

uptaken by APCs, which subsequently migrate via the lymphatics to the TdLNs, where they present these antigens to T cells. T lymphocytes, including CTLs, travel through the bloodstream and infiltrate the tumor, where they recognize tumor cells. CTLs exert their cytotoxic activity, leading to further antigen release. LN-like tertiary lymphoid structures, in which lymphoid and stromal cells accumulate, have been described in human cancer and are most of the time associated with good prognoses (Engelhard et al., 2018; Fridman et al., 2012; Goc et al., 2013; Joshi et al., 2015). Anti-tumor immunity is influenced by environmental, host and tumor factors, from which depend the magnitude and the kinetic of the anti-tumor immune response (Chen and Mellman, 2013, 2017). The TME negatively affects the immune system by three major mechanisms: immunoediting, immunosuppression and immunoevasion (Chen and Mellman, 2013, 2017; Dunn et al., 2004; Zitvogel et al., 2006). Mesenchymal, stromal and cancer cells play an important role in shaping the tolerogenic TME (Dunn et al., 2004; Turley et al., 2015; Zitvogel et al., 2006). Immune cells are greatly affected by the TME and conversely, immune cells can also contribute to tolerance in the TME, including tumor-associated (TA)-neutrophils, TA-macrophages, Tregs and myeloid-derived suppressor cells (MDSCs), which secrete anti-inflammatory cytokines (Zitvogel et al., 2006).

Figure 12. Anti-tumor immunity cycle. The induction of anti-tumor immunity is a self-propagating cycle, which leads to a gradual increase of immunostimulatory factors that enhance T cell responses, in principle. This cyclic process also involves immunoinhibitory factors, leading to mechanisms of immunoregulation that can prevent or limit anti- tumor immunity. The cycle comprises 7 main steps, it starts with antigen release from tumor cells and ends with tumor cell killing. This scheme depicts each step, including the primary cell types that are implicated and the anatomic location. APCs, antigen presenting cells; CTLs, cytotoxic T lymphocytes. Adapted from Chen and Mellman, Immunity, 2013 (Chen and Mellman, 2013).

III.1.a.ii. Cancer immunotherapies

A growing range of cancer immunotherapies Numerous types of cancer immunotherapies have been developed, aiming at inducing a robust tumor antigen-specific immune response [reviewed in (Galluzzi et al., 2014)] (Fig. 13). Passive immunotherapies include: monoclonal antibodies (mAbs) that target tumor cells; adoptive transfer of T cells that are selected/modified/expanded/activated ex vivo; and oncolytic viruses, which are non-pathogenic viruses selectively infecting tumor cells and having a direct cytotoxic activity (Galluzzi et al., 2014) (Fig. 13). Active immunotherapies include many different strategies, among which vaccines, such as DC-based and peptide- or DNA-based vaccines (Mellman et al., 2011; Yang, 2015). Of note, vaccines based on pDCs have been developed and will be described in the section describing the role of pDCs in tumors (Aspord et al., 2012; Tel et

al., 2013a; Tel et al., 2012b). Active immunotherapies also comprise immunostimulatory cytokines and chemotherapeutic/radiotherapeutic treatments that lead to an immunogenic cell death. They also include immunostimulatory mAbs, such as the ones blocking the immune checkpoint CTLA-4 and PD-1, and inhibitors of immunosuppressive metabolism, such as the enzyme IDO (Mellman et al., 2011; Yang, 2015). Finally, PRR agonists, including NLR and TLR agonists, are also used as immunotherapies, alone or in combination with other treatments (Galluzzi et al., 2014; Temizoz et al., 2016) (Fig. 13). Many other strategies have been developed but will not be described here.

Figure 13. A wide variety of cancer immunotherapies During the last three decades, a lot of cancer immunotherapies have been developed, such as monoclonal antibodies (mAbs) targeting the tumor or with immunomodulatory properties; anti-cancer vaccines based on peptides, DNA or dendritic cells (DCs); oncolytic viruses; pattern recognition receptor (PRR) agonists; cytokines with immunostimulatory properties; agents inducing an immunogenic cell death; immunosuppressive metabolism inhibitors; adoptive cell transfers. 1MT, 1-methyltryptophan; APC, antigen-presenting cell; IDO, indoleamine 2,3-dioxigenase; IFN, interferon; IL, interleukin; IMiD, immunomodulatory drug; NLR, NOD-like receptor; TLR, Toll-like receptor. Adapted from Galluzzi et al., Oncotarget, 2014 (Galluzzi et al., 2014).

Local administration of TLR ligands

Immunotherapies are most of the time administered systemically, hence often linked with damaging effects. Therefore, the local administration of immunotherapies, including intratumoral, topical, intranodal, intracranial and intradermal, is considered highly promising as it could solve many aspects that limit immunotherapeutic treatments (Fransen et al., 2013; Marabelle et al., 2014). Local administration requires lower dosage to reach the needed concentration, leading to decreased drug levels in the organism, hence, preventing autoimmune risks, and increasing the strength of the anti-tumor response (Fransen et al., 2013; Weiden et al., 2018). In this respect, one of the aims of locally administered immunotherapy is to overcome the TME inhibitory faculties to “warm up” the tumor and subsequently boost its immunogenicity, leading to the induction of effector immune cells. A possible approach is the local administration of TLR ligand agonists, converting the “cold” into “hot” TME, reactivating the immune cells directly at the site of injection and therefore boosting the potential for a strong immune response. As mentioned previously, PRRs, among which TLRs, sense danger signals through the recognition of PAMPs and DAMPs (Galluzzi et al., 2014; Thaiss et al., 2016). Activation of TLRs initiates signaling cascades leading to pro-inflamatory outcomes, such as the activation of NFκB and the secretion of pro-inflammatory cytokines, and promote the maturation of APCs (Galluzzi et al., 2014; Thaiss et al., 2016). The nature of the cytokines produced and of the immune response induced depends on the targeted TLR. In some cases, TLR agonists are instrumental to boost the efficacy of immune responses induced by other interventions such as radiotherapies, chemotherapies or immunotherapeutic strategies (Galluzzi et al., 2014; Mellman et al., 2011; Temizoz et al., 2016; Yang, 2015). In other cases, these agonists can be used as an immunotherapy per se (Galluzzi et al., 2014; Marabelle et al., 2014; Temizoz et al., 2016). Some TLR agonists have been approved by the American food and drug administration. This includes the vaccine Bacillus Calmette-Guérin (BCG) that contains a TLR2/4 agonist, used in the treatment of a subtype of bladder carcinoma, and imiquimod, a TLR7 agonist, used, for example, in the treatment of a subtype of skin cancer (Galluzzi et al., 2014; Hoffman et al., 2005). Of note, some malignant cells express TLRs, their cognate agonists might, therefore, directly modulate tumor cell functions (Galluzzi et al., 2014; Huang et al., 2008).

III.1.b. pDCs in tumor immunity

III.1.b.i. pDCs in tumor immunity: generalities

Plasmacytoid DCs have been found to infiltrate tumors in several types of cancer (Alculumbre et al., 2018a; Demoulin et al., 2013). This includes head and neck cancer, ovarian cancer, breast cancer and melanoma (Alculumbre et al., 2018a; Aspord et al., 2013; Demoulin et al., 2013; Hartmann et al., 2003; Labidi-Galy et al., 2011; Labidi-Galy et al., 2012; Sisirak et al., 2012; Treilleux et al., 2004). In several studies, infiltration of pDCs within the tumor correlated with poor prognoses (Aspord et al., 2013; Demoulin et al., 2013; Labidi-Galy et al., 2011; Labidi-Galy et al., 2012; Treilleux et al., 2004). In addition, pDC levels in the peripheral blood of non-small cell lung carcinoma (NSCLC) patients are increased compared with healthy donors, with higher levels of pDCs for late-stage compared to early-stage patients (Shi et al., 2014). Plasmacytoid DCs could therefore be involved in the pathogenesis of this cancer type. Of note, pDC levels were increased in smoking patients and in squamous cell carcinoma, the major subtype of NSCLC (Shi et al., 2014). Higher counts and frequencies were found in the BM of multiple myeloma (MM) patients, compared with healthy individuals (Chauhan et al., 2009). In ovarian cancer, a preferential recruitment of pDC in tumors was observed, as compared with peripheral blood, as pDC levels were important in the tumor but decreased in peripheral blood, in comparison with healthy individuals (Labidi- Galy et al., 2011). Recruitment of pDCs into the tumor has been suggested to be CCR6- dependent in melanoma and CCR6/CCR10-dependent in breast cancer, due to chemokines produced by cells in the TME (Charles et al., 2010; Labidi-Galy et al., 2011). In ovarian cancer, the production of CXC-chemokine ligand (CXCL)12 by tumor cells leads to the recruitment of pDCs into the tumor (Zou et al., 2001). Plasmacytoid DCs associated to tumors have been shown to acquire tolerogenic functions, which will be described in the following sections (Alculumbre et al., 2018a; Aspord et al., 2013; Chauhan et al., 2009; Demoulin et al., 2013; Lombardi et al., 2015; Sisirak et al., 2012). This tolerogenic phenotype is induced by molecules produced in the TME, such as TGF-β, IL-10, TNF-α, and IL-3 (Alculumbre et al., 2018a; Beckebaum et al., 2004; Ray et al., 2017; Sisirak et al., 2013; Terra et al., 2018).

III.1.b.ii. pDC innate functions in tumor immunity

IFN-I is an important molecule for the induction of potent anti-tumor immune responses, although it has a dichotomous effects, since prolonged signaling of this molecule can lead to immune dysfunctions (Demoulin et al., 2013; Dunn et al., 2004; Minn, 2015; Snell et al., 2017; Zitvogel et al., 2015). Several cancer immunotherapies are efficient only in the presence of IFN-I, which has been associated with favorable outcomes (Demoulin et al., 2013; Diamond et al., 2011; Rautela et al., 2015; Wang et al., 2011; Zitvogel et al., 2015). As previously mentioned, pDCs are professional IFN-I-producing cells (Reizis et al., 2011a). The ability of TA-pDC to secrete IFN-I is impaired in many types of cancers (Demoulin et al., 2013; Hartmann et al., 2003; Labidi-Galy et al., 2011; Perrot et al., 2007; Sisirak et al., 2012; Terra et al., 2018). TA-pDC capacity to secrete other cytokines, including TNF-α, macrophage inflammatory protein (MIP)-1β, IL-6 and regulated on activation, normal T cell expressed and secreted (RANTES), was also found altered in tumors (Labidi-Galy et al., 2011). Furthermore, pDCs were shown to directly interact with MM cells, leading to the production of IL-3 and subsequent increased MM growth, which in return also stimulates pDC survival (Ray et al., 2017). However, pDCs that are not immersed in the TME and are activated in vitro/ex vivo and injected directly into tumors retain the ability to induce antitumor responses (Liu, 2008). Moreover, pDCs can have a direct tumoricidal activity, they have the potential to directly induce the apoptosis of tumor cells (Drobits et al., 2012; Lombardi et al., 2015; Tel et al., 2014; Tel et al., 2012a).

III.1.b.iii. pDC antigen-presenting functions in tumor immunity

MHC-I and MHC-II antigen presentation by pDCs in the tumor context Whether pDCs were shown to induce the differentiation or expansion of T cells with a regulatory phenotype, or the suppression of effector T cell proliferation, pDCs associated with tumors and their MHC-I- and MHC-II-mediated antigen-presenting functions have been associated with poor clinical outcomes (Lombardi et al., 2015). In regard with MHC-I-mediated antigen-presenting functions, pDCs isolated from peripheral blood have been shown to have the ability to acquire membrane patches from cancer cells that harbored tumor antigenic peptide/MHC-I complexes, a mechanism resembling trogocytosis (Bonaccorsi et al., 2014). This occurred in a cell-cell contact-dependent manner and led to the presentation of exogenous antigens to CD8+ T cells. In addition, pDCs extracted from human colon cancer exhibited epithelial cell markers at their surface, suggesting this membrane transfer also happened in vivo (Bonaccorsi et al., 2014).

In human ovarian cancer, TA-pDCs have been suggested to promote IL-10-producing CD8+ Tregs (Wei et al., 2005). In addition, tumor-specific CD8+ T cells exhaustion has been correlated with high CD86+ pDC counts, in chronic myeloid leukemia (Schutz et al., 2017). The pDC expression of CD86, the ligand for CD28 (co-stimulatory) and CTLA-4 (co-inhibitory) molecules expressed by T cells, has been found to predict risk of disease recurrence after treatment (Schutz et al., 2017). Although it remains to be elucidated whether CD86+ pDCs migrate to the tumor and induce tumor-specific CD8+ T cell exhaustion, such a mechanism might occur, via CD86- CTLA-4 interaction. Plasmacytoid DCs from the BM of MM patients express high levels of PD-L1, compared to normal BM pDCs (Ray et al., 2015; Ray et al., 2014). Co-culture of pDCs from MM patients with autologous CD4+ or CD8+ T cells, in the presence of anti-PD-1 blocking antibody, led to increased T cell proliferation, compared with controls in absence of the antibody, suggesting a mechanism of T cell suppression by pDCs via PD-1/PD-L1 interaction (Ray et al., 2015). The accumulation of pDCs in tumors and TdLNs from melanoma patients and melanoma- bearing humanized mice is associated with a poor clinical outcome (Aspord et al., 2013, 2014a). These cells were shown to promote Th2 cells, producing IL-5, IL-13 and TNF-α, in an OX40-L- dependent manner (Aspord et al., 2013, 2014a). Another study, in the context of different breast cancer types, showed tumor-derived supernatant-induced activation of pDCs isolated from peripheral blood mononuclear cells (PBMCs), due to GM-CSF contained in the supernatants, which induced the upregulation of CD80, CD86 and ICOS-L (Ghirelli et al., 2015). GM-CSF- activated pDCs primed naïve T cells, leading to the differentiation of Th2 cells with a regulatory phenotype, producing IL-4, IL-5, IL-10, IL-13 and TNF-α, but low levels of IFN-γ. In vivo, high levels of GM-CSF and pDC infiltration correlated with the most aggressive breast cancer type (Ghirelli et al., 2015). In addition, TA-pDCs from ovarian cancer patients induced the production of IL-10 by allogeneic CD4+ T cells in ex vivo cultures (Labidi-Galy et al., 2011). Several studies, in human cancer and tumor mouse models, have shown an accumulation and colocalization of pDCs and CD4+ Tregs, correlating with poor prognoses (Le Mercier et al., 2013; Sisirak et al., 2012). A study in human breast cancer, although it did not demonstrate a direct role for pDC antigen presentation, showed that the decreased production of IFN-I by TA-pDCs was implicated in the expansion of Tregs (Sisirak et al., 2012). Furthermore, Treg expansion in breast and ovarian cancers is dependent on the expression of the immunosuppressive molecule ICOS-L by pDCs (Conrad et al., 2012; Faget et al., 2012). In both cases, pDC-induced Treg expansion was associated with disease progression and poor outcome. Moreover, TA-pDCs from melanoma patients induce the differentiation of IL-10-producing Tregs, in an ICOS-L-dependent

manner (Aspord et al., 2013, 2014a). In addition, the presence of Tr1 cells (Foxp3- IL-10+ IL-13-) in hepatocellular carcinoma or liver metastases from colorectal cancer was linked with pDC infiltration (Pedroza-Gonzalez et al., 2015). Increased IL-10 secretion by Tr1 cells was dependent on the expression of ICOS-L by pDCs. Finally, a subset of pDCs in the TdLNs of melanoma-bearing mice has been found to express the immunosuppressive molecule IDO, which led to the inhibition of T cell responses in vitro (Munn et al., 2004). In vivo, adoptive transfer of cells from TdLN led to T cell anergy, which was abrogated by IDO inhibitor administration. These IDO+ pDCs were subsequently demonstrated to directly activate Tregs in an IDO-dependent manner (Sharma et al., 2007). IDO+ pDCs have also been observed in the peripheral blood of late-stage melanoma patients (Chevolet et al., 2015). Whether they infiltrate the tumor and induce Treg expansion remains to be determined.

Harnessing the potential of antigen-presentation by pDCs to enhance anti-tumor immune responses Despite TA-pDCs having immunosuppressive properties, the potential of antigen presentation by pDCs that are not immersed in the TME, such as pDC lines, pDCs from peripheral blood or pDCs in non-invaded distal LNs, can be harnessed in order to enhance anti-tumor responses (Aspord et al., 2012; Guery et al., 2014; Lombardi et al., 2015; Tel et al., 2013a; Tel et al., 2012b). Aspord and colleagues used GEN2.2, a pDC line that harbor the MHC-I allele HLA-A*0201 (Aspord et al., 2010). Co-cultures of irradiated GEN2.2 loaded with tumor-derived antigens with PBMCs or tumor-infiltrating lymphocytes (TILs) from HLA-A*0201-matched melanoma patients led to enhanced CD8+ T cell degranulation and IFN-γ production and increased cytotoxicity towards patient tumor cells (Aspord et al., 2010; Aspord et al., 2012; Charles et al., 2018). The strategy of injecting tumor-derived peptide-pulsed GEN2.2 in vivo is now tested in clinical trial for late-stage melanoma patients4. In addition, Tel et al. performed therapeutic vaccination using pDCs isolated from melanoma patient peripheral blood (Tel et al., 2013a). Autologous pDCs were activated ex vivo by tick-borne encephalitis virus vaccine and loaded with tumor-antigenic peptides, inducing IFN-I secretion and upregulation of MHC-I, MHC-II, CD80, CD83 and CD86 expression. These cells were subsequently injected intranodally into patients, leading to enhanced CD4+ and CD8+ T cell responses (Tel et al., 2013a). Furthermore, in tumor mouse model, antigen delivery to pDCs via BST2, along with TLR agonists, led to antigen-presentation by pDCs in vivo, inducing a strong anti-tumoral response and tumor growth inhibition (Loschko et al., 2011b). Finally, Liu et al.,

4 http://pdc-line-pharma.com/clinical-trials/

injected ex vivo TLR9L (CpG-A)-activated pDCs in B16 melanoma-bearing mice, directly into the tumor, which led to regression of injected and untreated tumors (Liu et al., 2008a).

In an attempt to study a potential MHC-II-mediated antigen-presenting functions of pDCs in anti-tumor immunity, our group previously performed vaccination at a distal site from the tumor, i.e. subcutaneous injection in the right flank, while the tumor was established in the left flank (Guery et al., 2014). In this study, we used the TLR9 agonist CpG-B, along with a MHC-II- restricted peptide. Indeed, as mentioned previously, there are distinct classes of CpG, with different outcomes on pDCs (Gilliet et al., 2008). CpG-A forms aggregates that bind to TLR9 in the early endosome, leading to the production of IFN-I and IFN-III via IRF7 activation (Gilliet et al., 2008) (Fig. 14). In contrast, CpG-B, which exists in a monomeric form, traffics to the late endosome/lysosome, inducing the secretion of pro-inflammatory cytokines, such as IL-6 and TNF-α, and to the upregulation of the expression of genes involved in antigen presentation, such as MHC-II and the co-stimulatory molecules CD40, CD80 and CD86 (Gilliet et al., 2008) (Fig. 14). Finally, the binding of CpG-C on TLR9 leads to the expression of genes activated by CpG-A and of CpG-B-activated genes (Fig. 14) (Gilliet et al., 2008; Vollmer and Krieg, 2009). Therefore, CpG-B appeared to be the most appropriate class of CpG to analyze pDC antigen-presenting functions.

Following vaccination with CpG-B and MHC-II-resticted peptide, pDCs in distal LNs acquired an immunogenic phenotype and primed Th17 cells (Fig. 15) (Guery et al., 2014). Indeed, the use of mice in which MHC-II expression was selectively abrogated in pDCs allowed our group to conclude that Th17 differentiation was dependent, in this context, on MHC-II-mediated antigen- presenting functions of pDCs (Guery et al., 2014). Th17 cells subsequently migrated into the tumor, which led to immune cell infiltration, including CTLs, promoting tumor cell killing and inhibition of tumor growth (Fig. 15) (Guery et al., 2014). Our study, therefore, highlighted an anti-tumor role of pDCs, when activated at a distal site from the tumor, through their MHC-II- mediated antigen-presenting functions.

Figure 14. Signaling pathways of type A and B CpG oligodeoxynucleotides in early and late endosomes in plasmacytoid dendritic cells. Upper panel. In pDCs, the binding of CpG-A, which is aggregated, to Toll-like receptor (TLR)- 9 takes place in the early endosomal compartment in which markers, including early endosomal antigen 1 (EEA1) and transferrin receptor (TfR), are expressed. The prolonged interaction between CpG-A and TLR9 leads to the activation of myeloid differentiation primary response gene 88 (MyD88) and subsequently to interferon-regulatory factor 7 (IRF7), promoting a strong secretion of type I interferon (IFN-I). Lower panel. On the contrary, CpG-B (monomeric) bound to TLR9 traffic rapidly through the early endosome and enter the acidified late endosome/lysosome, in which LysoTracker and lysosomal- associated membrane protein 1 (LAMP1) are expressed. This leads to the activation of genes different from the ones activated by CpG-A, such as nuclear factor-κ B (NF-κB), mitogen-activated protein kinases (MAPK) and IRF5. It induces a different outcome for pDCs, with the upregulation of co-stimulatory molecules including CD80, CD40 and CD86, the production of pro-inflammatory cytokines such as tumor necrosis factor (TNF) and interleukin (IL)-6, and chemokines. BTK, Bruton's tyrosine kinase; IRAK, interleukin-1-receptor-associated kinase; OPN, osteopontin; TAK1, transforming-growth-factor-β-activated kinase 1; TRAF, tumor-necrosis factor (TNF) receptor-associated factor. Adapted from Gilliet*, Cao* & Liu, Nat Rev Immunol, 2008 (Gilliet et al., 2008).

Figure 15. Priming of Th17 cells by activated antigen-presenting plasmacytoid dendritic cells leads to tumor growth control: working model. 1. pDCs in tumor and tumor-draining lymph nodes (TdLNs) are rendered tolerogenic (purple) by the tumor microenvironment. 2. Nonetheless, tumor-bearing mice vaccination with CpG-B together with a MHC-II-restricted tumor antigenic peptide at a distal site from the tumor leads to the priming of Th17 cells by activated antigen-presenting pDCs (red). 3. Th17 cells migrate towards the tumor and induce immune cell infiltration into the tumor, including conventional DCs (cDCs), Th1 cells, regulatory T cells (Tregs) and cytotoxic T lymphocytes (CTLs). 4. Increased tumor antigen-specific CTL infiltration promotes tumor cell killing and tumor growth inhibition. Adapted from Guéry and Hugues, Oncoimmunology, 2015 (Guery and Hugues, 2015a), using data from Guéry et al., Cancer Res, 2014 (Guery et al., 2014).

Using TLR ligands to reprogram tolerogenic TA-pDCs Whether tolerogenic TA-pDCs can be reprogrammed using TLR agonists, in order to enhance their immunogenicity, is still a matter of debate. Topical administration of imiquimod (TLR7L) on skin neoplasms led to the infiltration of IFN-I-producing pDCs in lesions, inducing an important tumor regression (Stary et al., 2007; Urosevic et al., 2005). In addition, administration of imiquimod in a melanomal model of humanized mice induced activation of pDCs, with

increased cytotoxic functions and IFN-I production, and subsequent inhibition of tumor growth and metastasis (Aspord et al., 2014c). In another model of melanoma, topical imiquimod administration induced the recruitment of pDCs with direct tumoricidal activity thanks to granzyme B and TRAIL, leading to tumor growth control (Drobits et al., 2012). Finally, using an orthotopic mouse model of breast cancer, TLR7L intratumoral delivery led to a reprogramming of tolerogenic pDCs and to inhibition of tumor growth, which was dependent on IFN-I signaling (Le Mercier et al., 2013).

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.

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).

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.

Published OnlineFirst March 27, 2018; DOI: 10.1158/0008-5472.CAN-17-2549

Cancer Tumor Biology and Immunology Research

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 refractory to CpG-B stimulation and whose depletion did not harness the power of the immune system to eradicate cancer alter the efficacy of the treatment. Instead, tumor growth cells. Therapeutic vaccines, aimed at inducing active immune inhibition subsequent to i.t. CpG-B injection depended on responses against an existing cancer, are highly dependent on the recruitment of neutrophils into the milieu, resulting in the the immunological microenvironment, where many immune activation of conventional dendritic cells, subsequent cell types display high levels of plasticity and, depending on increased antitumor T-cell priming in draining lymph nodes, the context, promote very different immunologic outcomes. and enhanced effector T-cell infiltration in the tumor micro- Among them, plasmacytoid dendritic cells (pDC), known to environment. These results reinforce the concept that i.t. be highly immunogenic upon inflammation, are maintained delivery of TLR9 agonists alters the tumor microenvironment in a tolerogenic state by the tumor microenvironment. Here, by improving the antitumor activity of both innate and adap- we report that intratumoral (i.t.) injection of established tive immune cells. solid tumors with CpG oligonucleotides-B (CpG-B) inhibits Significance: Intratumoral delivery of CpG-B disrupts the tumor growth. Interestingly, control of tumor growth was tolerogenic tumor microenvironment and inhibits tumor growth. independent of tumor-associated pDC, which remained Cancer Res; 78(12); 3280–92. 2018 AACR.

Introduction sence of TA-pDC has been correlated to a bad prognosis, with little evidence showing their potential as efficient antigen- Tumor immunity is the result of complex interactions between presenting cells (APC). different cells immersed in the tumor microenvironment. However, pDC can be activated at a site distal from the tumor Although substantially inhibited by the tumor, effector immune and used as immunogenic APCs. Indeed, we previously showed responses can take place in particular conditions or experimental that contralateral vaccination of mice with CpG oligonucleotides- settings. It is not surprising, therefore, that novel therapeutic B (CpG-B), a TLR9 agonist (15), together with a MHCII-restricted strategies are aimed at boosting antitumor immune cell responses, tumor antigen, leads to the activation of distal pDC that induce in particular tumor-specific T-cell priming in lymph nodes (LN). tumor-specific Th17 cell differentiation. Immune cells are conse- Plasmacytoid dendritic cells (pDC) are involved in both innate quently recruited to the tumor, and cytotoxic T cells (CTL) (1) and adaptive immunity (2–9), and can exhibit either a subsequently eliminate the tumor (16). tolerogenic or immunogenic phenotype, depending on the Aiming at reversing the tolerogenic phenotype of TA-pDC and immunologic context (10). Although the role of pDC in antitu- other immune cells, by vaccinating directly at the tumor site, mor immunity is still debated, several studies have shown that might be a promising approach. Indeed, several studies have been tumor-associated pDC (TA-pDC), i.e., in the tumor and tumor- carried out using intratumoral (i.t.) delivery of different adju- draining lymph nodes (TdLN), exhibit a tolerogenic phenotype. vants, including CpG (17). Studies have used different classes of Indeed, TA-pDC are characterized by low type I interferon (IFN-I) CpG administered intratumorally (i.t.), peritumorally, and intra- secretion and costimulatory molecule expression and promote cranially (glioma/glioblastoma; refs. 18–21). In addition, CpG regulatory T-cell (Treg) induction (11–14). Therefore, the pre- has been injected systemically, vehicle in or in conjunction with molecules targeting the tumor, such as microparticles (22) or Department of Pathology and Immunology, University of Geneva Medical liposomes (23). Most of these studies have observed control of School, Geneva, Switzerland. þ tumor growth, infiltration of CD8 T cells, and decreased Treg Note: Supplementary data for this article are available at Cancer Research numbers in the tumor (22, 24–26). However, the mechanisms Online (http://cancerres.aacrjournals.org/). accounting for these observations have not been fully understood. Corresponding Author: S. Hugues, Geneva Medical School, Rue Michel-Servet 1, More importantly, CpG administration was, for the most part, 1211 Geneva, Switzerland. Phone: 41-22-379-58; Fax: 41-22-379-57-46; E-mail: combined with other treatments, such as cryosurgery, immune [email protected] checkpoint blockade, adoptive transfer of tumor-specific T cells, doi: 10.1158/0008-5472.CAN-17-2549 or cytokine injection. Therefore, the intrinsic effects of i.t. CpG 2018 American Association for Cancer Research. itself have not been deciphered.

3280 Cancer Res; 78(12) June 15, 2018

Intratumoral CpG-B Enhances Immune Cell Functions

Here, we investigated whether i.t. administration of CpG-B In vitro neutrophil migration assay alone or combined with a MHCII-restricted tumor antigen Tumor-conditioned media (TCM) were prepared as followed: could affect tumor growth. We show that i.t. CpG-B injection solid tumors were injected i.t. with PBS or CpG-B (30 mg) 11 days in mice leads to significant tumor growth inhibition that relies after tumor implantation. Twenty-four hours later, tumors were on T-cell responses, including an increased CTL infiltration and excised, cut in small pieces (25 mg/mL), and incubated in com- decreased Treg numbers. The CpG-B effect was enhanced by the plete RPMI at 37 C and 5% CO2. Twenty-four hours later, tumor addition of a tumor peptide. In both cases, pDC were not pieces were removed by centrifugation, and the supernatant, i.e., reactivated and were not involved in the control of tumor TCM, was filtered (0.22 mm). In parallel, bone marrow cells were growth, demonstrating that tolerogenic TA-pDC cannot be harvested from na€ve mice, followed by magnetic isolation of reprogrammed toward an immunogenic phenotype. In con- untouched neutrophils (Miltenyi Biotech). Bone marrow– trast, this treatment led to robust TA-conventional DCs (cDC) derived neutrophils were seeded in complete RPMI in the upper activation. Moreover, the i.t. CpG-B delivery induced a massive compartment (80% confluence on polycarbonate membrane i.t. infiltration of activated neutrophils. These cells are the first with 8 mm pores) of transwell chambers (Corning). The lower responders to sites of acute tissue damage and infection. In compartment contained either complete RPMI with indicated settings of chronic inflammation, neutrophils persist in tissues, CpG-B concentrations or medium conditioned from EG7 tumors and this has been associated with cancer progression (27). (TCM). After 24 hours of incubation at 37 C and 5% CO2, cells However,theroleofneutrophilsinthetumormicroenviron- were counted in the lower compartment. ment remains controversial, with evidence for both pro- and antitumor roles (28). Our results show that i.t. neutrophil pDC and neutrophil depletion recruitment significantly contributes to TA-cDC activation and For the depletion of pDC in vivo, diphtheria toxin (DT; Sigma; antitumor T-cell responses following i.t. CpG-B administration, 1 mg/mL in 100 mL of PBS) was injected i.p. in BDCA2-DTR mice, resulting in the control of tumor growth. Taken together, we 1 day prior to i.t. CpG-B OVAII administration and every 3 suggest that intratumoral delivery of vaccines that include TLR9 to 4 days over the indicated period. Plasmacytoid DC depletion agonists would benefit current immunotherapy anticancer was analyzed in the tumor on the day of i.t. CpG-B OVAII strategies by disrupting the tolerogenic tumor microenviron- administration by flow cytometry. ment and enhancing innate and adaptive tumor-associated Monoclonal anti-Ly6G (1A8) antibody and Rat IgG2a iso- immune cell responses. type control (2A3) were purchased from BioXcell and were injected i.v. (1 mg/mL in 100 mL of PBS) and i.t. (2.5 mg/mL in 40 mL of PBS), 1 day prior to i.t. CpG-B injection. Neutrophil Materials and Methods depletion was analyzed by flow cytometry in the blood and Mice tumors 2 days after depleting antibody injection. Mice were of pure C57BL/6 background and were bred and maintained under specific pathogen-free conditions at Geneva Preparation of single-cell suspensions Medical School animal facility or under specific and opportu- Tumors and TdLNs were excised from mice at indicated times. nistic pathogen-free conditions at Charles River, France or Italy. Cell isolation from tumors was performed as follows: tumors Rag2 / (29), BDCA2-DTR (30), and IFNAR / (31) mice have were digested chemically with collagenase D (1 mg/mL; Roche) been described elsewhere. All animal husbandry and experi- and DNAse I (10 mg/mL; Roche) and mechanically. Dead cells ments were approved by and performed in accordance with the were eliminated using lympholyte M (Cedarlane Laboratory). For guidelines of the animal research committee of Geneva canton, DC and neutrophil analyses and flow-cytometry cell sorting, Switzerland. TdLNs were subjected to the same treatment. For intracellular cytokine analyses, cells were restimulated for 4 hours at 37C and Tumor experiments 5% CO2, in complete RPMI medium in the presence of phorbol Ovalbumin-expressing tumor cell lines and EG7-OVA thymo- 12-myristate 13-acetate (Sigma; 100 ng/mL) and ionomycin ma cells were obtained from S. Amigorena laboratory in 2011 (Sigma; 1 mg/mL). GolgiPlug solution (BD Biosciences; 1 mL/mL) and used at below passage 20. Cells were not authenticated and was added to the culture medium for the last 2.5 hours. were tested negative for Mycoplasma. EG7-OVA cells were cultured at 37 C and 5% CO2 in complete RPMI medium [RPMI 1640 Blockade of T-cell egress from the LNs GlutaMAX with 10% heat-inactivated FBS, 1% penicillin/strep- Mice were injected i.p. with the sphingosine-1-phosphate tomycin (10,000 U/mL penicillin and 10 mg streptomycin/mL), receptor-1 antagonist FTY720 (Sigma; 100 mg/mL in 200 mLof 1% sodium pyruvate 100 mmol/L, 0.1% 2-b-mercaptoethanol PBS) 1 day prior to i.t. CpG-B OVAII administration, and every 50 mmol/L] supplemented with G418 (Mediatech; 0.4 mg/mL) day within the time period indicated. for the selection of OVA-expressing cells. G418 was removed from the medium 1 day prior to tumor cell implantation. Tumor Flow cytometry cells (1 106 in PBS) were injected s.c. into the left flank. Anti-CD62L (MEL-14), anti-CD69 (H1.2F3), anti-FasL OVAII peptide (ISQAVHAAHAEINEAGR) was purchased (MFL3), anti–ICAM-1 (YN1/1.7.4), anti-TCRb (H57-597), anti- from Polypeptide and CpG-B 1668 from Invivogen. Mice were CD4 (RM4-5), anti-CD8a (53-6.7), anti-FOXP3 (FJK-16s), anti- injected i.t. once tumors were established, at indicated times ICOS (15F9), anti-CD25 (PC61.5), anti-CD103 (2E7), anti-IL17 (D7 to D12 after tumor implantation), with CpG-B (30 mg), in (eBio17B7), anti-IFNg (XMG1.2), anti-CD11b (M1/70), anti– the presence or not of OVAII (10 mg), in PBS (40 mL). Tumor Siglec-H (eBio927), anti–PDCA-1 (eBio440c), anti-B220 (RA3- size was measured with a caliper [L (length) l(width)]every 6B2), anti-Ly6C (HK1.4), and anti-F4/80 (BM8) monoclonal 1 to 2 days over the indicated periods of time. antibodies were purchased from Thermo Fisher Scientific, and

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

Humbert et al.

anti-CD45 (30F11), anti-CD86 (PO3), anti–I-Ab (AF6-120.1), Unique gene model construction and gene coverage reporting. A anti-CD11c (N418), anti-CD40 (3/23), and anti-Ly6G (1A8) unique gene model was used to quantify reads per gene. Briefly, monoclonal antibodies and DRAQ7 were purchased from the model considers all annotated exons of all annotated protein Biolegend. coding isoforms of a gene to create a unique gene where the Cells were stained with Fixable viability dye eFluor 780 genomic region of all exons are considered coming from the same (Thermo Fisher Scientific) for 30 minutes, at 4C, for cell analyses, RNA molecule and merged together. or with DRAQ7 immediately before cell sorting. Pentamer and intracellular staining were performed on sepa- RNA-seq analysis. All reads overlapping the exons of each unique rate single-cell suspensions due to technical incompatibilities. gene model were reported using featureCounts version 1.4.6- p1 H-2kb SIINFEKL pentamer (OVAI; Proimmune) staining was (http://bioinf.wehi.edu.au/featureCounts/). Gene expressions performed in accordance with the manufacturer protocol. were reported as raw counts and in parallel normalized in reads Single-cell suspensions were incubated with anti-CD16/32 Fcg per kilobase million (RPKM) in order to filter out genes with low RII-RIII, a Fc receptor monoclonal blocking antibody (Thermo expression value (1 RPKM) before calling for differentially Fisher Scientific), for 10 minutes, at 4C, before staining with expressed genes. Library size normalizations and differential gene antibodies. expression calculations were performed using the package edgeR Intracellular stainings (cytokines and Foxp3) were performed (http://bioconductor.org/packages/release/bioc/html/edgeR.html) using the Intracellular Fixation and Permeabilization buffer set designed for the R software (http://www.R-project.org/). Only (Thermo Fisher Scientific). genes having a significant fold change (Benjamini–Hochberg cor- Data were acquired using Gallios (Beckman Coulter) and rected P value < 0.05) were considered for the rest of the RNA-seq analyzed using FlowJo Software (FlowJo, LLC). Flow cytometry analysis. cell sorting of tumor-associated cDC, pDC, and neutrophils was performed using MoFlo Astrios (Beckman Coulter). Doublets Gene ontology and/or KEGG analysis. Gene ontology (GO) term were excluded by gating on the cells along the diagonal in a FSC and Kyoto Encyclopedia of Genes and Genomes (KEGG) meta- height/FSC area dot plot. Dead cells were excluded by gating on bolic pathways enrichment were performed using homemade the live cells that have an intermediate fluorescence intensity scripts for the R software. for the viability dye eFluor 780 (cell analyses) or by excluding þ the DRAQ7 cells (cell sorting). cDC were defined as CD45hi GSEA: pathway enrichment. Gene sets were selected according to þ þ CD11chi F4/80 and pDC as CD45hi CD11cint B220 Siglec-H their known involvement in either "Antigen presentation," þ þ þ PDCA-1 . Neutrophils were defined as CD45 CD11b Ly6Cint "Immunogenicity," or "DC migration" (see Supplementary Table Ly6Ghi. In the experiments in which neutrophils were depleted S1). Genes were ranked by their calculated fold changes (decreas- using the anti-Ly6G antibody (1A8; BioXcell), neutrophils were ing ranking). A gene set analysis using the gene set enrichment þ þ defined as CD45 CD11b Ly6Cint as the use of the anti-Ly6G analysis (GSEA) package Version 2.2 (http://software.broadinsti (1A8, same clone used for in vivo depletion) antibody for flow tute.org/gsea/index.jsp; ref. 32) from the Broad Institute (MIT, cytometry would not have allowed a valid analysis of the Cambridge, MA) was used to analyze the pattern of differential depletion efficiency. gene expression between the two groups. Gene set permutations were performed 1,000 times for each analysis. The normalized RNA sequencing enrichment score (NES) was calculated for each gene set. Library preparation, sequencing, and read mapping to the reference genome. Flow cytometry–isolated TA-pDC and TA-cDC (see RT-qPCR above) were collected in RNAprotect Cell Reagent (Qiagen). RNA Flow cytometry–sorted tumor-associated neutrophils (TAN; was isolated using an RNeasy Plus Micro Kit (Qiagen), and three see above) were collected in RNAprotect Cell Reagent (Qiagen). to four replicates per condition were used. RNA integrity and Total RNA was isolated using the RNeasy Plus Micro Kit quantity were assessed with a Bioanalyzer (Agilent Technologies). (Qiagen), and RT-qPCR was performed as described (4). cDNA cDNA libraries were constructed by the Genomic platform of the was synthesized with random hexamers and M-MLV Reverse University of Geneva as follows: the SMARTer Ultra Low RNA Kit transcriptase (Promega). PCR was performed with CFX Connect from Clontech was used for the reverse transcription and cDNA Real-time System (Bio-Rad) and iQ SYBR green Super-mix (Bio- amplification according to the manufacturer's specifications, Rad). HPRT mRNA was used for normalization. Primer sequen- 0 starting with 1 ng of total RNA. cDNA (200 pg) was used for ces used were as follows: HPRT, forward, 5 -GAGGAGTCCTGTT- 0 0 library preparation using the Nextera XT Kit from Illumina. GATGTTGCCAG-3 and reverse, 5 -GGCTGGCCTATAGGCTCA- 0 0 Library molarity and quality were assessed with the Qubit and TAGTGC-3 ;CCL3,forward,5-TACAGCCGGAAGATTCCACG 0 0 0 Tapestation using a DNA High sensitivity chip (Agilent Technol- -3 and reverse, 5 -GTCTCTTTGGAGTCAGCGCA-3 ;CCL4,for- 0 0 0 ogies). Pools of 10 libraries were loaded for clustering on a single- ward, 5 -TTTCTCTTACACCTCCCGGC-3 and reverse, 5 -GAG- 0 0 read Illumina Flow cell. Reads of 50 bases were generated using GAGGCCTCTCCTGAAGT-3 ; CCL5, forward, 5 -GTGCCCACG- 0 0 the TruSeq SBS chemistry on an Illumina HiSeq 4000 sequencer. TCAAGGAGTAT-3 and reverse, 5 -TTCTCTGGGTTGGCACA- 0 FastQ reads were mapped to the ENSEMBL reference genome CAC-3 . (GRCm38.89) using STAR version 2.4.0j (https://github.com/ alexdobin/STAR) with standard settings, except that any reads Statistical analysis mapping to more than one location in the genome (ambiguous Statistical significance was assessed by one-way ANOVA test reads) were discarded (m ¼ 1). Sequences data have been sub- with Bonferroni post hoc test or by two-tailed Mann–Whitney mitted to the GEO database under the accession number test. Two-way ANOVA test with Bonferroni post hoc test was per- GSE112206. formed for the follow-up of tumor growth. All analyses were

3282 Cancer Res; 78(12) June 15, 2018 Cancer Research

Intratumoral CpG-B Enhances Immune Cell Functions

carried out using Prism 7.0 (GraphPad software; , P < 0.05; treated compared with PBS control tumors (Fig. 1G), showing , P < 0.01; , P < 0.001; , P < 0.0001; ns: nonsignificant). that CpG-BOVAII delivery significantly enhances anti–tumor- specific T-cell responses. In order to determine whether i.t. Results CpG-B OVAII treatments could elicit a systemic antitumor immune response, EG7 tumor cells were implanted in mice Intratumoral administration of CpG-BOVA leads to tumor II in both flanks.I.t.injectionofCpG-BorCpG-BþOVA in the growth inhibition mediated by tumor-specific T-cell responses II left flank led to a modest tumor growth inhibition in tumors Whether the tolerogenic phenotype of TA-pDC, and specifically implanted in the right flank, compared with the PBS control their MHCII-restricted antigen-presenting functions, can be group (Supplementary Fig. S3E), suggesting that a systemic reversed toward an immunogenic state remains unknown. To antitumor response could be induced, although not sufficient address this question, we used i.t. injection of the TLR9 ligand to control the growth of distal tumors as efficiently as injected CpG-B, known to be a potent activator of pDC, notably of their tumors. antigen-presenting functions (15), either alone or in combina- Altogether, these results show that i.t. injection of established tion with the MHCII-restricted model tumor antigenic peptide EG7 tumors with CpG-BOVA induces a potent specific and OVA . I.t. injection of CpG-B into established OVA-expressing II II antitumor T-cell response that leads to tumor growth control. We EG7 thymoma tumors led to a significant delayed tumor growth, further confirmed these results using another mouse model of compared with PBS-injected control tumors (Fig. 1A). Injection tumors expressing a different tumor antigen, MCA-101-Dby of OVA in addition to CpG-B (CpG-BþOVA vaccine) re- II II tumor cells (Supplementary Fig. S4A and S4D). sulted in a potentialization of this effect (Fig. 1A). EG7 tumor- bearing Rag2 / mice also received i.t. injection of CpG-B or þ CpG-B OVAII. In the absence of T and B cells, CpG-B OVAII Intratumoral administration of CpG-BOVAII does not i.t. administration did not affect tumor growth (Fig. 1B), suggest- reprogram pDC functions but induces cDC activation that ing that the effect was mediated by adaptive immune responses. precedes T-cell activation in the TdLNs / In addition, the absence of effect on tumor growth in Rag2 I.t. injection of CpG-B or CpG-BþOVAII ledtoapotent mice showed that CpG-B i.t. injection did not have any direct antitumor immune response (Figs. 1 and 2). We wondered þ þ þ cytotoxic activity on tumor cells (Fig. 1B). Accordingly, in vitro whether pDC, defined as CD11cint B220 PDCA1 Siglec-H treatment of EG7 cells with different concentrations of CpG-B cells (Supplementary Fig. S5), and known to be involved in did not affect cell growth (Supplementary Fig. S1). Similar tumoral peptide þ CpG-B vaccination at a site distal to the results on tumor growth were obtained following i.t. injection tumor (16), were playing a role in this response. To this of established OVA-expressing B16 (B16-OVA) melanoma purpose, we used BDCA2-DTR mice, in which pDC express tumors (Supplementary Fig. S2). the DT receptor, leading to pDC depletion upon DT injection As the control of tumor growth induced by i.t. administra- (Supplementary Fig. S5). Mice were administered i.t. with CpG- tion of CpG-B OVAII relied on adaptive immunity (Fig. 1B), BorCpG-BþOVAII 1 day after the first injection of DT. Both we analyzed the early T-cell response at day 5. Strikingly, tumor treatments led to a significant reduction in tumor sizes even in sizes and weights were already significantly lower in mice that the absence of pDC, suggesting these cells were not involved received CpG-B alone or CpG-BþOVAII as compared with the in the mechanism of action of the inhibition of tumor growth PBS control group, showing a rapid impact on tumor growth (Fig. 2). Moreover, the molecules important for antigen- (Fig. 1C; Supplementary Fig. S3A and S3B). Five days after i.t. presenting functions, such as MHCII and the costimulatory injection of CpG-B or CpG-BþOVAII,weobservedasignificant molecules CD40 and CD86, were not upregulated on TA-pDC þ þ þ fi decrease of Treg (CD4 Foxp3 CD25 ) numbers in ltrating 24 hours following i.t. CpG-BOVAII administration, neither the tumor compared with the PBS controlgroup(Fig.1D).The in the tumor nor in the TdLNs (Fig. 3A). As pDC are the main þ þ numbers of CD103 ICOS Tregs, which exhibit a suppres- producers of IFN-I and can indirectly activate other immune sive phenotype (33), were also strongly decreased (Fig. 1D). We cell types, such as cDC, we tested this hypothesis by adminis- fi þ / then analyzed the cytokinic pro le of CD4 effector T cells. The trating i.t. CpG-BþOVAII in IFNAR mice lacking the IFN-I þ frequency of IFNg-producing CD4 tumor-infiltrating lympho- receptor. In those mice, injection led to tumor growth control cytes (TIL) was massively increased (4-fold increase) 5 days (Supplementary Fig. S6), suggesting that IFN-I, and therefore þ after i.t. administration of CpG-B or CpG-B OVAII,bycontrast innate pDC functions, was not involved in i.t. CpG-BþOVAII with the IL17 production, for which there was no difference efficacy. Taken together, these results show that neither innate þ (Fig. 1E). IFNg and IL17 production by CD4 T cells in the nor adaptive pDC functions are mediating the antitumoral fi TdLNs was not signi cantly affected by the treatments (Sup- T-cell response induced by i.t. delivery of CpG-BOVAII. þ þ plementary Fig. S3C). The frequency of IFNg CD8 TILs was We next analyzed the cDC compartment, defined as CD11chi þ þ increased after i.t. injection of CpG-BOVAII, compared with cells. CD8a and CD103 cDC are specialized in the cross- þ þ the control group, together with a slight increase in the TdLNs presentation of antigens to CD8 T cells, whereas CD11b cDC þ (Fig. 1F; Supplementary Fig. S3D). Finally, the frequencies of are mainly involved in antigen presentation to CD4 T cells (34). þ OVAI-specificCD8 T cells were increased in the tumor and The most abundant cDC subtype present into the tumor was the þ þ þ TdLNs (Fig. 1F; Supplementary Fig. S3D). When normalized to CD11b , whereas both CD103 CD8a and CD11b were more tumor weights, Treg numbers became similar for untreated and frequent in the TdLNs, without or 24 hours after i.t. injection þ þ treated tumors (Fig. 1D), whereas effector CD4 and CD8 of CpG-BOVAII (Fig. 3B). I.t. CpG-B or CpG-BþOVAII did not T-cell numbers slightly increased (Fig. 1E and F), probably due lead to increased cDC frequencies or numbers in the tumor or þ þ to tumor weight heterogeneity. However, effector T-cell number/ TdLNs. As the CD103 CD8a cDC were very rare, we did not þ Treg number ratios were greatly increased in CpG-BOVAII– further characterize this population and focus on CD103 CD8a

www.aacrjournals.org Cancer Res; 78(12) June 15, 2018 3283

Humbert et al.

-/- A 300 WT B 300 Rag2 ) PBS ) PBS 2 2 m

CpG-B m CpG-B ns

(m CpG-B+OVA

(m CpG-B+OVA 200 II 200 II *** i.t. **** i.t. or size 100 m 100 ** Tumor size Tu

0 0 0 5 10 15 20 0 5 10 15 20 Time (days) Time (days) C D ** ) ** **

300 3

) D0 30 ns 8 3 2 * ** ** m )

D5 × 10 3

6 Tregs (m + 2 200 ns 20 4

10 ICOS 1 + Tregs (%) Tregs 2 100 (abs. nb x 10 Tumor size

0 Tregs (abs. nb 0 0 ns CD103 20 + 10 0 ns 8 PBS CpG-B CpG-B+OVAII 15 ICOS

+ 6 10 nb Tregs/ mg tumor mg 4 PBS CpG-B CpG-B+OVAII E 5 2 Tregs/mg tumor Tregs/mg ) nb CD103

3 * 150 30 ** 10 * 0 0 )

T ns ns 3 T/ T/ T (%) + 8 ns 60 + + 100 20 10 × 6 CD4 40 G CD4 CD4 + γ + + 4 5 *** γ γ T/

T (abs. nb x10 50 10 + + **

IFN 20

2 tumor mg (abs. nb (abs. nb 4 IFN CD4 CD4 0 0 0 nb IFN 0 3 ns + ns γ 100 4 1.5 8 2 T/ nb Tregs ) ns T + 3 ns + 80 T (%) 1

3 6 nb IFN + 1.0 × 10 60 CD4 0 CD4 + T/mg tumor T/mg 4 2 + CD4 + 17

40 + mg tumor mg 17 0.5 ns 17 20 1 IL 2 (abs. nb (abs. nb IL nb IL 250 T/

nb CD4 * 0 0 0 0 + 200 ) 3 F ns CD8 150 ns + γ

500 80 400 ns T/ 4,000 × 10 * + ) 100 T

* 3 nb Tregs

400 + T (%) 60 300 3,000

+ 50 CD8 nb IFN × 10

300 + CD8 40 γ 2,000 0 + 200 CD8 γ T (abs. nb nb T (abs. 200 + mg tumor mg + γ 20 1,000 100 IFN 100 nb IFN (abs. nb (abs. nb IFN CD8 T/ 80 * 0 0 + 0 0 * ns ns ns 6,000 25 T 150 60 60 CD8 T/ T (%) + )

P + = 0.08 + 3

+ * 20 ns 40

4,000 CD8 100 40 CD8 × 10 CD8

15 + nb Tregs + +

T/mg tumor 20

+ 10 2,000 50 20 mg tumor mg 0 5 nb Pentamer (abs. nb (abs. nb Pentamer Pentamer Pentamer nb CD8 0 0 0 0

3284 Cancer Res; 78(12) June 15, 2018 Cancer Research

Intratumoral CpG-B Enhances Immune Cell Functions

300 PBS T-cell egress from the LNs was blocked by the mean of sphingo-

) ns sine-1-phosphate receptor (S1PR) inhibition (35). Upon block-

2 CpG-B 200 ade of T-cell egress from the LNs using FTY720, the size of the CpG-B+OVAII

i.t. ** tumor 5 days after i.t. injection of CpG-B OVAII was not dimin- DT PBS **** ished in the groups that received CpG-BOVAII, compared with 100 DT CpG-B the two PBS-control group injected or not with S1PR inhibitor, Tumor size (mm DT CpG-B+OVAII suggesting that tumor growth control was not mediated by local 0 T-cell activation and was rather prevented when activated T cells 0 5 10 15 20 25 could not exit from TdLNs to inﬁltrate the tumor (not shown). In Time (days) agreement, TIL numbers in the groups that received S1PR inhib- Figure 2. itor were dramatically decreased compared with an i.t. PBS-

Control of tumor growth induced by i.t. administration of CpG-BOVAII injected control group, which did not receive the S1PR inhibitor does not rely on pDC. EG7 cells were implanted s.c. in BDCA2-DTR mice. At (Fig. 5A and B), demonstrating that activated tumor-specificT day 10 after tumor cell implantation, mice received an i.t. injection of cells were indeed sequestered in the LNs. Moreover, the frequen- PBS, CpG-B alone, or CpG-B along with the OVA peptide, CpG-BþOVA . þ þ þ II II cies of IFNg CD4 T cells and OVAI-specific CD8 T cells were One day prior to i.t. injection, indicated mice were injected i.p. with DT increased in LNs after i.t. CpG-BOVAII compared with the PBS and then every 3 to 4 days. Tumor growth was measured every 1 to 2 days. fi Two-way ANOVA test with Bonferroni post hoc test was performed. control group and were signi cantly increased compared with the Results show the mean SEM derived from 6 to 8 mice and are control group that did not receive S1PR inhibitor (Fig. 5A and B). þ þ representative of at least two independent experiments. , P < 0.05; This was concomitant to decreased IFNg CD4 T cells and OVAI- þ , P < 0.01; , P < 0.001; , P < 0.0001; and ns, nonsignificant. specific CD8 T-cell frequencies and numbers in the tumor (Fig. 5A and B). These results show that following i.t. CpG-BOVAII administration, activated APCs, likely cDC, migrate from the þ þ þ (referred to as CD8a ) cDC, also involved in the cross-presenta- tumor to the TdLNs and activate tumor-specific CD4 and CD8 þ tion of antigens to CD8 T cells (Fig. 3B). Twenty-four hours after T cells in the TdLNs. Whether cDC actually present antigens and injection with CpG-BOVAII, MHCII, CD86, and CD40 expres- induce effector T-cell differentiation remains to be determined. þ þ sion levels were upregulated by CD8a and CD11b cDC in the þ TdLNs and only by CD8a DCs in the tumor (Fig. 3A). Hence, i.t. Intratumoral CpG-B delivery increases the infiltration of administration of CpG-B or CpG-BþOVAII leads to the activation neutrophils that control tumor growth of TA-cDC. Our results were further confirmed by transcriptomic In order to determine whether other cell types were affected by fi experiments performed on cDC and pDC in ltrating the tumors. i.t. delivery of CpG-BOVAII,wefinally analyzed the myeloid Twenty-four hours following i.t. CpG-B, cDC preferentially upre- innate immune compartment. Whereas the frequencies of eosi- gulate genes implicated in antitumor T-cell immunity (antigen nophils, monocytes, and macrophages were unchanged (not þ presentation, immunogenicity, and DC migration; Fig. 4A and B; shown), we observed a 3-fold increase in neutrophil (CD11b Supplementary Table S1). In contrast, the distribution of gene Ly6Ghi Ly6Cint) frequency in the tumor 48 hours after i.t. expression is balanced for TA-pDC, without any preferential CpG-BOVAII (Fig. 6A). Also knowing that neutrophils express enrichment toward proimmunogenic genes (Fig. 4A and B). In TLR9 (36, 37), we further characterized their phenotype after i.t. particular, GSEA (32) highlighted that genes related to antigen injection. The expression of ICAM-1 was upregulated, whereas presentation are upregulated in TA-cDCs upon CpG-B treatment, that of CD62L was downregulated by neutrophils in untreated whereas they are downmodulated in TA-pDC. In addition, the (PBS) tumors compared with neutrophils in the TdLNs and enrichment of genes related to immunogenicity and DC migra- nondraining LNs (Supplementary Fig. S7A), suggesting that tion was more pronounced in TA-cDC compared with TA-pDC. neutrophils were already activated when infiltrating the tumor. Therefore, although i.t. CpG-B treatment does not convert Furthermore, we did not observe any significant changes in TA-pDC phenotype, it induces a reprogramming of TA-cDC neutrophil phenotype 48 hours after i.t. injection of CpG-B or toward an immunogenic phenotype. CpG-BþOVAII compared with control groups (Supplementary We hypothesized that the activation of tumor-specific T cells Fig. S7B). In addition, ex vivo exposure of TANs to CpG-B does not after i.t. injection (Fig. 1E and F) was initiated after antigen uptake affect their ability to produce ROS (Supplementary Fig. S7C). by cDC in the tumor and subsequent migration to the TdLNs Therefore, i.t. administration of CpG-B or CpG-BþOVAII induces for presentation to T cells. To determine whether the induction an increased recruitment of activated neutrophils into the tumor, of T-cell response was indeed induced in TdLNs or rather due without locally affecting their phenotype and function. In agree- þ þ to a local activation and/or reactivation of CD4 and CD8 TILs, ment, in vitro transwell experiments indicated that neutrophils

Figure 1.

Intratumoral delivery of CpG-BOVAII induces tumor growth inhibition and enhances antitumor T-cell responses. EG7 cells were implanted s.c. in WT (A and C–F)andRag2/ (B)mice.Seven(A), 10 (B), or 8–12 (C–G) days after tumor cell implantation, mice received an i.t. injection of either PBS, CpG-B alone, or CpG-B along with the OVAII peptide, CpG-BþOVAII. A and B, Tumor growth was measured every 1 to 2 days. Results show the mean SEM derived from 6 mice and are representative of at least two independent experiments. Two-way ANOVA test with Bonferroni post hoc test was performed. C, Tumor size was measured on the day of i.t. injection and 5 days later. Results are derived from at least 20 mice. D–G, Frequencies and/or absolute þ þ þ þ þ þ þ þ þ þ numbers of Tregs (Foxp3 CD25 CD4 Tcells)andCD103 ICOS Tregs (D); CD4 T cells, IFNg , and IL-17 cells among CD4 T cells (E); and CD8 T cells, þ þ þ IFNg ,andOVAI-specific(Pentamer )cellsamongCD8 T cells (F) were measured in cells isolated from the tumor by flow cytometry 5 days after i.t. injection. Ratios (indicated cell number/mg of tumor) are shown. G, Ratios (indicated cell number/Treg number) are shown. Results show the mean SEM (D–G) derived from 6 to 12 mice and are representative of at least two independent experiments. One-way ANOVA tests with Bonferroni post hoc tests were performed. , P < 0.05; , P < 0.01; , P < 0.001; , P < 0.0001; and ns, nonsignificant.

www.aacrjournals.org Cancer Res; 78(12) June 15, 2018 3285

Humbert et al.

A TdLN Tumor ns ** 10 8 *** 8 6

) **

6 )

3 ***

3 4 350 4 225 * 300 ** 200

× 10 2 2 × 10 175 250 ** 150 * * 0 200 0 125 ns 150 100 75 100 50 ** 50 ns 25 *** MHCII (MFI (MFI MHCII 0 (MFI MHCII 0 CD8α+ CD11b+ pDCs CD8α+ CD11b+ pDCs CD11chi CD11cint CD11chi CD11cint

** ns Figure 3. 3 ** 3 Intratumoral delivery of CpG-

BOVAII does not reprogram **** 2 2 TA-pDC tolerogenic phenotype but ) ) 3 25 3 25 ** promotes cDC activation. EG7 cells ** 1 20 *** 1 20 ** were implanted s.c. in WT mice. At × 10

*** × 10 ** ** days 9 to 10 after tumor cell 15 0 15 0 implantation, mice received an i.t. 10 10 injection of PBS, CpG-B alone, or ** CpG-B along with the OVA peptide, 5 ** 5 ns II CpG-BþOVAII. MHCII, CD86, and hi þ CD86 (MFI (MFI CD86 0 (MFI CD86 0 CD40 expression by CD11c CD8a , + + α+ + CD8α CD11b pDCs CD8 CD11b pDCs CD11chi CD11bþ, and pDC (A) and hi hi int CD11c CD11cint CD11c CD11c frequencies among CD45þ cells of þ þ CD11chi CD103 CD8a , CD11chi þ þ ns CD103 CD8a , CD11chi CD11b , and 50 30 ns pDC (CD11cint Siglec-Hþ PDCA1þ 40 B220þ, independent graphs are also 30 20 shown; B) were measured in the cells **** isolated from the tumor and TdLNs ) 20 ) ns 3 125 ** 3 60 10 by flow cytometry 24 hours after i.t. * 10 ns 100 50 injection. One-way ANOVA tests with × 10 × 10 0 post hoc 0 40 Bonferroni tests were 75 *** ns performed. Results show the ns 30 50 ** mean SEM derived from 4 mice 20 and are representative of at least 25 10 two independent experiments. CD40 (MFI (MFI CD40 0 (MFI CD40 0 , P < 0.05; , P < 0.01; , P < 0.001; + + α+ + CD8α CD11b pDCs CD8 CD11b pDCs , P < 0.0001; and ns, hi CD11chi CD11cint CD11c CD11cint nonsignificant. MFI, median fluorescence intensity.

PBS CpG-B CpG-B+OVAII B TdLN Tumor

0.3 2.0 * * ** 1.5 0.2 1.0 0.1 0.5 **

Dendritic cells0.0 (%) 0.0 CD103+ CD103- CD11b+ pDCs Dendritic cells (%) CD103+ CD103- CD11b+ pDCs CD8α+ CD8α+ CD8α+ CD8α+ CD11chi CD11cint CD11chi CD11cint

isolated from the bone marrow migrate efﬁciently toward TCM, To determine whether neutrophils were playing a role with a slight increase for CpG-B–injected tumors compared with in the mechanism of action of i.t. CpG-BOVAII in inhib- untreated tumors (Fig. 6B). In contrast, CpG-B alone did not iting tumor growth, we performed an antibody-mediated cell induce neutrophil migration (Fig. 6B). depletion (Supplementary Fig. S8A and S8B). As neutrophil

3286 Cancer Res; 78(12) June 15, 2018 Cancer Research

Intratumoral CpG-B Enhances Immune Cell Functions

Antigen A Immunogenicity DC migration presentation

2 TA-cDC TA-pDC 1

-1

-2 Normalized enrichment score 33 Genes 81 Genes 22 Genes B Antigen presentation Immunogenicity DC migration TA-cDC

0.5 0.6 0.7 0.5 0.3 0.4 0.3 E.S. 0.1 0.2 0.1 0 -0.1 -0.1

7.5 ‘na_pos’ (posively correlated) 7.5 ‘na_pos’ (posively correlated) 7.5 ‘na_pos’ (posively correlated) 2.5 2.5 2.5 -2.5 -2.5 -2.5 R.L.M. -7.5 ‘na_neg’ (negavely correlated) -7.5 ‘na_neg’ (negavely correlated) -7.5 ‘na_neg’ (negavely correlated) 0 2,000 4,000 6,000 8,000 10,000 0 2,000 4,000 6,000 8,000 10,000 0 2,000 4,000 6,000 8,000 10,000 TA-pDC

0.1 0.4 0.4 0.2 -0.1 0.2 0.0

E.S. 0.0 -0.3 -0.2 -0.2 -0.5

10 ‘na_pos’ (posively correlated) 10 ‘na_pos’ (posively correlated) 10 ‘na_pos’ (posively correlated)

0 0 0

R.L.M. ‘na_neg’ (negavely correlated) ‘na_neg’ (negavely correlated) ‘na_neg’ (negavely correlated) -10 -10 -10 0 2,500 5,000 7,500 10,000 0 2,500 5,000 7,500 10,000 0 2,500 5,000 7,500 10,000

Ranked in ordered dataset Ranked in ordered dataset Ranked in ordered dataset

Enrichment profile Hits Ranking metric scores

Figure 4. Enrichment scores and profiles of genes involved in Ag presentation, immunogenicity, and DC migration in TA-cDC and pDC following i.t. administration of CpG-BOVAII. EG7 cells were implanted s.c. in WT mice. At day 10 after tumor cell implantation, mice received an i.t. injection of PBS or CpG-B. One day after, TA-cDC (CD11chi F480)andTA-pDC(CD11cint Siglec-Hþ PDCA-1þ)werepurified by flow cytometry, and unbiased gene expression analysis was performed by RNA-seq (3–4 replicates per condition). GSEA was performed using the fold changes (in log2)calculatedbetweenCpG-B– and PBS-injected tumors for both purified TA-cDC and TA-pDC. Gene sets were selected according to their established involvement in different DC functions: antigen presentation, immunogenicity, or DC migration (see also Supplementary Table S1). A, NES for the gene categories "Antigen presentation," "Immunogenicity," and "DC migration." Number of genes analyzed is mentioned for each category. B, GSEA results for each gene category. ES, enrichment score; RLM, ranked listed metric.

www.aacrjournals.org Cancer Res; 78(12) June 15, 2018 3287

Humbert et al.

CpG-B administration and to analyze tumor weight and size A Tumor P B Tumor = 0.081 * 5 days later, similarly to what was done before (Fig. 1C;

) 300 Supplementary Fig. S3A and S3B). Neutrophil depletion ) 150 * 3

3 * (50% in tumors, 70%–80% in blood 1 day after i.t. injection; 10 × 10 * cells Supplementary Fig. S8A and S8B; 45% in tumors at day 5, not ×

T 200

T cells 100 ﬁ ﬁ + shown) led to a signi cant reduction in the ef cacy of i.t. CpG- + B delivery compared with the isotype control group, as the

CD8 100 (abs. nb nb (abs. CD4 50 decrease in tumor size and weight observed 5 days after injec- (abs. nb (abs. nb tion was not signiﬁcant upon neutrophil depletion (Fig. 7A; 0 0 Supplementary S8C). P = 0.053 Then, we wondered whether neutrophil depletion was affecting the induction of the tumor-speciﬁc T-cell response P P = 0.079 50 = 0.072 50 observed 5 days after i.t. CpG-B injection (Fig. 1E and F). þ ) )

3 ﬁ T cells 3 40 40 Increases in OVAI-speci cCD8 T-cell numbers in TdLNs and + T cells 10 frequencies in tumors following CpG-B administration were × 10 + × 30 30 ﬁ

CD8 signi cantly abrogated upon neutrophil depletion (Fig. 7B),

+ ﬁ CD4 20 20 suggesting that neutrophil in ltration in the tumor promotes

+ þ γ 10 antitumor CD8 T-cell responses after CpG-B delivery. (abs. nb nb (abs. (abs. nb nb (abs. 10 Furthermore, neutrophil depletion also resulted in a signif- IFN 0 0

Pentamer icantdecreaseofTA-cDCexpressionofCD80,CD86,and ns 40 25 ns MHCII promoted by CpG-B delivery, demonstrating that

+ cDC activation relies on neutrophil recruitment in tumors 20

+ 30 (Fig. 7C and D). Interestingly, qPCR experiments show that,

CD8 15

+ compared with untreated tumors, TANs isolated from tu-

CD4 20 ﬁ + 10 mors administered with i.t. CpG-B produced signi cantly γ more CCL3, CCL4, and CCL5 chemokines than that have T cells (%)

T cells (%) 10 5 IFN been involved in DC recruitment and/or activation (Fig. 7E; 0 Pentamer 0 refs. 39–41). TdLN TdLN Collectively, these results show that neutrophils contri- ** * bute to control of tumor growth, TA-cDC activation, and in- 4 creased tumor-speciﬁc T-cell responses induced by i.t. CpG-B 5 *** + administration.

+ 4 3 CD8 %) ( +

CD4 3 2 + γ

cells 2 T cells (%)

T 1 IFN A B 1 *** Pentamer 40 ** 2.0 ns 0 0 * ** 30 1.5 PBS PBS CpG-B CpG-B+OVAII ns Ctl FTY720 20 1.0

10 0.5 Fold change in cell count Figure 5. Neutrophils (%) Tumor-specific T cells are activated in the TdLNs after i.t. administration of 0 0.0 NT 0.5 μmol/L 2 μmol/L 8 μmol/L PBS CpG-B CpG-BOVA . EG7 cells were implanted s.c. in WT mice. At day 12 II PBS CpG-B CpG-B+OVAII CpG-B TCM after tumor cell implantation, mice received an i.t. injection of PBS, CpG-B alone, or CpG-B along with the OVA peptide, CpG-BþOVA .Oneday II II Figure 6. priortoi.t.injection,micewereinjectedi.p.withFTY720andthen fi every day for 5 days, except a control PBS group. Frequencies and/or Intratumoral injection of CpG-B induces i.t. in ltration of neutrophils. A, þ þ þ EG7 cells were implanted s.c. in WT mice. At day 11 after tumor cell absolute numbers of CD4 T cells and IFNg cells among CD4 T cells (A), CD8þ T cells, and OVA -specific(Pentamerþ)cellsamongCD8þ Tcells(B) implantation, mice received an i.t. injection of PBS, CpG-B alone, or CpG-B I þ þ hi were measured in cells isolated from the tumor and/or TdLNs by flow along with the OVAII peptide, CpG-B OVAII. Neutrophil (CD11b Ly6G int þ cytometry 5 days after i.t. injection. One-way ANOVA tests with Bonferroni Ly6C )frequencyamongCD45 cells was measured in the cells fl post hoc tests were performed. Results show the mean SEM derived from isolated from the tumor by ow cytometry 48 hours after i.t. injection. – 5 to 6 mice and are representative of at least two independent Results show the mean SEM derived from 7 mice. B, Bone marrow derived neutrophils were seeded in the upper compartment of transwell experiments. , P < 0.05; , P < 0.01; , P < 0.001; , P < 0.0001; and ns, nonsignificant. chambers. The lower compartment contained either indicated CpG-B concentrations or medium conditioned from EG7 tumors (TCM). After 24 hours, cells were counted in the lower compartment. Results show the mean SEM derived from three independent TCM. One-way ANOVA depletion leads to an increased neutrophil production in the tests with Bonferroni post hoc tests were performed. Results are bone marrow after a few days (38), it is technically challenging representative of at least two independent experiments. , P < 0.05; to deplete them on the long term. Thus, we decided to use a , P < 0.01; , P < 0.001; , P < 0.0001; ns, nonsignificant; and NT, one-shot injection of depleting antibodies 1 day prior to i.t. nontreated.

3288 Cancer Res; 78(12) June 15, 2018 Cancer Research

Intratumoral CpG-B Enhances Immune Cell Functions

A B 1.5 25 ns * +

ns ) 3 150 T cells 20 + CD8 ) 1.0 2 + 15 TdLN CD8 ns 100 + cells (%) 10

T 0.5 (abs. nb x10

Pentamer 5 50 0.0 Pentamer 0 Tumor size (mm 40 80 ns 0

+ * PBSCpG-B PBS CpG-B cells 30 T 60 + CD8 Isotype Anti-Ly6G + ns Tumor 20 CD8 40 + /mg tumor D0 D5 T cells (%) 10 20 Pentamer

0 Pentamer 0 D C CD8α+ cDCs CD11b+ cDCs ns ** ns 150 80 ** 150 80

) **** ) ** ** P 3 = 0.053 **** 3 ns 60 60 100 100 (%) ** *** (%) hi 40 hi 40 50 50 CD40 20 CD40 20 CD40 (MFI x10 0 0 CD40 (MFI x10 0 0 ns PBS 15 * 60 ** 15 ns 60 * ns ) ) 3 **** **** 3

** CpG-B Isotype ns ns

10 40 10 %) 40 ( (%) ** hi hi PBS 5 20 5 20 CD86 CD86 CpG-B CD86 (MFI x10 CD86 (MFI CD86 x10 0 0 0 0 ns Anti-Ly6G P = 0.089 60 ** 50 **** 80 50 ns )

) ns

3 ** **** 3 ns 40 60 40 40 (%) 30 ns (%) 30 hi hi 40 20 20 20 20 MHCII MHCII 10 10 MHCII (MFI x10 MHCII (MFI x10 0 0 0 0 E * * * 2.0 3 5 1.5 4 PBS 2 3 1.0 CpG-B 1 2 0.5

expression 1 expression expression Relative mRNA Relative mRNA Relative 0.0 0 Relative mRNA 0 CCL3 CCL4 CCL5

Figure 7. Enhanced i.t. neutrophil recruitment in CpG-B–treated tumors promotes TA-cDC activation and antitumor T-cell responses. A–D, EG7 cells were implanted s.c. in WT mice. At day 11 after tumor cell implantation, mice were injected i.v. with depleting anti-Ly6G antibodies or isotype control. One day later, mice received an i.t. injection of PBS or CpG-B. A, Tumor sizes were measured on the day of i.t. injection and 5 days later. B, Frequencies and/or absolute þ þ numbers of OVAI-specific(Pentamer )CD8 T cells were measured in cells isolated from the tumor and TdLNs by flow cytometry 5 days after i.t. injection. Ratios (indicated cell number/mg of tumor) are indicated. C and D, Expression levels (left histograms) and frequencies of high expresser cells (right þ þ histograms) of CD40, CD86, and MHCII among CD8a (C) and CD11b (D) DC (CD11chi F4/80 ) were measured in cells isolated from the TdLNs by flow cytometry 24 hours after i.t. injection. One-way ANOVA tests with Bonferroni post hoc tests were performed. Results show the mean SEM derived from 5 mice (A–D) and are representative of two independent experiments. E, EG7 cells were implanted s.c. in WT mice. At day 11 after tumor cell implantation, mice received an i.t. injection of PBS or CpG-B. After 24 hours, TA neutrophils were sorted by flow cytometry. Indicated chemokine mRNA levels were measured by qPCR. Results show the mean SEM derived from four pools of three individual mice and are representative of two independent experiments. Mann–Whitney tests were performed. , P < 0.05; , P < 0.01; , P < 0.001; , P < 0.0001; and ns, nonsignificant. MFI, median fluorescence intensity.

www.aacrjournals.org Cancer Res; 78(12) June 15, 2018 3289

Humbert et al.

Discussion performed by Nierkens and colleagues, a different CpG route of administration was used (peritumoral), and the combination Here, we investigated how the i.t. delivery of the TLR9 agonist with tumor cryosurgery might explain contradictory findings CpG-B could modulate the tumor microenvironment and repro- (20). Similarly, Lou and colleagues observed that antibody- gram immune cells to boost antitumor immunity. mediated depletion of pDC induced a decrease in chemokine A significant inhibition of OVA-expressing EG7 thymoma production and led to a partial decrease in T-cell accumulation tumor growth was observed after i.t. administration of CpG-B, in the tumor, accompanied by a modest loss of the protection confirming previous studies in mice and humans (19, 21, 26). conferred by CpG injection (49). Unfortunately, the CpG class However, in other reports, CpG-B was most frequently used in used in this model was not specified (49). conjunction with other treatments (18–20), thus making it dif- Having ruled out a possible role for pDC in the protection ficult to appreciate its inherent properties. The addition of the afforded by the i.t. CpG-BOVAII delivery, we analyzed the tumoral peptide (OVA ) to CpG-B enhanced the vaccine effect, þ þ II phenotype and frequency of cDC subsets. CD103 and CD8a most probably by promoting MHCII-restricted antigen presenta- þ cDC are key for antigen cross-presentation to CD8 T cells, tion by increasing tumor antigen availability. This observation is þ whereas CD11b cDC are known to be involved in antigen of importance because it suggests that when the tumor antigen is þ þ presentation to CD4 T cells (34). The frequency of CD103 available for presentation and its immunogenic potential is þ CD8a cDC infiltrating the tumors was extremely low, so we elevated, the delivery of TLR9 ligands may suffice to transform could not further characterize this cell population. Salmon and APCs from a tolerogenic to an immunogenic state, so enhancing þ colleagues showed, nonetheless, that CD103 cDC represent a antitumor T-cell responses. In agreement, similar observations rare DC subset in the tumor but are of major importance for were made using the MCA101-D as a tumor model, for which the þ transport of antigen to the TdLNs and cross-presentation to CD8 addition of an immunogenic tumor peptide to i.t. CpG-B delivery T cells (50). The frequency of cDC was not increased in tumor and resulted in a more pronounced potentization of antitumor T-cell TdLNs after i.t. CpG-BOVAII and was even decreased for immunity. In accordance, adaptive immunity was required for the þ CD11b DC, in accordance with previous studies, possibly due control of tumor growth following i.t. injection of both CpG-B to cell death or migration. Interestingly, and in agreement with and CpG-BþOVA . Importantly, Treg numbers were decreased in þ II previous studies in which CD11c DC were activated after cryo- the tumor after i.t. CpG-BOVA as suggested previously (25, 26), þ þ II surgery and CpG-B injection (20, 51), both CD11b and CD8a among which Tregs harbor a suppressive phenotype. These results cDC subsets upregulated MHCII, CD86, and CD40 molecules in are in agreement with studies using different immunodeficient TdLNs 24 hours after CpG-BOVA injection, suggesting that in mouse strains or depleting antibodies (22, 23, 25, 42–44). II contrast to TA-pDC, TA-cDC can be reactivated following local Highlighting the importance of the immunologic context in CpG-B administration. Accordingly, transcriptomic analyses effector T-cell polarization, we observed that whereas CpG- show that i.t. CpG-B reprograms the phenotype of TA-cDC, with BOVA delivered at a distal site to the tumor (16) induced II a preferential upregulation of several genes implicated in antigen potent Th17 responses, the i.t. injection promoted robust Th1 presentation, immunogenicity, and migration. Whether tumor responses. This was associated with increased IFNg production þ antigen presentation by these different cDC subtypes to T cells is and OVA -specificCD8 T cells in the tumor, confirming I actually enhanced in TdLNs after treatment remains to be deter- previous reports of similar findings (22, 24, 26). Delivery of mined. Control of tumor growth was rapid, 5 days after i.t. CpG- CpG-BOVA i.t. also slightly reduced the growth of a second II BOVA administration, suggesting that the T cells might be contralateral tumor. This limited effect on distal tumors is in II activated in the tumor itself (52). However, upon T-cell egress agreement with the study by Sagiv-Barfi I and colleagues, which blockade from the LN at the time of i.t. injection, the inhibition of shows that whereas i.t. CpG-C alone does not have a major þ þ tumor growth was abrogated. Moreover, IFNg CD4 T cells and impact on distant tumors, its combination with anti-OX40 þ OVA -specific CD8 T cells were sequestered in the TdLNs, where- agonistic antibodies efficiently cures established distant tumors I þ as the frequency and number of these cells were decreased in the (45). Moreover, increased tumor-specificCD8 T-cell responses tumor, compared with a control group in which T-cell migration were observed in the blood of tumor-bearing mice cotreated to the tumor was not blocked. This shows that i.t. CpG-BOVA with CpG-B and tumor-antigen transfected adenovirus (25), as II administration leads to tumor-specific T-cell activation in the well as in tumor patients treated with CpG-B combined to low TdLNs and subsequent migration to the tumor. These findings dose radiotherapy (18). also reinforce a possible role for activated cDC in the priming of Whether the tolerogenic phenotype of pDC immersed in the antitumor T cells following activation with CpG-BOVA . The tumor microenvironment can be reversed is still an important II mechanisms accounting for the activation of cDC also remain to matter for debate (11–14). We did not observe any tolerogenic- be determined. cDCs could be either directly activated via TLR9 to-immunogenic conversion of TA-pDC phenotype after i.t. ligation (53) or indirectly activated with other cell types mediat- CpG-BOVA . These cells were indeed refractory to activation, II ing the signaling between CpG-B and cDC activation. In contrast as they neither upregulated the expression of MHCII and costi- to previous observations (20), we excluded a possible role for mulatory molecules, nor they preferentially upregulated proim- IFN-I in the activation of cDC after i.t. CpG-BOVA delivery. munogenic genes, after treatment. These results, along with the II In an attempt to identify other cell types affected by i.t. absence of any impact of i.t. CpG-BOVAII on tumor growth in CpG-BOVA administration, no difference in the frequency of pDC-depleted (DT-treated BDCA2-DTR) and IFNAR / mice II innate immune cells in the tumor (not shown) was observed, compared with WT mice, may clarify previous conflicting obser- except for neutrophils that significantly increased after 48 hours. vations (46–48). Therefore, we propose that pDC are not Conditioned media from tumors actively attract bone marrow– involved in the antitumoral effect, which is in discrepancy with derived neutrophils, with a slight increase for CpG-B–treated previously published observations (20). However, in the study tumors, suggesting that, upon CpG-B delivery, some changes

3290 Cancer Res; 78(12) June 15, 2018 Cancer Research

Intratumoral CpG-B Enhances Immune Cell Functions

occurinthetumormicroenvironment, enhancing the produc- vention, to disrupt the tolerogenic tumor microenvironment. In tion of neutrophil attractants by either tumor cells or tumor- agreement, a recent study in mice has shown that the combination infiltrating cells. The expression of adhesion molecules by of CpG-C and an agonistic anti-OX40 antibody can trigger anti- tumor-infiltrating neutrophils was higher compared with LN tumor T-cell immunity and can cure multiple types of cancer or neutrophils. The modulation of adhesion molecule expression prevent spontaneous genetically driven cancers (45). Therefore, might allow the neutrophils to migrate from the periphery while i.t. TLR ligand administration might appear difficult to to the tumor. However, their phenotype was not altered by establish in the clinic, although successfully achieved on several the i.t. delivery of CpG-BOVAII,suggestingthati.t.CpG-B occasions (18), it remains a promising approach, especially with does not further activate TANs but rather promote their recruit- the development of new technologies allowing the local delivery ment. The role of neutrophils in antitumor immunity is con- of the adjuvant into the tumor. flicting, with some studies showing either pro- or antitumoral activity (27, 28). Hence, we determined whether neutrophil Disclosure of Potential Conflicts of Interest depletion could prevent tumor growth inhibition after i.t. CpG- No potential conflicts of interest were disclosed. B injection. Neutrophil depletion leads to an increased production of neutrophils in the bone marrow (38), rendering a Authors' Contributions proper depletion very challenging in the long term. Therefore, Conception and design: M. Humbert, S. Hugues we used one-shot injection of depleting antibody, which results Development of methodology: M. Humbert in an efficient neutrophil elimination for at least 5 days. Acquisition of data (provided animals, acquired and managed patients, Neutrophil depletion led to a significant loss of CpG-B anti- provided facilities, etc.): M. Humbert, L. Guery, D. Brighouse tumoral effect, demonstrating a major contribution of those Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): M. Humbert, L. Guery, S. Lemeille, S. Hugues cells in the inhibition of tumor growth. Importantly, it also led Writing, review, and/or revision of the manuscript: M. Humbert, S. Hugues fi þ to a decreased OVAI-speci cCD8 antitumor T-cell response, Study supervision: S. Hugues firmly establishing a role for enhanced i.t. neutrophil recruitment in the protective effect of CpG-B delivery. Furthermore, Acknowledgments neutrophil depletion resulted in impaired TA-cDC activation, The authors thank J.P. Aubry-Lachainaye and C. Gameiro for excel- and TANs isolated from CpG-B–injected tumors produced lent assistance in flow cytometry; C. Lippens, J. Dubrot, A. Schlaeppi, increased amounts of chemokines involved in cDC recruitment C. Kowalski, and A.O. Gkountidi for help in experiments; A. Marti-Lindez / and/or activation (54). Altogether, our results suggest that and W. Reith for providing Rag2 mice;L.Garnier,G.Harle, and neutrophils and cDC cooperate to promote antitumor T-cell W. Ferlin for critical reading of the article; and N. Bendriss-Vermare, F.Tacchini-Cottier,Y.Eimre,andA.Sidibe for helpful discussions. We immunity after i.t. CpG-BOVAII injection. Whether neutro- þ thank the iGE3 Genomics Platform of theUniversityofGeneva(https://ige3. phils could, in addition, attract CD8 T cells in the tumor as genomics.unige.ch) for RNA-seq experiments. This work was supported showninamodelofinfluenza virus infection (55) and/or by the Swiss National Science Foundation (PP00P3_152951 and þ directly cross-present antigens to CD8 Tcells(56)remainsto 310030_166541) and the European Research Council (281365). be determined. The cellular pattern of TLR expression may indeed differ The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked between mice and humans. However, further dissecting the advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate mechanisms of protection afforded by the i.t. CpG-B administra- this fact. tion will certainly improve future anticancer therapeutic strategies. This will include the combination of adjuvants with different Received August 23, 2017; revised February 13, 2018; accepted March 22, treatments, such as checkpoint inhibitors or mechanical inter- 2018; published first March 27, 2018.

References 1. Colonna M, Trinchieri G, Liu YJ. Plasmacytoid dendritic cells in immunity. induces protective T cell-mediated immunity. J Immunol 2011; Nat Immunol 2004;5:1219–26. 186:6718–25. 2. de Heer HJ, Hammad H, Soullie T, Hijdra D, Vos N, Willart MA, et al. 8. Ochando JC, Homma C, Yang Y, Hidalgo A, Garin A, Tacke F, et al. Essential role of lung plasmacytoid dendritic cells in preventing Alloantigen-presenting plasmacytoid dendritic cells mediate tolerance to asthmatic reactions to harmless inhaled antigen. J Exp Med 2004;200: vascularized grafts. Nat Immunol 2006;7:652–62. 89–98. 9. Villadangos JA, Young L. Antigen-presentation properties of plasmacytoid 3. Hadeiba H, Sato T, Habtezion A, Oderup C, Pan J, Butcher EC. CCR9 dendritic cells. Immunity 2008;29:352–61. expression deﬁnes tolerogenic plasmacytoid dendritic cells able to suppress 10. Guery L, Hugues S. Tolerogenic and activatory plasmacytoid dendritic cells acute graft-versus-host disease. Nat Immunol 2008;9:1253–60. in autoimmunity. Front Immunol 2013;4:59. 4. Irla M, Kupfer N, Suter T, Lissilaa R, Benkhoucha M, Skupsky J, et al. 11. Conrad C, Gregorio J, Wang YH, Ito T, Meller S, Hanabuchi S, et al. MHC class II-restricted antigen presentation by plasmacytoid dendritic Plasmacytoid dendritic cells promote immunosuppression in ovarian cells inhibits T cell-mediated autoimmunity. J Exp Med 2010;207: cancer via ICOS costimulation of Foxp3(þ) T-regulatory cells. Cancer Res 1891–905. 2012;72:5240–9. 5. Isaksson M, Ardesjo B, Ronnblom L, Kampe O, Lassmann H, Eloranta ML, 12. Faget J, Bendriss-Vermare N, Gobert M, Durand I, Olive D, Biota C, et al. et al. Plasmacytoid DC promote priming of autoimmune Th17 cells and ICOS-ligand expression on plasmacytoid dendritic cells supports breast EAE. Eur J Immunol 2009;39:2925–35. cancer progression by promoting the accumulation of immunosuppressive 6. Loschko J, Heink S, Hackl D, Dudziak D, Reindl W, Korn T, et al. Antigen CD4þ T cells. Cancer Res 2012;72:6130–41. targeting to plasmacytoid dendritic cells via Siglec-H inhibits Th cell- 13. Sharma MD, Baban B, Chandler P, Hou DY, Singh N, Yagita H, et al. dependent autoimmunity. J Immunol 2011;187:6346–56. Plasmacytoid dendritic cells from mouse tumor-draining lymph nodes 7. Loschko J, Schlitzer A, Dudziak D, Drexler I, Sandholzer N, Bourquin directly activate mature Tregs via indoleamine 2,3-dioxygenase. J Clin C, et al. Antigen delivery to plasmacytoid dendritic cells via BST2 Invest 2007;117:2570–82.

www.aacrjournals.org Cancer Res; 78(12) June 15, 2018 3291

Humbert et al.

14. Sisirak V, Faget J, Gobert M, Goutagny N, Vey N, Treilleux I, et al. Impaired tional cell surface Toll-like receptor 9. Eur J Immunol 2013;43: IFN-alpha production by plasmacytoid dendritic cells favors regulatory 2101–13. T-cell expansion that may contribute to breast cancer progression. Cancer 37. Schneberger D, Caldwell S, Kanthan R, Singh B. Expression of Toll-like Res 2012;72:5188–97. receptor 9 in mouse and human lungs. J Anat 2013;222:495–503. 15. Gilliet M, Cao W, Liu YJ. Plasmacytoid dendritic cells: sensing nucleic acids 38. Milanova V, Ivanovska N, Dimitrova P. TLR2 elicits IL-17-mediated RANKL in viral infection and autoimmune diseases. Nat Rev Immunol 2008; expression, IL-17, and OPG production in neutrophils from arthritic mice. 8:594–606. Mediators Inflamm 2014;2014:643406. 16. Guery L, Dubrot J, Lippens C, Brighouse D, Malinge P, Irla M, et al. Ag- 39. Bennouna S, Bliss SK, Curiel TJ, Denkers EY. Cross-talk in the innate presenting CpG-activated pDCs prime Th17 cells that induce tumor immune system: neutrophils instruct recruitment and activation of den- regression. Cancer Res 2014;74:6430–40. dritic cells during microbial infection. J Immunol 2003;171:6052–8. 17. Shirota H, Tross D, Klinman DM. CpG oligonucleotides as cancer vaccine 40. Charmoy M, Brunner-Agten S, Aebischer D, Auderset F, Launois P, Milon G, adjuvants. Vaccines 2015;3:390–407. et al. Neutrophil-derived CCL3 is essential for the rapid recruitment of 18. Brody JD, Ai WZ, Czerwinski DK, Torchia JA, Levy M, Advani RH, et al. In dendritic cells to the site of Leishmania major inoculation in resistant mice. situ vaccination with a TLR9 agonist induces systemic lymphoma regres- PLoS Pathog 2010;6:e1000755. sion: a phase I/II study. J Clin Oncol 2010;28:4324–32. 41. Trifilo MJ, Lane TE. The CC chemokine ligand 3 regulates 19. Carpentier A, Metellus P, Ursu R, Zohar S, Lafitte F, Barrie M, et al. CD11cþCD11bþCD8alpha- dendritic cell maturation and activation fol- Intracerebral administration of CpG oligonucleotide for patients with lowing viral infection of the central nervous system: implications for a role recurrent glioblastoma: a phase II study. Neuro Oncol 2010;12:401–8. in T cell activation. Virology 2004;327:8–15. 20. Nierkens S, den Brok MH, Garcia Z, Togher S, Wagenaars J, Wassink M, et al. 42. Houot R, Levy R. T-cell modulation combined with intratumoral CpG Immune adjuvant efficacy of CpG oligonucleotide in cancer treatment is cures lymphoma in a mouse model without the need for chemotherapy. founded specifically upon TLR9 function in plasmacytoid dendritic cells. Blood 2009;113:3546–52. Cancer Res 2011;71:6428–37. 43. Marabelle A, Kohrt H, Sagiv-Barfi I, Ajami B, Axtell RC, Zhou G, et al. 21. Sharma S, Karakousis CP, Takita H, Shin K, Brooks SP. Cytokines and Depleting tumor-specific Tregs at a single site eradicates disseminated chemokines are expressed at different levels in small and large murine tumors. J Clin Invest 2013;123:2447–63. colon-26 tumors following intratumoral injections of CpG ODN. 44. Sharma S, Dominguez AL, Manrique SZ, Cavallo F, Sakaguchi S, Neoplasia 2004;6:523–8. Lustgarten J. Systemic targeting of CpG-ODN to the tumor microenvi- 22. Sato T, Shimosato T, Ueda A, Ishigatsubo Y, Klinman DM. Intrapulmonary ronment with anti-neu-CpG hybrid molecule and T regulatory cell delivery of CpG microparticles eliminates lung tumors. Mol Cancer Ther depletion induces memory responses in BALB-neuT tolerant mice. 2015;14:2198–205. Cancer Res 2008;68:7530–40. 23. Brignole C, Marimpietri D, Di Paolo D, Perri P, Morandi F, Pastorino F, 45. Sagiv-Barfi I, Czerwinski DK, Levy S, Alam IS, Mayer AT, Gambhir SS, et al. et al. Therapeutic targeting of TLR9 inhibits cell growth and induces Eradication of spontaneous malignancy by local immunotherapy. Sci apoptosis in neuroblastoma. Cancer Res 2010;70:9816–26. Transl Med 2018;10. 24. Carpentier AF, Xie J, Mokhtari K, Delattre JY. Successful treatment of 46. Le Mercier I, Poujol D, Sanlaville A, Sisirak V, Gobert M, Durand I, et al. intracranial gliomas in rat by oligodeoxynucleotides containing CpG Tumor promotion by intratumoral plasmacytoid dendritic cells is reversed motifs. Clin Cancer Res 2000;6:2469–73. by TLR7 ligand treatment. Cancer Res 2013;73:4629–40. 25. Geary SM, Lemke CD, Lubaroff DM, Salem AK. Tumor immunotherapy 47. Tel J, Aarntzen EH, Baba T, Schreibelt G, Schulte BM, Benitez-Ribas D, et al. using adenovirus vaccines in combination with intratumoral doses of CpG Natural human plasmacytoid dendritic cells induce antigen-specific T-cell ODN. Cancer Immunol Immunother 2011;60:1309–17. responses in melanoma patients. Cancer Res 2013;73:1063–75. 26. Grauer OM, Molling JW, Bennink E, Toonen LW, Sutmuller RP, Nierkens S, 48. Tel J, Sittig SP, Blom RA, Cruz LJ, Schreibelt G, Figdor CG, et al. Targeting et al. TLR ligands in the local treatment of established intracerebral murine uptake receptors on human plasmacytoid dendritic cells triggers antigen gliomas. J Immunol 2008;181:6720–9. cross-presentation and robust type I IFN secretion. J Immunol 2013;191: 27. Nicolas-Avila JA, Adrover JM, Hidalgo A. Neutrophils in homeostasis, 5005–12. immunity, and cancer. Immunity 2017;46:15–28. 49. Lou Y, Liu C, Lizee G, Peng W, Xu C, Ye Y, et al. Antitumor activity mediated 28. Fridlender ZG, Sun J, Kim S, Kapoor V, Cheng G, Ling L, et al. Polarization by CpG: the route of administration is critical. J Immunother 2011;34: of tumor-associated neutrophil phenotype by TGF-beta: "N1" versus "N2" 279–88. TAN. Cancer Cell 2009;16:183–94. 50. Salmon H, Idoyaga J, Rahman A, Leboeuf M, Remark R, Jordan S, et al. 29. Shinkai Y, Rathbun G, Lam KP, Oltz EM, Stewart V, Mendelsohn M, et al. Expansion and activation of CD103(þ) dendritic cell progenitors at the RAG-2-deficient mice lack mature lymphocytes owing to inability to tumor site enhances tumor responses to therapeutic PD-L1 and BRAF initiate V(D)J rearrangement. Cell 1992;68:855–67. inhibition. Immunity 2016;44:924–38. 30. Swiecki M, Gilfillan S, Vermi W, Wang Y, Colonna M. Plasmacytoid 51. den Brok MH, Sutmuller RP, Nierkens S, Bennink EJ, Toonen LW, Figdor dendritic cell ablation impacts early interferon responses and antiviral NK CG, et al. Synergy between in situ cryoablation and TLR9 stimulation and CD8(þ) T cell accrual. Immunity 2010;33:955–66. results in a highly effective in vivo dendritic cell vaccine. Cancer Res 31. Muller U, Steinhoff U, Reis LF, Hemmi S, Pavlovic J, Zinkernagel RM, et al. 2006;66:7285–92. Functional role of type I and type II interferons in antiviral defense. Science 52. Peske JD, Thompson ED, Gemta L, Baylis RA, Fu YX, Engelhard VH. Effector 1994;264:1918–21. lymphocyte-induced lymph node-like vasculature enables naive T-cell 32. Subramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, Gillette MA, entry into tumours and enhanced anti-tumour immunity. Nat Commun et al. Gene set enrichment analysis: a knowledge-based approach for 2015;6:7114. interpreting genome-wide expression profiles. Proc Natl Acad Sci U S A 53. Puttur F, Francozo M, Solmaz G, Bueno C, Lindenberg M, Gohmert M, 2005;102:15545–50. et al. Conventional dendritic cells confer protection against mouse 33. Barthlott T, Bosch AJ, Berkemeier C, Nogales-Cadenas R, Jeker LT, Keller cytomegalovirus infection via TLR9 and MyD88 signaling. Cell Rep MP, et al. A subpopulation of CD103(pos) ICOS(pos) Treg cells occurs 2016;17:1113–27. at high frequency in lymphopenic mice and represents a lymph node 54. Schuster S, Hurrell B, Tacchini-Cottier F. Crosstalk between neutrophils specific differentiation stage. Eur J Immunol 2015;45:1760–71. and dendritic cells: a context-dependent process. J Leukoc Biol 2013;94: 34. Dalod M, Chelbi R, Malissen B, Lawrence T. Dendritic cell maturation: 671–5. functional specialization through signaling specificity and transcriptional 55. Lim K, Hyun YM, Lambert-Emo K, Capece T, Bae S, Miller R, et al. programming. EMBO J 2014;33:1104–16. Neutrophil trails guide influenza-specific CD8(þ) T cells in the airways. 35. Cyster JG, Schwab SR. Sphingosine-1-phosphate and lymphocyte egress Science 2015;349:aaa4352. from lymphoid organs. Annu Rev Immunol 2012;30:69–94. 56. Beauvillain C, Delneste Y, Scotet M, Peres A, Gascan H, Guermonprez P, 36. Lindau D, Mussard J, Wagner BJ, Ribon M, Ronnefarth VM, et al. Neutrophils efficiently cross-prime naive T cells in vivo. Blood Quettier M, et al. Primary blood neutrophils express a func- 2007;110:2965–73.

3292 Cancer Res; 78(12) June 15, 2018 Cancer Research

Supplementary material and methods

Tumor experiments

B16-OVA and Dby-expressing MCA-101 (MCA-101-Dby), methylcholanthrene-induced fibrosarcoma expressing the MHC-II-restricted male Dby peptide (NAGFNSNRANSSRSS) 5, were cultured and implanted as described in the main material and methods. Dby HY peptide was purchased from Polypeptide. Mice were injected i.t. once tumors were established, at D10 post-tumor implantation, with CpG-B (30 µg), in the presence or not of HY peptide (10 µg), in PBS (40 µl).

CpG-B toxicity in vitro assay

EG7-OVA cells were seeded at an initial density of 5 x 104 cells/ml in complete RPMI medium and cultured for three days, at 37°C and 5% CO2, in the presence of CpG-B at the indicated concentrations.

Cells were counted every 24 hours.

Effect of i.t. CpG-B treatment on the growth of a distal tumor

Tumor cells were implanted (as described in the main material and methods) in both flanks. Mice were injected i.t. in the left tumors (D12 post-tumor implantation) with CpG-B (30 µg), in the presence or not of OVAII (10 µg) in PBS (40 µl). Tumor size was measured in the right flank (distal tumor), with a caliper [L (length) x l (width)] every one to two days over the indicated periods of time.

Reactive oxygen species (ROS) detection

Flow cytometry cell sorted tumor-associated neutrophils (see main material and methods) were seeded in complete RPMI, treated with indicated CpG-B concentrations, and cultured for 24h at 37°C and 5%

CO2. ROS production was assessed using the CellROX deep red flow cytometry Assay kit

(Thermofisher), following manufacturer’s instruction. Anti-oxydant (N-acetyl cystein, NAC), and oxydant (tert-burtyl hydroperoxide, TBHP) were used as negative and positive controls, respectively.

5 Joncker NT, Marloie MA, Chernysheva A, Lonchay C, Cuff S, Klijanienko J, et al. Antigen-independent accumulation of activated effector/memory T lymphocytes into human and murine tumors. International journal of cancer. 2006 Mar 1;118(5):1205-14. PubMed PMID: 16152614.

III.3. Discussion

Plasmacytoid DCs are extremely plastic cells and their phenotype highly depends on the immunological context (tissue, stimulus, inflammatory conditions) (Alculumbre et al., 2018a; Guery and Hugues, 2013). The MHC-II-mediated antigen-presenting functions of pDCs can therefore be either tolerogenic or immunogenic (Guery et al., 2014). In the context of tumor immunity, pDCs infiltrate tumors in many types of human cancer, and several studies show a correlation between pDC infiltration and poor clinical outcome (Aspord et al., 2013; Hartmann et al., 2003; Labidi-Galy et al., 2011; Labidi-Galy et al., 2012; Sisirak et al., 2012; Treilleux et al., 2004). TA-pDCs harbor a tolerogenic phenotype; they possess decreased co-stimulatory molecule expression, reduced ability to secrete IFN-I, along with increased ability to induce Tregs (Aspord et al., 2013; Conrad et al., 2012; Faget et al., 2012; Hartmann et al., 2003; Sharma et al., 2007; Sisirak et al., 2012). Nonetheless, the immunogenicity of pDCs that are not immersed in the TME can be harnessed to elicit efficient anti-tumor responses (Aspord et al., 2012; Liu et al., 2008a; Loschko et al., 2011b; Tel et al., 2013a; Tel et al., 2012b). Consequently, pDCs are interesting candidates for the use of specific immunotherapies in order to enhance the immune response directed against tumors. We had previously demonstrated that, after contralateral vaccination of mice with CpG-B, in combination with a MHC-II-restricted tumor antigenic peptide, pDCs were activated in LNs distal from the tumor, where they acted as immunogenic APCs, inducing tumor antigen-specific Th17 cells. This induced the recruitment of immune cells into the tumor and subsequent CTL- mediated tumor cell death (Fig. 15) (Guery et al., 2014). Whether TA-pDC tolerogenic phenotype can be reverted following intratumoral injection of CpG-B, and how this approach impacts the TME remains unclear. Here, we aimed at determining whether TA-pDCs could undergo a tolerogenic-to-immunogenic reprogramming following intratumoral administration of CpG-B (alone or along with a MHC-II- restricted tumor antigenic peptide) in murine established tumors (Fig. 16). We originally sought to characterize the MHC-II-mediated antigen-presenting functions of pDCs immersed in the TME. However, this study led us to also investigate how this local immunotherapeutic approach would impact, besides pDCs, other leukocytes infiltrating the tumor.

Result summary Using thymoma or melanoma tumor models, we found a significant tumor growth control after intratumoral CpG-B delivery, which is in agreement with other studies (Carpentier et al., 2010; Grauer et al., 2008; Sharma et al., 2004). However, in most other studies, CpG was used in combination with other treatments, making difficult a precise comparison between the mechanisms involved (Brody et al., 2010; Carpentier et al., 2010; Nierkens et al., 2011). The addition of the MHC-II-restricted peptide led to different outcomes depending on the tumor model used. This could be due to different immunodominant epitopes, depending on the tumor. Adaptive immunity was required for the vaccine to be effective, which is in agreement with other studies (Brody et al., 2010; Geary et al., 2011). We also observed a decrease in the [effector T cells/Treg] ratio, in accordance with other studies (Geary et al., 2011; Grauer et al., 2008). However, this treatment did not lead to a reversion of the tolerogenic TA-pDC phenotype. Moreover, in the thymoma model, genetic pDC depletion did not abrogate the efficacy of the treatment in controlling tumor growth. Furthermore, in this model, TA-pDCs remained refractory to TLR9 ligand treatment and were not implicated in the mechanism of action of the vaccine. On the contrary, conventional DCs, i.e. cDC1 (CD8α+ cDCs) and cDC2 (CD11b+ cDCs), were activated following i.t. CpG-B administration. The control of tumor growth occurred rapidly, 5 days after i.t. CpG-B. Therefore, T cells could be activated locally in the tumor (Peske et al., 2015). However, upon blockade of T cell egress from the TdLNs, there was an abrogation of tumor growth inhibition after i.t. CpG-B. Effector CD4+ and CD8+ T cells were sequestered in TdLNs, while effector CD4+ and CD8+ T cells infiltration was reduced in the tumor. This suggests that APCs, likely cDCs, travel from the tumor to the TdLNs, where they present tumor antigens to T cells, which subsequently migrate to the tumor, in order to perform their effector functions. Finally, we observed an increased recruitment of activated neutrophils in the tumor and their depletion abrogated the effect of CpG-B i.t. delivery on tumor growth. It also altered cDC activation and CD8+ T cell response priming. These results raise a certain number of questions that are addressed in the following paragraphs.

Discussion and perspectives

Phenotype and implication of pDCs In our model, we did not observe any tolerogenic-to-immunogenic conversion of TA-pDCs. Intratumoral CpG-B (alone or along MHC-II-restricted tumor antigenic peptide) did not induce

an upregulation of MHCII or co-stimulatory molecules expression (FACS) nor did it induce an immunogenic phenotype (RNA-seq). Several explanations may account for these results. A first possibility would be that pDCs cannot act as APCs and contribute to immunity exclusively as innate immune cells producing IFN-I. Indeed, two recent studies, using single-cell RNA-seq and/or CyTOF, have raised concerns about the potential of pDCs to present antigens (See et al., 2017; Villani et al., 2017). The authors concluded that pDCs cannot activate T cells and that previously published results demonstrating a role for pDCs as APCs are a consequence of a contamination of pDCs with another cell population (i.e. pre-DCs or DC5, depending on the publication), which gives rise to cDCs and has the capacity to present antigens (See et al., 2017; Villani et al., 2017). The assumption that pDCs cannot present antigens is based on in vitro presentation assays using human blood pDCs in the steady state, which are, therefore, not activated (See et al., 2017; Villani et al., 2017). The observation that they cannot induce T cell expansion is therefore expected in this setting. This interpretation does not take into account the possibility that pDCs may acquire antigen-presenting functions in other immunological contexts in which inflammatory signals could induce a maturation of pDCs. Hence, we rule out the possibility that pDCs can never act as APCs as an explanation for our results. Another possibility is that a small fraction of TA-pDCs could have undergone a tolerogenic-to- immunogenic conversion but that it could not be observed by analyzing the bulk of pDCs by FACS or RNA-seq; results of a potential activation of a subpopulation might have been “diluted”. As developed in the introduction, human pDCs form a heterogeneous population due to reasons including tissue specialization and developmental origins (Alcantara-Hernandez et al., 2017; Alculumbre et al., 2018b; Rodrigues et al., 2018; Zhang et al., 2017). A subpopulation of human pDCs (CD2hi CD5+ CD81+), with a corresponding population in mice, expressing classical pDC markers and having increased ability to activate T cells in response to stimulation has been characterized (Zhang et al., 2017). Furthermore, only human pDCs derived from CDP were found to process and present antigens, as compared with CLP-derived pDCs (Rodrigues et al., 2018). Finally, three distinct human pDC populations were characterized, based on the expression of PD-L1 and CD80, after stimulation with influenza virus (Alculumbre et al., 2018b). The population P3 (CD80+ PD-L1-) had increased capacity to induce the expansion of CD4+ T cells. The three populations were stable after secondary stimuli and did not depend on pre- existing variability before stimulation (Alculumbre et al., 2018b). However, whether distinct

subpopulations are similarly generated after exposure of pDCs to tolerogenic situations remains to be elucidated. The possibility that the results associated with a putative re-activation of a small pDC subpopulation, which might have been diluted by analyzing the bulk of pDCs in our model, cannot be excluded, although the above-mentioned subsets were found in human and there is no evidence for such subsets in mouse. It would, therefore be interesting to determine whether there are different pDC subpopulations in our model, raising additional questions if it was the case: What would be the phenotype of these pDC subsets in the TME? Would they all be maintained in a tolerogenic state by the TME? - If yes, would all these pDC subsets be refractory to i.t. CpG-B treatment or could some of the subsets undergo a tolerogenic-to-immunogenic activation? In which case, why some subsets could be re-activated while other could not (different ontogeny? pre-activation with another stimulus rendering them unresponsive/responsive to a subsequent distinct stimulus?)? What would be the properties rendering pDC subsets responsive to re-activation, compared with non-responsive pDC subsets? - If not, what would make some pDC subsets resistant to/protected against the immune- suppressive TME?

Plasmacytoid DC depletion, using BDCA2-DTR mice, prior to i.t. CpG-B delivery did not impact the effect of this treatment on tumor growth. These results are in discrepancy with two studies, in which pDCs were depleted using antibodies, leading to an abrogation of the effects of i.t. CpG administration on tumor growth (Lou et al., 2011; Nierkens et al., 2011). However, Nierkens and colleagues used CpG-B administered peritumorally in combination with cryosurgery, which might lead to a different mechanism (Nierkens et al., 2011). Lou et al. did not specify the class of CpG used (Lou et al., 2011). In both cases, the pDC phenotype after treatment was not assessed. Our results regarding pDC depletion and the absence of pDC phenotype modulation analyzed by FACS and RNA-seq are in concordance. Consequently, we believe that TA-pDCs are not implicated in the mechanism of action leading to tumor growth inhibition following i.t. CpG-B delivery and that these cells are refractory to a tolerogenic-to-immunogenic reprogramming by CpG-B. Moreover, the effect of i.t. CpG-B administration on tumor growth was not impaired in IFNAR-/- mice. This rules out a possible role for TA-pDC IFN-I production in our model.

Activation and implication of cDCs Two distinct subsets of cDCs have been described: murine cDC1 are defined as CD103+ and/or CD8+ and are particularly efficient at cross-presenting exogenous antigens via MHC-I to CD8+ T cells, while cDC2 are defined as CD11b+ and are specialized in the presentation of exogenous antigens to CD4+ T cells via MHC-II (Anderson et al., 2018; Dalod et al., 2014; Guilliams et al., 2014; Merad et al., 2013). Following i.t. CpG-B administration, the expression of MHC-II and co-stimulatory molecules was upregulated (FACS) in both tumor-associated CD8+ cDCs and CD11b+ cDCs, in agreement with other studies (den Brok et al., 2006; Nierkens et al., 2011). These results were supported by RNA-seq analysis, which showed the upregulation of several genes in TA-cDCs with implications in immunogenicity, antigen presentation and migration. Unfortunately, we analyzed the bulk of TA-cDCs and did not assess the phenotype of TA-cDC1 and TA-cDC2 separately by RNAseq, which may have brought insight into the mechanism of action of i.t. CpG-B delivery and the role of TA-cDCs. In addition, the frequency of CD103+ cDCs (CD103+ CD8+ or CD103+ CD103-) in tumors was very low, preventing to further analyze this subset. Therefore, the role of CD103+ cDCs, which has been shown to be important for the transport of antigens from the tumor to the TdLNs, might have been overlooked in our setting, due to technical issues (Salmon et al., 2016). Altogether our results show that, unlike TA-pDCs, TA-cDCs are activated following i.t. CpG-B. Whether these cells migrate to the TdLNs, in order to present antigens to T cells, remains to be investigated. It would also be interesting to use mice such as CD11c-DTR mice, in which cDCs can be depleted, in order to determine whether these cells are implicated in the mechanism of action of i.t. CpG-B delivery. Our results also raise the question as to how TA-cDCs are activated. The possibility that IFN-I might activate cDCs is unlikely, as i.t. CpG-B administration in IFNAR-/- mice did not impact the effect of this treatment on tumor growth, although, as above- mentioned, the implication of cDCs in the mechanism leading to tumor growth inhibition remains to be directly demonstrated. In mouse, several cell types express TLR9, including cDCs, in the contrary to human, for which TLR9 expression is restricted to pDCs and B cells (Jarrossay et al., 2001; Puttur et al., 2016; Takeuchi and Akira, 2010; Xu et al., 2015). Therefore, in our model, cDCs may be directly activated by CpG-B ligation on TLR9, although this remains to be elucidated. In this regard, an activation of cDCs through the TLR9/MyD88 signaling pathway has been implicated in antiviral

100

immune responses, with pDC depletion having no impact on antiviral response (Puttur et al., 2016; Xu et al., 2015). Finally, TA-cDCs might be indirectly activated following i.t. CpG-B administration by other immune cell types, including neutrophils. This idea is developed below.

Why are TA-pDCs refractory to i.t. CpG-B? Whether TA-pDCs can be re-activated is still a matter of debate. In our model, we did not observe any tolerogenic-to-immunogenic reconversion of TA-pDCs following i.t. CpG-B administration. Conflicting results might be explained by several reasons. First, there might be differences depending on the type of cancer, which could lead to different results as the TME highly depends on the type of tumor (Michea et al., 2018). Furthermore, discrepancies might be due to distinct experimental settings. For instance, different results might be obtained by studying the phenotype of TA-pDCs isolated from tumors and treated with TLR ligands ex vivo, compared with TA-pDCs isolated from tumors treated in vivo with intratumorally- administered TLR ligands. Indeed, being extremely plastic cells, TA-pDCs treated ex vivo might loose their unresponsiveness when taken out of the TME. As previously mentioned, the TME maintains pDCs in a tolerogenic state, which may prevent their activation. Molecules found in the TME and involved in the tolerogenic phenotype of TA-pDCs include TGF-β, TNF-α, IL-10 and IL-3 (Alculumbre et al., 2018a; Beckebaum et al., 2004; Ray et al., 2017; Sisirak et al., 2013; Terra et al., 2018). Some studies may have shown that TA-pDCs could be reprogrammed ex vivo, but it is not the case if isolated TA-pDCs are in presence of tumor-conditioned medium or specific cytokines found in the TME (Chauhan et al., 2009; Ray et al., 2014; Terra et al., 2018). In this regard, Terra et al. showed that TGF-β in the TME is responsible for the inability of TA-pDCs to produce IFN-I, and other cytokines and chemokines, in response to TLR9L (Terra et al., 2018). Finally, conflicting results may arise from the use of different TLR ligands, with certain TLR ligands able to activate TA-pDCs, while others are not; use of TLR7L vs TLR9L, or use of CpG- B vs CpG-A or CpG-C. Studies have shown that specific TLR7L could reprogram TA-pDCs, although the ligands responsible for activation were not the same (Le Mercier et al., 2013; Perrot et al., 2007). Perrot and colleagues showed in human NSCLC that R848 (TLR7/TLR8 ligand) could induce a modest upregulation of CD86 in TA-pDCs ex vivo, and LeMercier et al. demonstrated that Flu (TLR7) could activate TA-pDCs ex vivo and in vivo in a mouse model of breast cancer, while CL075, R848 and SM360320 (other TLR7L) could not (Le Mercier et al., 2013; Perrot et al., 2007). The reason why some but not all TLR7L can reprogram TA-pDCs remains to be elucidated.

101

Furthermore, Imiquimod, a TLR7L approved for the treatment of skin cancers (topical application), was shown to induce the recruitment of pDCs in skin neoplasms (Drobits et al., 2012; Kobold et al., 2014; Stary et al., 2007; Urosevic et al., 2005). In some of these studies, recruited pDCs acquired direct tumoricidal activity through the secretion of TRAIL and/or granzyme B (Drobits et al., 2012; Kobold et al., 2014; Stary et al., 2007). The acquisition of direct cytotoxic functions by pDCs following topical imiquimod application was also found in a humanized mouse model of melanoma (Aspord, 2014 24751730). Finally, efficient production of IFN-I by pDCs was observed in skin neoplasms following the same treatment, although pDCs were recruited after topical imiquimod application, therefore not demonstrating a local reprograming of TA-pDCs (Kobold et al., 2014; Urosevic et al., 2005). Altogether, these studies suggest that tolerogenic TA-pDCs may be reprogrammed by certain types of TLR7L, although it has been demonstrated by very few studies.

The reason why TA-pDCs are refractory to i.t. CpG-B administration remains to be elucidated. One of the possible explanations would be a downregulation of TLR9 expression by TA-pDCs (Hartmann et al., 2003). Another possibility could be the upregulation of inhibitory receptors by TA-pDCs compared to pDCs unexposed to the TME, such as LAG-3, protein-tyrosine phosphatase receptor (PTPRS in human; PTPRF in mouse) or receptors similar to brain and dendritic cell-associated lysosome-associated membrane protein (BAD-LAMP; found in human) (Bunin et al., 2015; Camisaschi et al., 2014; Defays et al., 2011). Finally, TA-pDCs could have an exhausted phenotype. Plasmacytoid DC exhaustion due to sustained TLR7 signaling, leading to downregulation of E2-2 and subsequent pDC loss of function, with poor IFN-I production, has been observed in a mouse model of chronic viral infection (Macal et al., 2018). It is unclear whether pDCs can be activated by DAMPs in the TME (Alculumbre et al., 2018a). Molecules such as HMBG1 (TLR4 and TLR9 ligand) and oxidized mitochondrial DNA (TLR9 ligand), released by dying tumor cells, could be implicated in sustained TLR9 signaling, leading to a putative exhaustion and subsequent unresponsiveness to posterior TLR9 stimuli (Alculumbre et al., 2018a; Caielli et al., 2016). Prolonged TLR signaling could lead to epigenetic silencing of genes required for the activation of pDCs and stable “imprinting” of an exhausted phenotype. Furthermore, TLR7L, TLR9L or cytokines in the TME could induce a pDC phenotype that remains stable even after stimulation with a secondary stimulus, in the same line as the above- mentioned stable subpopulations of human pDCs characterized by Alculumbre et al. (Alculumbre et al., 2018b).

102

The fact that TA-pDCs remained refractory to i.t. CpG-B administration, while this treatment led to an activation of TA-cDCs, is also puzzling. The TME has been demonstrated to differentially impact DC subsets (Michea et al., 2018). As mentioned above, TLR9 downregulation, which could be implicated in TA-pDCs hyporesponsiveness to TLR9 ligand, could specifically occur in TA-pDCs but not in TA-cDCs (Hartmann et al., 2003). In addition, TA-cDCs could be activated via TLRs not expressed by pDCs, i.e. other than TLR7 and TLR9 (Takeuchi and Akira, 2010). TA-cDCs could also be activated by an interaction with neutrophils (explained in the next paragraph), which would not take place between TA-pDCs and neutrophils, for example interaction of macrophage-1 antigen (Mac-1), on neutrophils, with dendritic cell-specific intercellular adhesion molecule-3-grabbing non-integrin (DC-SIGN), on cDCs (Schuster et al., 2013; van Gisbergen et al., 2005). In addition, the TME could induce an upregulation of inhibitory receptors selectively expressed by pDCs but not by cDCs (Bunin et al., 2015; Combes et al., 2017; Defays et al., 2011). Another possible explanation could be a differential metabolic reprograming taking place in pDCs vs cDCs upon TLR ligation (Basit et al., 2018)6. Indeed, human DC subsets from peripheral blood undergo metabolic reprogramming that are required for their activation. It has recently been demonstrated that the metabolic reprogramming occurring in pDCs and cDCs upon TLR7/8 ligation in vitro are distincts (Basit et al., 2018)5. Therefore, the TME could prevent metabolic reprograming of pDCs, due for example to the lack of availability in specific nutrients, such as glutamine (pDC metabolic reprogramming and activation involves glutaminolysis), preventing TA-pDC activation, while keeping TA-cDCs responsive to activation, as these specific nutrients would not be necessary for cDC activation. Whether and how different TMEs affect the metabolic reprograming of DC subsets, and their subsequent ability to be activated, remain to be investigated and are of major interest, since it could lead to the identification of novel therapeutic targets.

Recruitment and implication of neutrophils Neutrophils have a diverse range of biological functions (Yang et al., 2017). They have direct cytotoxic activity; they phagocyte, perform intracellular degradation, release cytotoxic granules and form neutrophil extracellular traps (NETs) (Yang et al., 2017). In addition, they can orchestrate immune responses and directly interact with immune cells from the innate and

6 Basit et al. (2018) Human dendritic cell subsets undergo distinct metabolic reprogramming for immune response. Front. Immunol. 9:2489

103

adaptive compartments (Yang et al., 2017). The function of TA-neutrophils (TANs) is a matter of debate. These cells, which are plastic, can play either an anti- or pro-tumoral role, depending on different factors, such as the tumor stage and the type of tumor (Massara et al., 2017; Nicolas- Avila et al., 2017; Shaul and Fridlender, 2018; Zilio and Serafini, 2016). We observed an increased infiltration of neutrophils in the tumor after i.t. CpG-B±MHC-II- restricted peptide. One-shot injection of neutrophil depleting antibodies led to an abrogation of the effect of i.t. CpG-B on tumor growth, showing that neutrophils were implicated in the mechanism of action of the treatment. This depletion resulted in decreased cDC activation (both cDC1 and cDC2), and decreased CD8+ T cell activation. Moreover, TANs isolated from CpG-B- injected tumors produced higher amounts of chemokines (CCL3, CCL4 and CCL5) that can be involved in the recruitment and/or activation of cDCs (Bennouna et al., 2003; Charmoy et al., 2010; Trifilo and Lane, 2004). Neutrophils have been shown to regulate the recruitment and/or the activation of cDCs by different mechanisms (Schuster et al., 2013). Our findings raise the question whether it is the case in our setting. First, the above-mentioned chemokines could be implicated in cDC recruitment and/or activation (Bennouna et al., 2003; Charmoy et al., 2010; Trifilo and Lane, 2004). In addition, neutrophils could activate cDCs via a cell-cell contact-dependent mechanism involving interaction between Mac-1 and DC-SIGN or between CD18 and intercellular adhesion molecule 1 (ICAM-1), at the surface of neutrophils and cDCs, respectively (Schuster et al., 2013; van Gisbergen et al., 2005). Neutrophils can also release DAMPs (Schuster et al., 2013). Finally, they can release NETs, which could induce an amplification of a putative CpG-B-induced activation of cDCs through TLR9, due to a release of self-DNA and antimicrobial peptides, as observed in autoimmune settings with the activation of pDCs by NETs (Diana et al., 2013; Lande et al., 2011; Schuster et al., 2013). Neutrophils could also secrete cytokines that activate cDCs, or be phagocytosed by cDCs after having acquired tumor antigens. Furthermore, whether the effect of neutrophils on anti-tumor CD8+ T cell response activation is direct, or mediated by cDCs, remains to be determined. Regarding a possible direct effect of neutrophils on CD8+ T cell response, antigen cross-presentation to CD8+ T cells by neutrophils in vivo has been reported in the literature (Beauvillain et al., 2007). In addition, neutrophils could be implicated in the recruitment of CD8+ T cells in the tumor (Lim et al., 2015). Finally, neutrophils could have a direct cytotoxic activity against tumor cells (Massara et al., 2017; Yang et al., 2017; Zilio and Serafini, 2016). Indeed, we observed a very rapid effect of i.t. CpG-B on tumor growth. However, the effect of the treatment was completely abrogated in Rag2-/- mice, suggesting that neutrophils are involved in the orchestration of the adaptive immune response.

104

Mouse neutrophils have been reported to express TLR9 at the cell surface, as opposed to endosomes for pDCs, raising the question whether they could be directly activated by CpG-B (Lindau et al., 2013; Schneberger et al., 2013). However, we did not observe any effect of CpG-B on ex vivo TAN ROS production, used as readout for neutrophil activation. Nonetheless, we observed an increased frequency of neutrophils in the tumor after i.t. CpG-B. Adhesion molecule expression was higher in TANs in comparison with neutrophils in the TdLNs, which suggests that this upregulation could enable neutrophils to migrate from the periphery to the tumor. However, there was no difference between the TANs from tumors injected or not with CpG-B. This suggests that i.t. CpG-B delivery did not modify the activation status of TANs following i.t. CpG-B delivery, but rather induced changes in TME that promote the recruitment of neutrophils. Accordingly, tumor-conditioned media (TCM) led to increased in vitro migration of BM-derived neutrophils compared to normal media. The attraction was slightly increased with TCM from CpG-B-injected tumors compared with TCM from control tumors, although the difference was not significant. The changes in the TME induced by i.t. CpG-B and leading to a recruitment of neutrophils remain to be determined, i.e. whether they result from changes in immune cells, stromal cells or tumor cells themselves. Neutrophils could also be attracted by infiltrating T cells (Appelberg, 1992). A recently published study shows that neutrophil migration into tumors is mediated by ATP signaling via purinergic receptors (Patel et al., 2018). An analysis of tumor supernatants could shed light on soluble factors potentially implicated in neutrophil recruitment. A complete characterization of neutrophil phenotype after i.t. CpG-B injection would be of interest, as well as determining, if it was the case, how these neutrophil phenotypic/functional changes would be related to tumor growth inhibition. Neutrophil features to be investigated include their polarization (N1 vs N2), with nitric oxide production, NETosis and arginase expression (Nicolas-Avila et al., 2017). Our study highlights an anti-tumoral role of neutrophils in the context of solid tumors, which is of major interest as the anti- or pro-tumoral role of neutrophils is widely debated in the current literature (Massara et al., 2017; Nicolas-Avila et al., 2017; Shaul and Fridlender, 2018; Zilio and Serafini, 2016). Altogether, our results on neutrophils contribute to this debate. We also suggest a novel potential for anti-cancer therapeutic approaches harnessing the anti- tumor activity of neutrophils (Heemskerk and van Egmond, 2018; Massara et al., 2017; Shaul and Fridlender, 2018; Treffers et al., 2016; Zilio and Serafini, 2016).

105

Role of other immune cells Analysis of other immune cells could also be of interest in order to understand the mechanism of action of i.t. CpG-B. For instance, increased infiltration of macrophages or at least partial requirement of macrophages has been implicated in studies, although the type of CpG was not specified (Auf et al., 2001; Carpentier et al., 2000; Schettini et al., 2012). Macrophage recruitment in tumors could also contribute to the anti-tumoral effect of i.t. CpG-C combined with anti-PD-1 treatment (Sagiv- Barfi et al., 2018). In our model, we did not observe any change in the frequency of macrophages in tumor injected with CpG-B or controls. Nonetheless, we did not analyze the phenotype of TA-macrophages that could convert from M2 (pro-tumoral) to M1 (anti-tumoral), as observed after i.t. administration of TLR7L in combination with anti-PD-1 treatment in a mouse model of head and neck cancer (Sato-Kaneko et al., 2017). In addition, it might be of interest to analyze the role and phenotype of NK cells in our setting. Indeed, NK cells were suggested to be at least partially required for antitumoral effect of i.t. CpG, or increased NK infiltration was observed after this treatment, although the type of CpG was not always specified (Carpentier et al., 2000; Geary et al., 2011; Schettini et al., 2012; Sharma et al., 2008). In addition, the TLR7 agonist Imiquimod led to activation of NK cells required for tumor cell killing (Doorduijn et al., 2017). Furthermore, cDCs could activate NK cells (Puttur et al., 2016). Conversely, if NK cells played a role in the mechanism of action of i.t. CpG-B, they could be implicated in the recruitment of cDCs (Bottcher et al., 2018).

Working model

The working model we propose involves a cooperation between neutrophils, cDCs and T cells (Th1 and CTLs) that leads to tumor growth control (Fig. 17) [Appendix 2 (Humbert and Hugues, 20187)]. The TME is affected by i.t. CpG-B administration, promoting an increased production of molecules attracting neutrophils, either by tumor-infiltrating immune cells or by tumor cells themselves (Fig. 17). Consequently, there is a rapid recruitment of neutrophils into the tumor after local CpG-B delivery, associated with an enhanced cDC activation (Fig. 17). Upon i.t. CpG-B administration, activated cDCs are recruited from the tumor to the TdLNs, in order to activate tumor antigen-specific CTLs and Th1 cells (Fig. 17). As a consequence, effector T cells infiltrate the tumor, which also contains decreased Treg numbers, modulating the [effector T cell/Treg] ratio in favor of effector T cells and resulting in tumor cell killing and tumor growth

7 Appendix 2: Humbert and Hugues (2018) Warming up the tumor microenvironment in order to enhance immunogenicity. Oncoimmunology, e1510710.

106

control (Fig. 17). Noteworthy, this cascade of events depends on the intratumoral recruitment of neutrophils following i.t. CpGB delivery, as both the activation of cDCs and the priming of tumor antigen-specific T cells are abrogated after neutrophil depletion. We noticed that the efficacy of CpG-B administration regarding tumor growth control was potentialized when used in combination of a tumor antigenic peptide. This observation suggests that in conditions where tumor antigens are available for presentation to T cells, in situ delivery of TLR9 agonist could lead to a dysfunctional-to-immunogenic conversion of APCs, generating strong anti-tumor T cell responses.

Figure 17. Intratumoral administration of CpG-B and tumor antigenic peptide induces tumor cell death: working model 1. Intratumoral administration of CpG-B and tumor antigenic peptide 2. Activated neutrophil recruitment into the tumor 3. Neutrophils are involved in conventional dendritic cell (cDC) activation 4. Activated cDCs migrate towards the tumor-draining lymph nodes (TdLNs) 5. Activated cDCs prime Th1 cells and cytotoxic T lymphocytes (CTLs) in the TdLNs 6. CTLs and effector Th1 cells infiltrate the tumor and produce IFN-γ 7. Treg numbers in the tumor are reduced 8. Tumor cells are killed by CTLs Adapted from Humbert and Hugues, Oncoimmunology, 2018 [Appendix 2 (Humbert and Hugues, 2018)], using data from Humbert et al., Cancer Res, 2018 (Humbert et al., 2018).

107

Importance of the immunological context

This study highlights the importance of the immunological context regarding pDC phenotype, as well as T cell polarization. Indeed, as mentioned previously, pDCs are extremely plastic cells and their phenotype and functions are highly context-dependent (Alculumbre et al., 2018a; Guery and Hugues, 2013). Regarding pDC activation, here we show that TA-pDCs in our model are refractory to a tolerogenic-to-immunogenic reprogramming by i.t. delivery of CpG-B±MHC-II- restricted peptide. On the contrary, our group previously showed that pDCs in distal LNs could be activated by the same treatment delivered s.c. at a site distal from the tumor (Guery et al., 2014). This contrast in pDC response to treatment, depending on whether they are immersed in the TME or not, shows that TA-pDCs are deeply “imprinted” by the TME, preventing their reprogramming, while “untouched” pDCs in distal LNs can be responsive to the treatment. These observations suggest that pDCs may lose their plasticity when they are strongly imprinted in a specific immunological microenvironment. Regarding T cell polarization, s.c. injection of CpG-B±MHC-II-restricted peptide at a distal site from the tumor led to Th17 cell differentiation, while i.t. delivery led to a Th1 polarization, likely due to the differences in APCs implicated, but also due to the pre-existing microenvironment (Fig. 15 and Fig. 17) (Guery et al., 2014).

Local treatment administration: clinical application

Different outcomes depending on mouse vs human TLR expression pattern and on TLR ligands types The cellular patterns of TLR9 expression differ between mice and humans. Indeed, in human, TLR9 expression is restricted to pDCs and B cells (Jarrossay et al., 2001; Takeuchi and Akira, 2010). On the contrary, in mouse it is expressed by other cell types, including cDCs and neutrophils (Lindau et al., 2013; Puttur et al., 2016; Schneberger et al., 2013; Xu et al., 2015). In addition, we can expect a different outcome on the immune response, depending on the class of CpG administered i.t. Class A CpG-ODN (CpG-A) leads to the secretion of IFN-I by pDCs, while CpG-B induces cytokine secretion, including IL-12 and IL-6, and pDC maturation (Gilliet et al., 2008; Honda et al., 2005; Vollmer and Krieg, 2009). CpG-C promotes both IFN-I production and pDC maturation (Vollmer and Krieg, 2009). We used CpG-B because our primary goal was to study the role of antigen-presenting functions of pDCs. In many studies, the effects of i.t. CpG were mediated by IFN-I production. Sagiv-Barfi et al., Gallota et al. and Wang et al. used CpG-C and the anti-tumor effects were at least partly mediated by IFN-I (Gallotta et al., 2018; Sagiv-Barfi et al., 2018; Wang et al., 2018a). Therefore, the

108

immune response induced was not the same as the one triggered by i.t. CpG-B, as this treatment also works in IFNAR-/- mice, in our hands. Despite these differences, further dissecting the mechanisms of protection afforded by i.t. CpG- B administration may improve future anticancer therapeutic strategies.

Advantages of local administration Malignancy treatments, which include immunotherapies, are most of the time administered systemically. Therefore, they are often linked with damaging effects. Therefore, the local administration of immunotherapies, including intratumoral, topical, intranodal, intracranial and intradermal, is considered highly promising and could solve many aspects that limit immunotherapeutic treatments (Fransen et al., 2013; Marabelle et al., 2014). Local administration requires lower dosage to reach the desired concentration, leading to decreased drug levels in the organism, hence, preventing autoimmune risks, and increasing the strength of the anti-tumor response (Fransen et al., 2013; Weiden et al., 2018).

Limitations of local administration In our model, the local administration of CpG-B did not lead to an efficient systemic anti-tumor immune response, also called abscopal effect, in our models, as we observed a rather modest inhibition of non-injected contralateral tumors. This result is in accordance with a recent study demonstrating that intratumoral CpG-C alone does not affect distal tumor growth (Sagiv-Barfi et al., 2018). This is an important drawback of local administration. Indeed, the induction of a systemic effect is important in order to avoid or reduce metastases. However, as this intratumoral approach has many advantages, its optimization in order to induce an abscopal effect with the same or increased magnitude as induced by systemic treatment while limiting adverse effects, seems promising.

Warming up the TME The goal of immunotherapy is to induce a robust tumor antigen-specific immune response. Nonetheless, as previously mentioned, the TME negatively affects the immune system (Zitvogel et al., 2006). In this respect, one of the aims of locally administered immunotherapy is to overcome the TME inhibitory properties in order to “warm up” the tumor and subsequently boost its immunogenicity, leading to the induction of effector immune cells [Appendix 2

109

(Humbert and Hugues, 2018)8]. A possible approach is the local administration of TLR ligand agonists or other drugs, to reactivate the immune cells directly at the site of injection. Intratumoral immunotherapies may convert “cold” into “hot” TME (Schaedler et al., 2017; Wang et al., 2018a), therefore boosting the potential for a strong immune response following its combination with systemic immunotherapies. In this regard, sialic acid sugars that promote an immunosuppressive TME can be blocked by i.t. injection of inhibitor, leading to an immune-permissive TME favorable for the use of adoptive T cell transfer (Bull et al., 2018). In addition, i.t. CpG-C administration in combination with agonistic anti-OX40 ligand leads to tumor eradication and systemic anti-tumor response in mouse (Sagiv-Barfi et al., 2018). The issue of resistance to immune checkpoint blockade represents one of the main actual challenge in immunotherapy, which is otherwise a revolutionary approach9. Checkpoint inhibitor blockade acts by removing a “brake” for anti-tumor T cell response, therefore, a pre-existing anti- tumor T cell response in patients is required in order to respond to this therapy (Sharpe and Pauken, 2018). This is a reason why several patients (non-responders) fail to respond to this treatment, although it works very well for other patients (responders) (Sharpe and Pauken, 2018). Therefore, warming up the TME is of particular importance to overcome immune checkpoint blockade resistance in non-responder patients (Sharpe and Pauken, 2018). TLR1L/TLR2L or TLR7L/TLR9L, in combination with anti-PD-L1, led to tumor eradication in melanoma or head and neck tumors, respectively, in mice (Sato-Kaneko et al., 2017; Wang et al., 2018b). Finally, i.t. CpG-C along with anti-PD-1 treatment in mice led to tumor growth inhibition in several types of tumor mouse models (Gallotta et al., 2018; Wang et al., 2016).

Technical limitations Certain limitations actually refrain the development of local immunotherapeutic approaches. These limitations include the necessity to optimize the delivery methods, such as the route, dose and volume, as well as the tumor accessibility. Indeed, i.t. TLRL administration may seem difficult to achieve in the clinic, depending on the type of cancer. Different techniques can be used: topical (e.g. melanoma), intradermal, intratumoral, intracranial (e.g. glioma), peritumoral, intranodal (e.g. lymphoma), via the respiratory tract (e.g. lung tumor) or through systemical administration of adjuvants coupled with molecules enabling tumor-targeting. Development of

8 Appendix 2: Humbert and Hugues (2018) Warming up the tumor microenvironment in order to enhance immunogenicity. Oncoimmunology, e1510710. 9 Nobel prize in physiology or medicine 2018 was awarded for the discovery of the immune checkpoint molecules CTLA-4 and PD-1 (https://www.nobelprize.org/prizes/medicine/2018/summary/)

110

new technologies, such as monoclonal antibody-coated microparticles, radiation-targeted delivery of nanoparticles, scaffolds, polymers and lipids, will allow a widen use of intratumoral administration, without being too invasive (Appelbe et al., 2017; Bahmani et al., 2018; Bookstaver et al., 2018; Smith et al., 2017; Weiden et al., 2018).

Recent clinical trials Despite these limitations, in situ vaccination strategies are possible and can lead to immune cell activation and tumor growth inhibition, with reduced risk of adverse systemic effects. For instance, it has been demonstrated that melanoma patients that received a single local injection of low-dose CpG-B, at the site of tumor excision after surgery, were protected against relapse (Koster et al., 2017). This protection was associated with LN cDC activation. In addition, two recently (October 2018) published clinical trials show that combining TLR9L administered locally, which converts the TME into an immune-permissive environment, with another immunotherapeutic strategy is possible and gives very good results in human. Franck and colleagues showed that in situ administration of SD-10110 (CpG-C) in combination with local low- dose radiotherapy led to a reduction of tumors, at the injection site, of most of the indolent lymphoma patients included in the clinical trial (Frank et al., 2018). The treatment led to an increased [effector T cells/Tregs] ratio in the TME and to systemic immune response, with tumor growth inhibition in untreated sites (Frank et al., 2018). Furthermore, intratumoral injection of SD-101 in combination with systemically administered anti-PD-1 treatment in advanced (metastatic or unresectable) melanoma patients induced tumor regression at treated and untreated sites, with 78% overall response rate in patients previously naïve to PD-1 blockade (Ribas et al., 2018).

Conclusion

Further investigations are necessary to elucidate the mechanisms accounting for the protection observed after intratumoral CpG-B delivery, and may improve anti-tumor immunotherapies. As mentioned above, the combination of different interventions including the blockade of checkpoint inhibitors and techniques that lead to tumor antigen release seem a promising approach. The development of new technologies for i.t. delivery should also improve anticancer therapies, in order to treat cancer types that are more difficult to reach.

10 http://www.dynavax.com/our-pipeline/cancer-immunotherapy/sd101/

111

IV. CHAPTER B

IV.1. Introduction

IV.1.a. Multiple sclerosis and its animal model, experimental autoimmune encephalomyelitis

IV.1.a.i. Multiple sclerosis

Multiple sclerosis (MS) is a chronic inflammatory autoimmune demyelinating disease of the CNS, which affected 2.5 million people worldwide in 2015, although its incidence varies greatly across world areas (Dendrou et al., 2015). The onset occurs usually in early- to mid-adulthood, and the prevalence is higher in women (Dendrou et al., 2015; Whitacre, 2001). MS is a heterogeneous disease, with clinical characteristics that depend on the location of lesions within the CNS (Roman and Arnett, 2016). These lesions originate from an autoimmune attack against myelin, which is produced by oligodendrocytes and protects neuron fibers, leading to a rupture in neuron electric signals (Compston and Coles, 2008; Dendrou et al., 2015; Hemmer et al., 2015). Patients are divided into three main clinical groups. About 85% of patients suffer from the relapsing-remitting form of MS (RRMS) at onset, which consists in acute relapses during several days, followed by remitting periods (Antel et al., 2012; Roman and Arnett, 2016). About 80% of the RRMS patients evolve towards a secondary progressive MS (SPMS) form, in which there is no remitting phase and that is associated with CNS atrophy and axonal loss. The remaining 20% stay in the RRMS clinical form. Finally, about 15% of patients suffer from a primary progressive (PPMS) form, in which the disease is in constant progression since onset (Antel et al., 2012; Roman and Arnett, 2016).

IV.1.a.ii. Causes

Although the causes of MS are not fully understood, the consensus points towards genetic susceptibilities, accounting for 30% of the pathogenesis, in combination with environmental factors, which account for 70% of the disease, with none of the genetic or environmental factors that could cause the disease alone (Brown, 2016; Compston and Coles, 2008; Olsson et al., 2017).

112

Genetic susceptibilities

MS is not a hereditary disease. For instance, MS patient monozygotic twins do not necessarily develop the disease. Nonetheless, offsprings and siblings, especially monozygotic twins, of these patients are at higher risks, compared with the general population (Hemminki et al., 2009; Wekerle, 2015). However, this is not due to one particular gene but rather linked with several genes, a majority of which being related to immune functions (Axisa and Hafler, 2016; Inshaw et al., 2018). Genome-wide association studies (GWAS) have lately revealed many variants associated with increased risks to develop MS. Major risk alleles are linked with MHC-II molecules, especially HLA-DRB1*1501 (Dendrou et al., 2018; International Multiple Sclerosis Genetics et al., 2007; International Multiple Sclerosis Genetics et al., 2011; Jersild et al., 1972; Naito et al., 1972). In addition, many single nucleotide polymorphisms (SNPs) have been assigned to genes involved in various immune pathways such as adhesion molecules, cytokine production, central tolerance mechanisms, and T cell homeostasis, activation, proliferation and differentiation (Dendrou et al., 2015; International Multiple Sclerosis Genetics et al., 2013; International Multiple Sclerosis Genetics et al., 2011).

Environmental factors As above-mentioned, the incidence of MS varies greatly depending on world regions (Ascherio and Munger, 2007; Dendrou et al., 2015; Milo and Kahana, 2010). Immigration studies have shown that people emigrating from high to low incidence areas had decreased risks to develop MS, while the ones migrating in the opposite direction had the same risks as in their country of origin (Dean and Elian, 1997; Ebers, 2008; Gale and Martyn, 1995; Milo and Kahana, 2010). These changes in the risk of developing MS were dependent on the ethnicity and on the age at immigration; exposure to environmental factors, such as infectious agents, before a certain age could be implicated (Dean and Elian, 1997; Ebers, 2008; Gale and Martyn, 1995; Milo and Kahana, 2010). This idea is linked with the hygiene hypothesis, which postulates that the reduced frequency of infections observed in the developed world is implicated in the increased frequency of autoimmune and allergic diseases (Ascherio and Munger, 2007; Bach, 2018; Gale and Martyn, 1995). Two environmental factors are, in particular, thought to play a major role in MS pathogenesis: low vitamin D levels and Epstein-Barr virus (EBV) infection (Brown, 2016; Milo and Kahana, 2010). Other factors such as cigarette smoking, obesity and low melatonin levels, as well as other viruses have been suggested to be implicated in the pathophysiology of MS, although their roles will not be detailed here (Bar-Or, 2016; Leibovitch et al., 2018; Olsson et al., 2017).

113

Epstein-Barr virus The role of infection by EBV, a herpes virus that is causing mononucleosis, has not been fully elucidated, but increasing evidence suggests it would be a causative agent of MS (Brown, 2016; Geginat et al., 2017). Most of MS patients are infected by EBV, and have higher titers of antibodies directed towards EBV compared with infected individuals from the general population (Ascherio and Munger, 2007; Haahr and Hollsberg, 2006; Pender and Burrows, 2014). Of note, HLA-DRB1*1501 individuals with an infectious mononucleosis had higher risks to develop MS, these two factors having synergistic effects (Disanto et al., 2013). Different theories have been formulated regarding the impact of EBV infection on MS prevalence, including: molecular mimicry, i.e. T cell clone recognizing both EBV and myelin antigens, leading to the activation of myelin antigen-specific T cells following EBV infection; breakdown of tolerance and subsequent bystander activation of autoreactive T cells; and infection of B cells by EBV (B cells are the primary target of EBV), B cells being involved in the pathogenesis of MS, the role of which will be briefly described later in this section (Brown, 2016; Geginat et al., 2017; Lang et al., 2002; Li et al., 2018; Pender, 2003; Wucherpfennig and Strominger, 1995).

Vitamin D The fact that MS incidence increases with the latitude could be due to differences in sunlight intensity (Milo and Kahana, 2010). Indeed, correlations have been found between a lack of sunlight exposure and increased risks to develop MS (Alonso et al., 2011; Dalmay et al., 2010; Goldacre et al., 2004). A possible explanation for this latitude gradient of MS incidence might be the effects of ultra violet (UV) on vitamin D release, UV converting the precursor of Vitamin D into its active form. MS patient sera contain lower vitamin D levels, and low vitamin D serum levels correlate with higher risks to develop MS. However, reduced serum vitamin D levels in MS patients might not only be due to a lack of sunlight exposure, but also to polymorphisms for genes involved in vitamin D metabolism and diet (Ascherio et al., 2012; Brown, 2016; Shaygannejad et al., 2010; Smolders et al., 2009). Dietary vitamin D uptake has been correlated with decreased risks to develop MS (Munger et al., 2004). Vitamin D can affect immune responses in many different ways and the link between a lack of vitamin D and increased risk of MS development might be due to the induction of immune tolerance by vitamin D, such as the induction of tolerogenic DCs and subsequent Treg generation (Adorini, 2003; Adorini and Penna, 2009; Griffin et al., 2001).

114

IV.1.a.iii. Experimental autoimmune encephalomyelitis model

EAE is a murine model for MS (Baxter, 2007). There are two main ways to induce EAE, i.e. active or passive immunization. Active immunization consists in subcutaneously injecting an emulsion containing complete Freund’s adjuvant (CFA) along with a MHC-II-restricted myelin antigen, such as myelin basic protein (MBP), myelin oligodendrocyte protein (MOG) or myelin proteolipid protein (PLP), depending on the mouse strain used (Baxter, 2007; Rangachari and Kuchroo, 2013). CFA enables the activation of APCs that subsequently prime myelin antigen- specific T cells in LNs draining the site of injection (Baxter, 2007). Pertussis toxin (PTX) is also injected intravenously, although its role has not been fully deciphered. PTX could be involved in the permeabilization of the blood-brain barrier (BBB), therefore facilitating the infiltration of immune cells in the CNS, in the activation of myeloid cells to prime encephalitogenic T cells, and in altering the Treg compartment (Bruckener et al., 2003; Cassan et al., 2006; Dumas et al., 2014; Hou et al., 2003; Linthicum et al., 1982; Ronchi et al., 2016). The distinct mouse strains, which have been reviewed by Rangachari and Kuchroo, are used to model the different forms (relapsing-remitting or progressive) of MS (Rangachari and Kuchroo, 2013). The mouse strain we used in our study, C57BL/6, can lead to a chronic form without remitting phase (mimicking PPMS) or to a RRMS form, depending on the dose of MOG injected, low dose vs high dose of MOG, respectively (Mendel et al., 1995; Rangachari and Kuchroo, 2013). Passive induction of EAE can be achieved by injecting into naïve mice in vitro-primed monoclonal 2D2 cells, CD4+ T cells expressing a MOG-specific TCR or polyclonal T cells from actively-immunized mice that are re-activated ex vivo before their adoptive transfer (Bettelli et al., 2003; Hohlfeld and Steinman, 2017; Peters et al., 2015). Spontaneous and humanized models of EAE also have been developed (Krishnamoorthy et al., 2006; Wekerle et al., 2012).

IV.1.a.iv. Immunopathophysiology of MS and EAE

It has not been demonstrated whether MS is initiated in the periphery or in the CNS, although the consensus among scientists classifying MS as an autoimmune disease assumes that MS is initiated in the periphery (Dendrou et al., 2015). MS is considered as a CD4+ T cell-mediated autoimmune disease, based on the fact that certain MHC-II alleles, such as HLA-DRB1*1501, are major risk alleles for MS and also since EAE can be induced in mice by injecting MHC-II- restricted myelin antigens (Dendrou et al., 2015; Dendrou et al., 2018; International Multiple

115

Sclerosis Genetics et al., 2007; International Multiple Sclerosis Genetics et al., 2011; Rangachari and Kuchroo, 2013). In the EAE model, autoreactive T cells are activated by APCs in the periphery, in SLOs (Becher and Greter, 2012; Dendrou et al., 2015). The idea that autoreactive T cells are primed in the LNs is supported by the efficacy of Fingolimod, used to treat MS, which sequesters T cells in the LNs by blocking their egress (Chun and Hartung, 2010). Encephalitogenic T cells migrate to the CNS and enter the subarachnoid space, which contains CSF, after having crossed the BBB at the choroid plexus (Becher and Greter, 2012; Dendrou et al., 2015). Autoreactive T cells are subsequently re-activated by APCs in the CNS and produce pro-inflammatory cytokines, leading to immune cell recruitment in the parenchyma, such as monocytes, macrophages, neutrophils and other T cells, which leads to the destruction of the myelin sheath (Becher and Greter, 2012; Dendrou et al., 2015). These processes are detailed in the following paragraphs.

Priming of cells in the periphery Role of DCs and other APCs The old dogma stipulates that, after subcutaneous immunization with CFA and myelin antigen for EAE induction, DCs residing at the injection site capture the antigen and migrate to the dLNs, where they prime naïve autoreactive T cells that have escaped central tolerance (Becher and Greter, 2012; Dendrou et al., 2015; Huang et al., 2012). Regarding the role of cDCs, this dogma has been challenged by studies using transgenic mice in which DCs are depleted, such as the CD11c-DTR (diphtheria toxin receptor) mouse model (Becher and Greter, 2012; Birnberg et al., 2008; Isaksson et al., 2012; Ohnmacht et al., 2009; Yogev et al., 2012). Some studies concluded that cDCs were not required for T cell priming or that they played a tolerogenic role in EAE, raising the question whether other APCs primed encephalitogenic T cells in the SLOs (Becher and Greter, 2012; Ohnmacht et al., 2009; Yogev et al., 2012). However, these conflicting results might be partly explained by depletion protocols that were not fully optimized, by distinctive roles of cDCs in SLOs in the periphery and in the CNS, and finally, by a failure to distinguish the functions of different cDC subsets (Becher and Greter, 2012). Indeed, as described in the general introduction, there are several subsets of cDCs, which have distinct roles (Anderson et al., 2018; Dalod et al., 2014; Guilliams et al., 2014; Merad et al., 2013). For example, monocytes could be recruited to the site of immunization and differentiate into inflammatory monocyte-derived (mo)DCs that would subsequently prime encephalitogenic T cells (Becher and Greter, 2012). Indeed, human inflammatory DCs, which share characteristics with in vitro-differentiated moDCs were demonstrated to induce the

116

differentiation of Th17 cells, a T cell subset known to play a pathogenic role in MS (Segura et al., 2013b). Dermal CD103+ DCs or macrophages could also prime encephalitogenic T cells (Becher and Greter, 2012). Macrophages might also be involved in the priming, while other cDC subsets play an anti-inflammatory role (Becher and Greter, 2012; Marta et al., 2008; Ohnmacht et al., 2009; Yogev et al., 2012). Regarding pDCs in EAE, the role of IFN-I production is controversial, as this cytokine can have anti- or pro-inflammatory effects (Guery and Hugues, 2013; Isaksson et al., 2009; Kasper and Reder, 2014; Loschko et al., 2011a). Inded, EAE severity is increased in IFNAR-/- mice, and IFN- β, the first-line treatment of MS, delays relapses and dampens the severity of the disease in MS patients, while pDC depletion and IFN-I neutralization have been shown to ameliorate early phase of EAE (Guery and Hugues, 2013; Isaksson et al., 2009; Kasper and Reder, 2014). Regarding APC functions of pDCs, we have demonstrated that pDCs, as MHC-II-restricted APCs, have a tolerogenic role in EAE and promote Treg generation (Duraes et al., 2016; Guery and Hugues, 2013; Irla et al., 2010; Lippens et al., 2016). Furthermore, antigen-targeting to pDCs via Siglec-H dampens EAE severity (Loschko et al., 2011a).

Other APCs, beside monocytes, macrophages and DCs, could be involved in the priming of encephalitogenic T cells, such as B cells (Li et al., 2018). B cell depletion ameliorates EAE independently of antibody production, suggesting that their APC functions may play a pathogenic role (Fillatreau and Anderton, 2007). For instance, it was recently demonstrated that memory B cells from the at-risk allele HLA-DR15 MS patients present antigens to autoreactive Th1 cells, inducing their proliferation (Jelcic et al., 2018). Moreover, it was recently shown that B cells were the major APCs implicated in naïve CD4+ T cell activation after immunization with a monoparticle antigen derived from a virus, showing that DCs are not necessarily the primary initiator of CD4+ T cell responses (Hong et al., 2018).

Th1/Th17 Th1 and Th17 cells are strongly implicated in the immunopathophysiology of EAE. The frequencies of myelin-antigen specific CD4+ T cells are similar in MS patients and healthy donors, however, they are functionally different, with increased production of IFN-γ, IL-17 and GM-CSF in MS patients (Cao et al., 2015). However, EAE severity in IFN-γ-/- and IL-17-/- mice was increased or similar, compared with control mice, respectively, showing that IFN-γ and IL-17 are not crucial cytokines for EAE induction (Ferber et al., 1996; Haak et al., 2009). On the contrary, IL-1β and IL-23, which drive

117

Th17 differentiation, are essential cytokines, without which mice are resistant to EAE (Cua et al., 2003; El-Behi et al., 2011; Langrish et al., 2005). Using GM-CSF-/- mice, it has been demonstrated that this cytokine is essential to induce EAE (McQualter et al., 2001). In addition, Th17 cells expansion and GM-CSF expression by T cells are increased in MS patients, while these phenomena are reduced by IFN-β therapy (Durelli et al., 2009; Rasouli et al., 2015). In EAE, encephalitiogenic or non-encephalitogenic Th17 can be distinguished based on their production of GM-CSF (Codarri et al., 2011; El-Behi et al., 2011). Moreover, the production of GM-CSF by IL-17-/- IFN-γ-/- CD4+ T cells is sufficient to induce EAE (Codarri et al., 2011). Although Th17 cells play an important role, it does not exclude a role for Th1 cells, which had long been thought to be the main player in EAE, before the discovery of Th17 cells (Bettelli et al., 2004; Lovett- Racke et al., 2004). Indeed, similarly to Th17 cells, the adoptive transfer of Th1 cells can induce EAE (Jager et al., 2009). Nonetheless, Th1 and Th17 do not use the same mechanisms to infiltrate the CNS (Stromnes et al., 2008). Th1 cells, via the production of TNF-α and IFN-γ, induce the expression of adhesion molecules, such as vascular cell adhesion molecule 1 (VCAM- 1), facilitating the transmigration of these cells into the CNS, preferentially the spinal cord (Stromnes et al., 2008). Th17 cells, by their production of IL-17, enhance the permeabilization of the BBB and preferentially infiltrate the brain (Huppert et al., 2010; Stromnes et al., 2008).

Tregs The Treg compartment has been shown to be altered in MS patients; tTregs isolated from MS patient peripheral blood exhibit impaired suppressive abilities (Viglietta et al., 2004). It has been suggested that MHC-II risk alleles, such as HLA-DR15, might be implicated in dysfunctional negative selection in the thymus (Dendrou et al., 2018). It has been recently shown, in patients suffering from Goodpasture disease, a CD4+ T cell-mediated autoimmune disease in which HLA- DR15 is also a risk allele, that HLA polymorphism shapes the abundance or lack of self-specific Tregs, leading to protection or susceptibility to the disease (Ooi et al., 2017). The same conclusions were drawn using HLA-DR15 transgenic mice in a model of Goodpasture disease (Ooi et al., 2017). It is possible that this discovery also applies for MS. Other phenomenon such as Treg instability and Treg plasticity could be implicated in MS pathogenesis (Dominguez-Villar and Hafler, 2018). It has been demonstrated, in mice, that the adoptive transfer or depletion of Tregs before EAE induction decreases or exacerbates EAE severity, respectively (Irla et al., 2010; Lippens et al., 2016; McGeachy et al., 2005).

118

Immune cell migration into the CNS The entry of T cells in the CNS is believed to occur in two phases (Engelhardt and Ransohoff, 2012). A first wave of Th17 cells entry, for which the expression of CCR6 is important to reach the subarachnoid space (Reboldi et al., 2009). CCR6 ligand, CCL20, is expressed by choroid plexus epithelial cells. This is followed by a second wave of immune cell infiltration, including Th1 cells, which enter the CNS parenchyma, independently of CCR6. The infiltration of Th1 rather depends on the expression of CXC-chemokine receptor (CXCR)3 (Lalor and Segal, 2013). The entry of Th1 and Th17 cells in the spinal cord, as well as the entry of Th1 in the brain, is dependent on the interaction of the integrin very late antigen 4 (VLA-4) with VCAM-1, expressed by BBB endothelial cells (Baron et al., 1993; Glatigny et al., 2011). Of note, Natalizumab is a monoclonal antibody used in the treatment of MS that blocks the VLA- 4/VCAM-1 interaction (Polman et al., 2006; Schwab et al., 2015; Shirani and Stuve, 2017). On the contrary, the entry of Th17 cells in the brain does not depend on VLA-4 but on the expression of lymphocyte function-associated antigen 1 (LFA-1) (Glatigny et al., 2011; Rothhammer et al., 2011). Moreover, it has recently been demonstrated, using a lewis rat EAE model, that the leptomeninges act as a checkpoint from where activated myelin-specific T cells can enter the CNS parenchyma, while non-activated T cells are released in the CSF (Bartholomaus et al., 2009; Schlager et al., 2016). The attachment of activated T cells to the leptomeninges is due to the interaction of VLA-4 and LFA-1 with their ligands expressed by resident macrophages, CXCR5/CXCR3 signaling and antigen presentation by macrophages, while non-activated T cells are flushed by the CSF (Schlager et al., 2016). In addition to these well-established routes for T cell entry into the CNS, a putative role of meningeal LVs could also be implicated. As mentioned previously, the recent (re)-discovery of LVs in rodent CNS, also visualized by MRI in human and non-human primate, has induced a reassessment of the immune privilege dogma, in which the BBB and blood-leptomeningeal barrier (BLMB), at the surface between the brain and spinal cord, separating the blood and the CSF, were thought to prevent the entry of immune cells into the CNS (Absinta et al., 2017; Aspelund et al., 2015; Engelhardt et al., 2017; Louveau et al., 2015a; Louveau et al., 2017; Louveau et al., 2015b; Mezey and Palkovits, 2015). These finding are of major importance for MS research. Indeed, meningeal LVs, which link the CNS and deep cervical LNs, might be involved in CNS inflammatory processes that take place in MS (Meyer et al., 2017).

119

Furthermore, endothelial cells of the BBB are strategically positioned and are implicated in the control of adaptive immune responses in the CNS, especially for the entry of T cells in the CNS (Galea et al., 2007; Meyer et al., 2017).

Chronic CNS inflammation and damages Reactivation of T cells in the CNS Autoreactive T cells that infiltrate the CNS are re-activated by APCs, the nature of which is widely debated (Dendrou et al., 2015). In the steady-state, several types of myeloid cells co-exist in the CNS, including microglia and different subsets of macrophages and DCs (Mrdjen et al., 2018). T cell reactivation might occur in the cervical LNs, in the CNS parenchyma by microglia or astrocytes, or in the perivascular and meningeal spaces by macrophages or DCs (Dendrou et al., 2015). The role of antigen presentation by astrocytes and microglia in MS and EAE is a matter of debate and will not be developed here (Colombo and Farina, 2016; Goldmann and Prinz, 2013; Ponath et al., 2018). Conventional DCs have been recently demonstrated to be of crucial importance for the reactivation of self-reactive T cells in the CNS (Giles et al., 2018).

Chronic CNS inflammation In MS, pro-inflammatory cytokines are secreted by leukocytes that have infiltrated the CNS (Becher et al., 2017). As mentioned earlier, GM-CSF produced by re-activated T cells play a detrimental role in MS and EAE (Croxford et al., 2015b). It has been shown to be implicated in the recruitment of myeloid cells in the CNS (Spath et al., 2017). Among these cells, monocytes and CD103+ DCs are recruited (Croxford et al., 2015a; King et al., 2009; King et al., 2010). IL-17 has been found in CNS lesions and in the CSF of MS patients (Matusevicius et al., 1999). IL-17 is involved in the induction of cytokine and chemokine secretion, as well as the production of reactive oxygen species (ROS) due to microglial activation, which leads to demyelination and axonal loss (Gilgun-Sherki et al., 2004; Kawanokuchi et al., 2008). IL-17 was also implicated in the formation of tertiary lymphoid structures (TLS) (Becher, 2015; Pikor et al., 2015). Myeloid cells recruited to the CNS produce IL-1β, which is critical for EAE induction via the differentiation of Th17 cells (Levesque et al., 2016). This cytokine can be produced by Th17 themselves, inducing an autocrine loop of Th17 cell maintenance (El-Behi et al., 2011; Martin et al., 2016). Altogether, these inflammatory processes result in white and grey matter damages (Dendrou et al., 2015).

120

Tregs in the CNS During EAE, Tregs infiltrate the CNS and dampen the inflammation (Kohm et al., 2002). In addition, in EAE mice, remitting phases are associated with an accumulation of IL-10-producing Tregs within the CNS, although Tregs in the inflamed CNS environment were shown to have impaired suppressive capacity (Korn et al., 2007; Zhang et al., 2004). Finally, neurons (MHC-II- negative cells) were demonstrated to induce the conversion of encephalitigenic CD4+ T cells into Tregs via the production of TGF-β (Liu et al., 2006).

Beyond CD4+ T cells Although MS is considered as a CD4+ T cell-mediated disease, CD8+ T cells and B cells also have been shown to be implicated in the immunopathophysiology of MS and EAE (Li et al., 2018; Rangachari et al., 2017). However, their role will not be developed here. Briefly, B cells are involved in the production of auto-antibodies that target the myelin sheath, and are APCs, with paradoxical effects, as the outcome can be either encephalitogenic T cell priming/re-activation or Treg generation (Fillatreau and Anderton, 2007; Jelcic et al., 2018; Li et al., 2018). Nonetheless, a rationale for the pathogenic role of B cells in MS is that the use of monoclonal antibodies depleting B cells induces a delay in relapses in RRMS patients (Gasperini et al., 2013; Hauser et al., 2008; Sorensen and Blinkenberg, 2016). Although CD8+ T cells have a higher frequency than CD4+ T cells in MS lesions, their role has been underestimated and is still debated, with studies concluding for a pathogenic role while others for a regulatory role in MS (Salou et al., 2015; Sinha et al., 2015)

IV.1.a.v. MS therapies targeting the immune system

There are several existing disease-modifying therapies, with immunomodulatory or immunosuppressive properties (Diebold and Derfuss, 2016; Fyfe, 2016). However, these treatments allow the reduction of relapses in RRMS patients but are not efficient to treat all patients (Lopez-Diego and Weiner, 2008; Ransohoff et al., 2015).

First-line treatments The first-line treatments in MS are IFN-β and glatiramer acetate (GA). IFN-β has pleiotropic effects on the immune system, it affects T cells, B cells and APCs, shifts the cytokine network due to anti-inflammatory effects and limits immune cell trafficking across the BBB (Kasper and Reder, 2014). GA is a polymer that is analog to a MBP epitope, which leads to a competition between GA and MBP for the binding to MHC molecules and for presentation to T cells

121

(Neuhaus et al., 2000; Ziemssen and Schrempf, 2007). Treatment with GA induces a shift from Th1 to Th2 cell response.

Therapies blocking cell migration Fingolimod (FTY720) induces the sequestration of lymphocytes in the LNs, the cells are therefore unable to migrate from the periphery towards the CNS. This drug is an antagonist of S1P1R and blocks the egress of lymphocytes, which is dependent on the S1P/S1P1R interaction (Chun and Hartung, 2010; Kappos et al., 2006; Lublin et al., 2016). Natalizumab is a monoclonal antibody targeting VLA-4, therefore blocking the transmigration of immune cells across the BBB and their entry into the CNS (Polman et al., 2006; Schwab et al., 2015; Shirani and Stuve, 2017).

Therapies targeting T and B cell functions Teriflunomide and Mitoxantrone are cytotoxic agents that, by using different mechanisms, suppress the proliferation of T and B cells (Bar-Or et al., 2014; Fox, 2004; O'Connor et al., 2011). Daclizumab is a monoclonal antibody targeting IL-2Rα, also known as CD25, expressed by T cells, leading to a decreased secretion of GM-CSF, a molecule that has a crucial role in the pathogenesis of EAE, as mentioned previously (Hartmann et al., 2014; Wynn et al., 2010).

Antigen-specific therapies These treatments consist in nanoparticles covered by myelin-related peptides and linked to MHC- II. This allows the presentation of self-peptides in a tolerogenic fashion, converting encephalitogenic T cells into Tregs (Clemente-Casares et al., 2016; Hohlfeld et al., 2016).

Therapies targeting B cell functions Rituximab, Ocrelizumab and Ofatumumab are monoclonal antibodies targeting CD20, inducing the depletion of B cells, and leading to a reduction of relapses in RRMS patients and a decrease in active lesions (Gasperini et al., 2013; Hauser et al., 2008; Sorensen and Blinkenberg, 2016)

Therapies targeting multiple pathways Laquinimod targets multiple immune pathways. Among others, it is reducing the infiltration of T cells into the CNS and also inhibits MHC-II-mediated antigen presentation (Bruck and Wegner, 2011). Alemtuzumab is a monoclonal antibody targeting CD52, expressed at the surface of T and B cells, NK cells and macrophages, leading to the cytolysis of these cells (Hartung et al., 2015).

122

Improving our understanding of immune pathways for which the disruption may lead to or aggravate EAE/MS may pave the way for the discovery of new treatments, and to more accurate diagnoses.

IV.1.b. Role of self-antigen presentation by LECs in autoimmunity and peripheral T cell tolerance

As mentioned in the general introduction, LECs can modulate T cell responses indirectly, by impacting different mechanisms, as well as directly, by presenting exogenously-acquired antigens to CD8+ and CD4+ T cells, leading to their dysfunction. For instance, we previously showed that they can acquire peptide/MHC-II complexes from DCs, subsequently inducing cell death and anergy of CD4+ T cells (Fig. 18) (Dubrot et al., 2014). In this section, we will describe how LECs also affect T cell responses by presenting endogenously-express self-antigens, and how this phenomenon is involved in peripheral T cell tolerance and autoimmunity. We have discussed our current knowledge on this topic in a review [Appendix 1 (Humbert et al., 2016)].

IV.1.b.i. Endogenous-expression of peripheral tissue-restricted antigens by LECs

The discovery of mTEC ability to ectopically-express peripheral tissue-restricted antigens (PTAs) and present them to T cells was the first example of endogenously-expressed self-antigens presentation to T cells by non-hematopoietic cells (Derbinski et al., 2001; Kyewski et al., 2000). Intestinal fatty acid-binding protein (iFABP)-tOVA (truncated ovalbumin) and glial fibrillary acidic protein (GFAP)-HA (hemagglutinin) are transgenic mouse models in which tOVA or HA are expressed as self-antigens in mature intestinal epithelial cells (IECs) or enteric glial cells (EGCs), respectively. In those mice, the expression of EGC-associated HA or IEC-associated tOVA proteins have been unexpectedly observed in CD45-negative stromal cells, in addition to EGCs or IECs, and was not restricted to draining (mesenteric) LNs but was found in all LNs (Lee et al., 2007; Magnusson et al., 2008). LNSCs (CD45-negative) had the ability to process endogenously-expressed HA or tOVA into antigenic peptides, and to subsequently present them in SLOs to CD8+ T cells. This function of LNSCs can therefore be considered as peripheral counterpart of mTECs in the thymus, i.e. presenting endogenously-expressed PTAs to SP thymocytes during the process of negative selection (Collier et al., 2008; Hirosue and Dubrot, 2015; Lee et al., 2007; Magnusson et al., 2008). In addition, the expression of PTAs and their

123

direct presentation to CD8+ T cells by LNSCs have been described in non-transgenic mouse models. It was observed that each subtype possesses a distinct but partially redundant PTA expression pattern. Indeed, some PTAs are expressed exclusively by one LNSC subtypes, while other PTAs are expressed in a redundant manner (Cohen et al., 2010; Fletcher et al., 2010) (Fig. 18). For example, LECs are the only LNSC subset that ectopically expresses tyrosinase (Tyr), an antigen whose expression is normally restricted to melanocytes (Cohen et al., 2010; Fletcher et al., 2010; Nichols et al., 2007). The fact that some PTAs are expressed specifically by one LNSC subset suggests a non- redundant role for the different LNSC subsets regarding the tolerization of a variety of self- specific T cells. Moreover, PTA expression by LECs is compartmentalized subanatomically; PTAs are only highly expressed in LN medullary sinus LECs (Cohen et al., 2014). In mTECs, the expression of the vast majority of PTAs is regulated by the transcription factor Aire (Anderson et al., 2002; Derbinski et al., 2005). In the LNs, extrathymic Aire-expressing cells (eTACs), a rare bone-marrow-derived population, have been described to express Aire and named after it. These cells express various PTAs in an Aire-dependent manner (which might therefore be redundant with PTAs expressed by mTECs in the thymus) and present them directly through MHC-I and MHC-II molecules, leading to CD4+ T cell anergy and CD8+ T cell deletion, respectively (Fig. 18) (Gardner et al., 2008; Gardner et al., 2013). On the other hand, PTAs that are expressed by LNSCs, cells of non-hematopoietic origin, do not depend on Aire but on other transcription factors (Cohen et al., 2010). For example, the regulation of pancreatic polypeptide (Ppy) expression, a pancreatic self-antigen, by LECs in pancreatic LNs is dependent on deformed epidermal autoregulatory factor 1 (DEAF-1), which belongs to the SAND (named after Sp100, AIRE-1, NucP41/75, DEAF-1) gene family, together with Aire (Gibson et al., 1998; Yip et al., 2009). Of note, Deaf1 variant isoforms found in human and in mice display an impaired expression of Ppy and have been linked with autoimmune type I diabetes; decreased DEAF1 function and subsequent loss of eukaryotic translation initiation factor 4 gamma 3 (Eif4g3) expression indeed affects PTA expression (Yip et al., 2013; Yip and Fathman, 2014; Yip et al., 2009). The low overlapping between PTA expression in LNSCs and mTECs may be explained by the fact that Aire is not expressed by LNSCs (Metzger and Anderson, 2011). This, therefore, suggests a complementary contribution of mTECs and LNSCs in inducing and maintaining T cell tolerance. Future work is needed to characterize other transcription factors, commonly or exclusively expressed by the different LNSC subtypes, and promoting a non-redundant expression of PTAs in these cells.

124

Figure 18. Role of antigen-presentation by lymph node stromal cells in peripheral T cell tolerance. Following negative selection in the thymus, SP thymocytes that successfully passed the selection process, along with thymus-derived regulatory T cells (tTregs), exit from the thymus and reach the periphery. Some self-reactive T cells manage to escape the mechanism of central tolerance and also enter the periphery. Hence, self antigen-specific T cell tolerance also needs to be maintained in the periphery. Exogenous antigens are acquired from peripheral tissues (yellow) by conventional dendritic cells (cDCs) and plasmacytoid DCs (pDCs) that subsequently migrate to the LNs where they present self-antigenic peptides to autoreactive T cells. The antigens expressed by lymphatic endothelial cells (LECs) can also be acquired by cDCs. LECs, blood endothelial cells (BECs) and fibroblastic reticular cells (FRCs) present endogenously-expressed peripheral tissue-restricted antigens (PTAs) (pink) as well as antigenic peptide/major histocompatibility complex (MHC) complexes acquired from cDCs. Therefore, LNSCs contribute to peripheral T cell tolerance through various mechanisms. Extrathymic Aire-expressing cells (eTACs) also present endogenously- expressed PTAs to T cells. Antigen transfers and cell migration are depicted in dashed and dotted arrows, respectively. The outcomes regarding antigen presentation by cDCs, pDCs, LNSCs and eTACs on CD4+ and CD8+ T cell responses are illustrated in the figure. References regarding the contributions of LNSCs to T cell responses are depicted: 1. (Fletcher et al., 2010); 2. (Cohen et al., 2010); 3. (Baptista et al., 2014); 4. (Gardner et al., 2008); 5. (Gardner et al., 2013); 6. (Rouhani et al., 2015); 7. (Dubrot et al., 2014). Ags, Antigens; exo Ags, exogenous antigens; PTAs, peripheral tissue-restricted antigens; pTreg, peripherally-induced Treg; thym. cDC, thymus-resident cDC; tTregs, thymus-derived regulatory T cells. Adapted from Humbert, Hugues* and Dubrot*, Front Immunol, 2016 [Appendix 1 (Humbert et al., 2016)].

125

IV.1.b.ii. Impact of LEC presentation of endogenously-expressed PTAs on T cell responses

Not only LNSCs endogenously express PTAs, which can be acquired by DCs from LECs (Rouhani et al., 2015), but LNSCs have the ability to process these antigens and directly present the PTA-derived peptides to CD8+ T cells, in the context of MHC-I molecules, leading to their elimination by clonal deletion and to tolerance induction (Fig. 18 and 19) (Lee et al., 2007; Magnusson et al., 2008; Nichols et al., 2007). In the iFABP-tOVA and GFAP-HA transgenic models mentioned previously, the lack of presentation of tOVA or HA to tOVA- or HA-specific CD8+ T cells by enteric stromal cells was linked with enteric autoimmunity (Lee et al., 2007; Magnusson et al., 2008). LECs, among the other LNSC subtypes, are involved in CD8+ T cell clonal deletion and are necessary and sufficient for peripheral tolerance to certain self-antigens, such as Tyr, which has been associated with autoimmune vitiligo, showing a major role for LECs in peripheral tolerance maintenance (Cohen et al., 2010; Fletcher et al., 2010; Fletcher et al., 2011; Yip et al., 2009).

Figure 19. Antigen presentation-dependent role of lymph node stromal cells in peripheral T cell responses. Fibroblastic reticular cells (FRCs), lymphatic endothelial cells (LECs) and blood endothelial cells (BECs) express peripheral tissue-restricted antigens (PTAs). It has been demonstrated that FRCs and LECs are able to directly present those antigens to CD8+ T cells, inducing their deletion. Whether LNSCs can directly present endogenously-expressed PTAs to CD4+ T cells and the subsequent outcome on CD4+ T cell responses remain a matter of debate. Adapted from Turley*, Fletcher* and Elpek*, Nat Rev Immunol, 2010 (Turley et al., 2010).

126

The capability of LNSCs to present endogenously-expressed PTAs in a direct manner to CD4+ T cells, in the context of MHC-II molecules, and whether it is inducing CD4+ T cell dysfunction and/or Treg differentiation/maintenance remains under debate (Fig. 18 and 19). We previously demonstrated that CIITA pIV regulates MHC-II molecules endogenous expression in LECs, BECs and FRCs (Dubrot et al., 2014). A study has shown that the adoptive transfer of the transgenic 6.5 CD4+ T cells with a HA-specific TCR in GFAP-HA mice, in which EGCs express HA as an autoantigen, did not dampen enteric autoimmunity (Magnusson et al., 2008). The absence of direct HA peptide presentation by LNSCs to HA-specific CD4+ T cells in this model did nonetheless not rule out a potential MHC-II molecule upregulation by LNSCs upon inflammation, and subsequent presentation of the antigenic peptide (Magnusson et al., 2008). Indeed, as mentioned previously, few studies have suggested that the expression of MHC- II-molecules at the surface of LNSCs is upregulated under pro-inflammatory conditions (Dubrot et al., 2014; Malhotra et al., 2012). On the other hand, using models in which endogenously- expressed PTAs were β-galactosidase (β-gal), membrane-bound HA or I-Eα, Engelhard and colleagues concluded that LECs were not able to present these PTAs to CD4+ T cells (Rouhani et al., 2015). This outcome was not due to antigen localization, but rather due to the absence of H2-M expression in LECs that could hamper peptide loading onto MHC-II molecules, H2-M being the chaperone that allows antigenic peptide loading onto MHC-II molecules. Nonetheless, this study was undertaken in steady state conditions, while the expression of CIITA pIV by LECs, BECs and FRCs is IFN-γ-inducible (Reith et al., 2005). Thus, IFN-γ might be required to induce the upregulation of H2-M, as it is the case for the expression of MHC-II expression. Indeed, H2-M and MHC-II molecule expression are co-regulated by CIITA. In addition, Baptista et al. found H2-M mRNA transcripts in LECs, among other MHC-II-related molecules (Baptista et al., 2014). They also observed an unexpected expression of OVA in LECs in K14mOVA transgenic mice, in which OVA is expressed under the control of the keratin 14 promoter. Moreover, OVA+ LECs had the ability to present OVA peptides, in the context of MHC-II molecules, to OVA-specific OT-II cells in vitro, which was crucial for the maintenance of Foxp3+ OT-II Tregs (Baptista et al., 2014). Baptista and colleagues, by performing LN transplantation experiments, further suggested that endogenously-expressed self-antigen presentation by LNSCs could contribute to Foxp3+ CD4+ Treg maintenance, in vivo (Fig. 18) (Baptista et al., 2014). Moreover, another team observed that endogenously-expressed PTA presentation by LNSCs led to CD4+ and CD8+ T cell hyporesponsiveness/anergy. They used lentiviral vectors that allow the selective transduction of MHC-II-expressing cells of non-hematopoietic origin with MHC-II-

127

and MHC-I-restricted HY male-derived epitopes in female mice (Cire et al., 2016). Effector CD4+ T cell conversion into CD25+ Foxp3+ pTregs was found increased in Marylin mice, in which CD4+ T cells express a HY-specific transgenic TCR (Cire et al., 2016). However, whether these observations were due to the direct presentation of HY endogenously-expressed by gp38+ LNSCs, i.e. LECs and FRCs, to CD4+ T cells remains to be investigated. The possibility that (non-DC) hematopoietic cells might contribute to HY antigen presentation cannot be ruled out, because of unwanted transduction and subsequent direct antigen presentation or due to antigen transfer from (non-DC) hematopoietic cells to stromal cells (Cire et al., 2016; Dubrot et al., 2014). It also cannot be excluded that transduced stromal cells could transfer MHC-II-restricted HY peptide to DCs, which could subsequently present the antigen to CD4+ T cells (Rouhani et al., 2015). Although the direct antigen presentation by gp38+ LNSCs was not fully demonstrated and the contribution of LECs and FRCs could not be distinguished, this study is in agreement with the results found by Baptista et al. (Baptista et al., 2014). Finally, using K14tgpIVKO mice, in which MHC-II expression is abrogated in LNSCs, thanks to CIITA pIV deletion, our group very recently showed that elderly K14tgpIVKO mice present signs of spontaneous systemic autoimmunity (Dubrot et al., in press). This was associated with an impaired Treg compartment, with decreased Treg frequencies and ability to suppress T cell responses, leading to enhanced activation of effector CD8+ and CD4+ T cell in SLOs. It induced a subsequent increase in T cell infiltration in peripheral tissues, compared with K14tg age- matched controls, along with enhanced production of auto-antibodies. In addition, the adoptive transfer of T cells from LNs of elderly K14tgpIVKO mice in Rag2-/- mice, which lack T and B cells, recapitulated the autoimmune phenotype. Moreover, in LNs, the proliferation of Tregs, observed by microscopy, in contact with LECs from WT mice was increased in elderly mice or in IFN-γ-administered mice, due to an enhanced MHC-II expression by LECs in these mice. This effect was abrogated in K14tgpIVKO mice or in Prox-1CreERT2MHCIIfl/fl mice, a model that will be described later in this chapter, in which MHC-II is selectively abrogated in LECs. These results show the importance of endogenous MHC-II expression by LNSCs in T cell tolerance, as it provides a brake in spontaneous autoimmunity observed in elderly mice in absence of LNSC MHC-II expression, and are in agreement with the studies mentioned previously (Baptista et al., 2014; Cire et al., 2016). Of note, this analysis supports a particular role for LECs in maintaining the Treg compartment through MHC-II-restricted antigen presentation.

128

IV.1.b.iii. Molecular pathways implicated in peripheral T cell tolerance mediated by LECs

The molecular pathways implicated in CD8+ T cell clonal deletion by LNSCs have not been fully elucidated. Using the iFABP-tOVA mouse model, in which tOVA is expressed in the intestinal epithelium as a self-antigen, it was demonstrated that PD-1:PD-L1 interaction is required for the induction of CD8+ T cell tolerance, as a breakdown of this pathway could lead to severe enteric autoimmune disorder (Reynoso et al., 2009). Tewalt and colleagues performed an adoptive transfer of Tyr-specific transgenic TCR CD8+ FH cells into bone marrow chimeric mice expressing Tyr, and lacking or not the expression of PD-L1 in either radiosensitive (hematopoietic) or radioresistant (non-hematopoietic) cells. The expression of PD-L1 by the non-hematopoietic compartment specifically allowed the clonal deletion of FH cells (Tewalt et al., 2012). In addition, among the different LNSC subtypes, the highest levels of PD-L1 expression were found in LECs, particularly in LECs localized in the LN medullary sinuses. Moreover, costimulatory molecules are not expressed by LECs. The administration of agonistic antibodies targeting the costimulatory molecule 4-1BB hampered the clonal deletion of FH cells. The absence of 4-1BB co-stimulation by LECs has been suggested to induce the upregulation of PD-1 expression by FH cells. Indeed, the presentation of Tyr by LECs led, in this study, to an increased PD-1 expression by FH cells, an effect that was abrogated upon the administration of agonistic anti-4-1BB antibody. This would subsequently prevent the upregulation of CD25 expression that is necessary for the survival of CD8+ T cells. Indeed, FH cell CD25 expression was upregulated uniquely upon administration of agonistic anti-4-1BB or anti-PD-L1 antibodies, after the presentation of Tyr by LECs (Tewalt et al., 2012). Therefore, in this model, LECs can be accounted for the endogenously-expressed Tyr presentation, which along with the provision of co-inhibitory signal and the absence of costimulation, induces Tyr-specific CD8+ T cell deletion (Tewalt et al., 2012). As anti-lymphotoxin β receptor (Ltβr) antibody administration in mice leads to decreased expression of PD-L1 in LECs, it is likely that PD-L1 expression is regulated by Ltβr in this LNSC subset (Cohen et al., 2014). T cells seemed to prevent PD-L1 expression in LECs, while B cells were necessary for mucosal vascular addressin cell adhesion molecule 1 (MadCAM-1) expression on LEC surface in the medulla, which was itself required for PD-L1 expression (Cohen et al., 2014). These finding were observed using Rag1−/−, CD3ε−/− and μMT−/− mice, respectively lacking T and B cells, only T cells and only B cells. Finally, the authors suggested that MHC-II expression in LECs could be implicated in the induction of CD8+ T cell tolerance towards endogenously-expressed PTAs in LECs by binding to lymphocyte activation gene 3 (LAG-3), a co-inhibitory molecule. Indeed,

129

Rouhani et al. performed an adoptive transfer of Bg1 cells, CD8+ T cells harboring transgenic β- gal-specific TCR, into Prox-1 x βgal mice, in which β-gal is exclusively expressed by LECs. It led to an enhanced Bg1 cell proliferation after the administration of blocking anti-LAG-3 antibody, which was working in synergy with anti-PD-L1 blockade (Rouhani et al., 2015). In our hand, a high PD-L1 expression in LECs is associated with the unique capability to induce CD4+ T cell death following the presentation of peptide-MHC-II complexes acquired from DCs, as compared to other LNSC subtypes (Dubrot et al., 2014). The molecular mechanisms responsible for the induction of tolerance towards endogenously- expressed MHC-II-restricted PTAs that are directly presented to CD4+ T cells by LECs still remain to be deciphered. Nonetheless, they are likely to also involve the expression of PD-L1 in LECs, similarly to the case of CD8+ T cells.

IV.1.c. Specific aim

As above-mentioned, LECs have been shown to be directly involved in CD8+ T cell tolerance. Dysfunctions in the ability of LECs to present endogenously-expressed MHC-I-restricted PTAs to CD8+ T cells have been associated with autoimmunity. Emerging evidence is in favor of a role for LNSCs in CD4+ T cell tolerance to MHC-II-restricted PTAs, although this phenomenon still needs to be fully elucidated and the specific contributions of each LNSC subsets to CD4+ T cell tolerance remain to be characterized. For instance, the abrogation of MHC-II expression in all LNSC subsets (LECs, BECs and FRCs) in murine LNs alter the Treg compartment, leading to increased CD4+ and CD8+ T cell activation and effector functions, which correlates with signs of spontaneous autoimmunity in mice (Dubrot et al., in press). These findings are in agreement with other studies that have been described in the previous section (Baptista et al., 2014; Cire et al., 2016). Of note, LEC MHC-II-restricted antigen-presenting abilities seem to have a particularly important role in promoting the maintenance of Treg populations (Dubrot et al., in press).

As described previously, EAE is a mouse model of multiple sclerosis, a chronic inflammatory autoimmune disease of the CNS. As it is a CD4+ T cell-mediated pathological model, it therefore appeared to be suited for the study of a hypothetical induction of tolerance to encephalitogenic T cells through the presentation of MHC-II-restricted antigenic peptides. Conversely, a better understanding of the immunopathophysiology of EAE may shed light on new targets for the design of therapeutic drugs to treat MS.

130

The aim of this project was to dissect the precise contribution of MHC-II-mediated antigen- presenting functions of LECs to self-reactive CD4+ T cell responses and to determine the impact of this phenomenon on the development of autoimmunity. To this end, we used genetically- engineered mice, in which MHC-II-restricted antigen-presenting functions of LECs were selectively abrogated or reinforced, in the EAE model.

131

IV.2. Results

MHC class II-restricted myelin antigen presentation by lymphatic endothelial cells dampens central nervous system autoimmunity

Marion Humbert1, Camille Kowalski1 and Stéphanie Hugues1

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

Results presented as a manuscript in preparation [Discussion of the results obtained so far, as well as future experiments to be performed in order to address different hypotheses, are provided in the discussion part of Chapter B (section IV.3.)]

Objective: Lymphatic endothelial cells (LECs) affect T cell responses in many ways, including the drainage of macromolecules, fluids, and antigen presenting cells (APCs), from tissues to the lymph nodes (LNs). In addition, LECs function as unconventional major histocompatibility complex class I (MHC-I)-restricted APCs, therefore being directly implicated in the modulation of peripheral T cell responses. For instance, LECs from the LNs present endogenously-expressed peripheral tissue-restricted antigens (PTAs) via MHC-I molecules to CD8+ T cells, which leads to their elimination by clonal deletion. The participation of LECs in peripheral CD4+ T cell tolerance, as MHC-II-restricted APCs presenting PTA-derived peptides, is debated. The aim of this study was to determine whether LECs function as MHC-II-restricted APCs to modulate autoreactive CD4+ T cell responses in experimental autoimmune encephalomyelitis, a mouse model for multiple sclerosis.

Personal contribution: For this research, I wrote the manuscript and I, along with Camille Kowalski, participated in the conception/design of the study, development of the methodology, acquisition and analyzis of the data, all under the supervision of Stéphanie Hugues.

132

MHC class II-restricted myelin antigen presentation by lymphatic endothelial cells dampens central nervous system autoimmunity

Marion Humbert1, Camille Kowalski1 and Stéphanie Hugues1

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

Corresponding author: Stephanie Hugues, Department of Pathology and Immunology, University of Geneva Medical School, Rue Michel-Servet 1, 1211, Geneva, Switzerland. Phone: 41-2-23-79-58-93; Fax: 41-2-23-79-57-46; E-mail: [email protected]

Disclosure: The authors declare no potential conflicts of interest.

- 1 -

Abstract

Beside their role in directing fluid, macromolecules and antigen presenting cells (APCs) from tissues to the lymph nodes (LNs), lymphatic endothelial cells (LECs) can also directly impact peripheral T cell responses as major histocompatibility complex class I (MHC-I)-restricted unconventional APCs. Indeed, LECs located in LNs (LN-LECs) present endogenously expressed peripheral tissue-restricted antigens (PTAs) through MHC-I molecules to CD8+ T cells, inducing their clonal deletion. Whether LECs could participate in peripheral CD4+ T cell tolerance by presenting PTA-derived peptides loaded onto MHC-II is still a matter of debate. Here, we investigated whether LECs contribute as MHC-II-restricted APCs to shape autoreactive CD4+ T cell responses in a mouse model for multiple sclerosis, experimental autoimmune encephalomyelitis (EAE). MHC-II expression in LN-LECs was upregulated during EAE induced inflammation. Moreover, the selective genetic ablation of MHC-II in LECs resulted in disease severity exacerbation. In contrast, the genetic enforcement of the presentation of a myelin-derived peptide through MHC-II in LECs led to delayed onset and dampened EAE severity, accompanied by a drastic decrease in the spinal cord CD4+ T cell infiltrate and a modulation of these cells towards a tolerogenic phenotype. CD4+ T cell numbers and phenotype in the draining LNs were found unaffected, suggesting that this pathway may modulate effector T cell functions rather than the priming of naïve T cells in LNs. Altogether, our results highlight a role for LECs as unconventional tolerogenic APCs in EAE.

- 2 -

Introduction

Lymph node (LNs) stromal cells (LNSCs) are essential to the structure and functions of LNs (1). These cells are from non-hematopoietic origin and include fibroblastic reticular cells (FRCs), blood endothelial cells (BECs) and lymphatic endothelial cells (LECs), and were long thought to function as simple organ scaffolds (1, 2). However, LECs, BECs and FRCs were recently described as strong modulators of immunity, and in particular T cell responses (1-4). LECs compose the lymphatic vessels, which convey lymph fluid, containing proteins and immune cells, drained from interstitial tissues, and help to get rid of toxins and undesired components from the organism (2, 5). T cell responses are indirectly impacted by LECs in many ways: LECs control antigen delivery and dendritic cell (DC) migration to the LNs; regulate DC functions; are implicated in T cell homeostasis; modulate T cell egress from the LNs and their migration through lymphatic vessels (5-8).

LN-LECs also have a direct impact on peripheral T cell responses as unconventional antigen- presenting cells (APCs). They express major histocompatibility complex (MHC) class II (MHC-II) molecules in the presence of IFN-γ (3, 9, 10) (Dubrot et al., in revision). LECs are negative for CD40, CD80 and CD86 co-stimulatory molecules in both steady-state and inflammatory conditions. On the contrary, they express CD48 and herpes virus entry mediator (HVEM), as well as the co-inhibitory molecule programmed cell death 1 ligand 1 (PD-L1) at steady-state, and further up-regulate these molecules upon inflammation (9-12) (Dubrot et al., in revision). LECs can acquire exogenous antigens and present them through MHC-I molecules to induce CD8+ T cell dysfunction (13-15), and also present peptide/MHC-II complexes acquired from DCs to promote CD4+ T cell apoptosis and anergy (9). Interestingly, LECs, BECs and FRCs endogenously express peripheral tissue-restricted antigens (PTAs), each subtype exhibiting a distinct but partially redundant PTA expression pattern (2, 10, 16-19). Moreover, these cells process endogenously-expressed PTAs and present PTA-derived peptides loaded onto MHC-I to CD8+ T cells, leading to their elimination by clonal deletion. Importantly, dysfunctions in these mechanisms have been linked with autoimmunity in mouse models (2, 16, 17, 19). For instance, LECs are necessary and sufficient for peripheral tolerance to certain self-antigens, such as tyrosinase, which has been associated with autoimmune vitiligo, showing a major and direct role for LECs in peripheral tolerance maintenance (10, 18, 20, 21).

- 3 -

The ability of LNSCs to present PTA-derived peptides loaded onto endogenous MHC-II molecules to CD4+ T cells, and its subsequent outcome on T cell responses, is still a matter of debate (2, 7). Endogenously-expressed MHCII-restricted PTA presentation by LNSCs has been suggested to contribute to Treg homeostasis and peripheral T cell tolerance (22, 23). In addition, we recently showed that elderly naïve mice, in which MHC-II expression was abrogated in LNSCs, presented signs of spontaneous systemic autoimmunity, associated with an impaired Treg compartment, and subsequent increase in T cell infiltration in peripheral tissues in vivo (Dubrot et al., in revision). Of note, MHC-II expression by LN-LECs was important for the proliferation of Tregs in situ (Dubrot et al., in revision). LECs do not only impact T cell outcome in LNs. For instance, liver sinusoidal endothelial cells (LSECs) cross-present antigens to CD8+ T cell to maintain tolerance in the liver, using a mechanism involving the expression of PD-L1 (24-26). Furthermore, LSECs present MHC-II- restricted antigens to CD4+ T cells, leading to different outcomes, including Treg differentiation (27, 28). In addition, tumor-associated LECs in skin melanoma present tumor- derived antigens through MHC-I molecules and upregulate PD-L1 to inhibit T cell activation (29). Multiple sclerosis (MS) is a chronic inflammatory autoimmune demyelinating disease of the central nervous system (CNS), mediated by autoreactive CD4+ T cells, which affects more than 2.5 million individuals worldwide (30). A system of lymphatic vessels has been recently identified in rodent and visualized in human and non-human primate meninges, along the perisinusal space (31-34). So far, mainly DCs have been detected in meningeal LECs, but it is possible that T cells use intralymphatic routes to travel in and out the CNS, and recirculate in cervical LNs (35-37). Here, we investigated whether LN-LECs and/or meningeal LECs contribute as MHC-II- restricted APCs to autoreactive CD4+ T cell responses in experimental autoimmune encephalomyelitis (EAE), a murine model for MS (38). We observed an upregulation of LN- LEC MHC-II expression during EAE early phase. In addition, the selective genetic abrogation of MHC-II expression in LECs enhanced disease severity. Using genetically-engineered mice in which the presentation of the MHC-II-restricted myelin-derived MOG35-55 peptide was enforced in LECs, we further showed that EAE onset was delayed and disease severity was attenuated. No effect was observed in the quality and the amplitude of encephalitogenic effector CD4+ T cells (Th1 and Th17) in LNs draining EAE immunization site during T cell priming phase. However, at peak disease, the number of cells infiltrating the spinal cord (SC),

- 4 - as well as the ratios [effector T cells (Th1 or Th17) / Treg], were drastically reduced in mice in which LECs present MOG35-55 through MHC-II molecules. These data suggest that this pathway might be involved in the modulation of effector T cells rather than the priming of naïve T cells in LNs draining EAE immunization sites. Therefore, by presenting myelin- derived peptides on MHC-II, LECs contribute to the tolerization of antigen-specific CD4+ T cells. Altogether these results highlight a role for LECs as unconventional tolerogenic APCs, in impacting peripheral encephalitogenic T cell responses and EAE pathogenesis. Further experiments need to be performed in order to dissect the relative contribution of LECs in LNs and meninges. A better understanding of the implication of LECs as unconventional APCs in autoimmunity may shed light on new targets for the design of therapeutic drugs.

- 5 -

Material and methods

Mice Mice were of pure C57BL/6 background and were bred under specific and opportunistic pathogen-free conditions at Charles River, France. IiMOGfl/fl, Prox1CreERT2 and MHCIIfl/fl have been described elsewhere (39-41). Experiments were performed under specific pathogen-free conditions at Geneva medical school animal facility. All animal husbandry and experiments were approved by and performed in accordance with the guidelines of the animal research committee of Geneva canton, Switzerland. Prox1CreERT2 were crossed with MHCIIfl/fl and with IiMOGfl/fl. Genotyping were performed on ear biopsies by PCR with the Phire Tissue Direct PCR Master Mix (F170L, Thermo Fisher Scientific). We used mice heterozygous or negative for Prox1 gene for experiments. In order to selectively abrogate the expression of MHC-II in LECs or to enforce the presentation of myelin oligodendrocyte peptide MOG35-55 via MHC-II selectively by LECs (Fig. 3), Prox1CreERT2xMHCIIfl/fl and MHCIIfl/fl (control), or Prox1CreERT2xIiMOGfl/fl and IiMOGfl/fl (control), received tamoxifen (Tx). Tx was injected twice a day intraperitoneally (i.p.) for four consecutive days, for a total of 2mg/mouse/day, solubilized in PBS containing ethanol (10%) and castor oil (10%; Kolliphor®; Sigma-Aldrich).

EAE induction

Mice were immunized subcutaneously (s.c.) in both flanks with MOG35-55 peptide (MEVGWYRSPFSRVVHLYRNGK; Biotrend) in PBS emulsified in incomplete Freund's adjuvant (dilution ½; BD Diagnosis) supplemented with Mycobacterium tuberculosis (MT)

H37Ra (BD Diagnosis), for a total of 100 µg of MOG35-55 peptide and 400 µg of MT per mouse. At the time of immunization and two days later, mice were injected intravenously (i.v.) with pertussis toxin (300 ng; Sigma-Aldrich).

EAE monitoring Mice were weighted and clinical symptoms for EAE were evaluated daily and blindly. Scores correspond to the following clinical signs: 1, flaccid tail; 2, impaired righting reflex and hind limb weakness; 3, complete hind limb paralysis; 4, complete hind limb paralysis with partial fore limb paralysis; 5, moribund.

- 6 -

Preparation of single cell suspensions Skin draining LNs (dLNs) and spinal cords (SCs) were isolated from mice at indicated times. For T cell analyses LNs were scratched, and LN cell suspensions were filtered on a 70-µm cell strainer. SCs were mechanically and chemically digested with an enzymatic mix in HBSS medium containing collagenase D (1 mg/mL; Roche) and DNAse I (10 μg/mL; Roche). SC single cell suspensions were centrifuged through a discontinuous 30%:70% percoll gradient (Invitrogen) and further disaggregated using a 70-µm cell strainer. In the case of subsequent intracellular cytokine analysis, LN and SC cells were re-stimulated for 4 h at 37°C and CO2 (5%), in complete RPMI medium [RPMI 1640 GlutaMAX with 10% heat-inactivated FBS, 1% penicillin/streptomycin (10,000 U/mL penicillin and 10 mg streptomycin/mL), 1% sodium pyruvate 100 mmol/L, 0.1% 2-β mercaptoethanol 50 mmol/L] in the presence of phorbol 12- myristate 13-acetate (PMA) (100 ng/ml; Sigma) and ionomycin (1 µg/ml; Sigma). GolgiPlugTM solution (1 µl/ml; BD Biosciences) was added to the culture medium for the last 2.5 h. For LNSC analyses and sorting LNs were digested with an enzymatic mix in RPMI medium containing collagenase P (200 µg/mL; Roche), dispase (800 µg/mL; Gibco) and DNAse I (100 μg/mL; Roche). LN pieces were incubated for 10 min at 37°C, with gently tube shaking and replacement of the supernatant with fresh enzymatic medium once the pieces had settled back at the tube bottom. Suspensions were then subjected to vigorous pipetting. These steps were repeated every 10 min until complete digestion. LNs single cell suspensions were further disaggregated using a 70-µm cell strainer and incubated with anti-CD16/32 antibody (2.4G2) FcgRII-RIII, a Fc receptor monoclonal blocking antibody (Thermo Fisher Scientific). CD45-negative cells were isolated from the LN cell suspensions using CD45 microbeads (negative selection) and magnetic bead column separation (Miltenyi Biotec).

Flow cytometry Anti-TCRβ (H57-597), anti-CD4 (GK1.5), anti-CD25 (PC61.5), anti-IL-17 (eBio17B7) and anti-IFN-γ (XMG1.2) monoclonal antibodies were purchased from Thermo Fisher Scientific. Anti-CD45 (30F11), anti-FOXP3 (MF-14), anti-gp38 (8.1.1), anti-CD31 (390) and anti-PDL1 (10F.962) were purchased from Biolegend, and anti-I-Ad/I-Ed (MHC-II; 2G9) and anti-CD8 (53-6.7) were purchased from BD Biosciences.

- 7 -

Cells were stained with Fixable viability dye eFluorTM 780 (Thermo Fisher Scientific) 30 min, at 4 °C, for cell analyses. Single cell suspensions were incubated with anti-CD16/32. Intracellular stainings (cytokines and Foxp3) were performed using the Intracellular Fixation & Permeabilization buffer set (Thermo Fisher Scientific). Data were acquired using GalliosTM (Beckman Coulter) or LSRFortessaTM (BD Biosciences), and analyzed using FlowJo® Software (FlowJo, LLC). 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 intermediate fluorescence intensity for the viability dye eFluorTM 780 (Thermo Fisher Scientific). LNSC subsets (CD45-negative) were defined as followed: LECs, CD31+ gp38+; BECs, CD31+ gp38-; FRCs, CD31- gp38+ (CD31 is also called PECAM-1 and gp38, podoplanin).

Statistical analyses Statistical significance was assessed by one-way ANOVA tests with Bonferroni post Hoc test or by two-tailed Mann-Whitney test. Two-way ANOVA tests with Bonferroni post Hoc test were performed for the follow up of EAE clinical scores. All analyses were carried out using Prism 7.0 (GraphPad software). *: p<0.05; **: p<0.01; ***: p<0.001; ****: p<0.0001; ns: non-significant.

- 8 -

Results

MHC-II expression is upregulated in LECs during EAE

LNSCs (CD45-negative) can be distinguished based on the expression of gp38 and CD31: FRCs, gp38+ CD31-; LECs, gp38+ CD31+; and BECs, gp38- CD31+) (Figure 1A). LECs have been shown to upregulate surface MHC-II expression in presence of IFN-γ. In order to determine whether LECs expressed MHC-II during the course of EAE, we immunized WT mice. In contrast to FRCs, for which MHC-II remained expressed at low levels, LECs and BECs significantly up-regulated MHC-II molecules in draining LNs compared with naïve mice, as soon as 5 days post-immunization. In both populations, the increase in MHC-II expression reached a peak at day 15 (Figure 1B). At D25, MHC-II levels decreased in LECs and BECs, most likely reflecting a reduced inflammation in draining LNs at this later time-point (Figure 1B). Therefore, MHC-II expression in LECs seems particularly sensitive to the increased inflammation occurring in draining LNs until D15 post-immunization.

Genetic MHC-II abrogation in LECs exacerbates disease severity

After showing that the expression of MHC-II was upregulated in LECs during the course of EAE (Figure 1B), we next hypothesized that LECs could function as APCs involved in shaping self-reactive T cell responses during the course of EAE. To test this, we generated mice in which MHC-II expression is abrogated in LECs. We crossed transgenic mice expressing the tamoxifen (Tx)-inducible CreERT2 recombinase under the promoter Prox-1, which is a transcription factor expressed by LECs, Prox1CreERT2+/- mice, with MHCIIfl/fl mice. Prox1CreERT2+/- x MHCIIfl/fl mice, called MHCII-/- LEC mice herein, and MHCIIfl/fl control mice were treated with tamoxifen (Tx) to induce the deletion, and were injected with IFN- to enhance MHC-II expression in LNSCs. As expected, MHC-II expression was upregulated in LECs, BECs and FRCs in LNs draining the site of IFN-γ injection in control mice. In contrast, MHC-II expression was selectively abrogated in LECs, but not in BECs and FRCs, in MHCII-/- LEC mice, thus validating our model (Figure 2A). 4 weeks after Tx treatment, control and MHCII-/- LEC mice were immunized to induce EAE. We observed an exacerbation of the disease severity in MHCII-/- LEC mice, compared with control mice (Figure 2B). The expression of Cre recombinase alone in LECs had no

- 9 - significant impact on the development of EAE, as observed using Prox1CreERT2+/- mice (not shown). Altogether these results suggest a tolerogenic role for MHC-II-restricted antigen-presentation by LECs in EAE.

MOG35-55 presentation by LECs modulates the T cell infiltrate towards a tolerogenic phenotype in the spinal cord and dampens EAE severity

In order to enforce the presentation of MOG35-55, used as a model antigen in EAE, through MHC- II by LECs, we crossed IiMOGfl/fl mice with Prox1CreERT2+/- mice, inducing the artificial presentation of MOG35-55 through endogenous MHC-II by LECs upon Tx treatment (Figure 3A-

C). IiMOG mice express a mutant invariant chain (Ii) locus containing the peptide MOG35-55, expressed as a replacement peptide for class II-associated Ii peptide (CLIP), under the control of the Rosa26 promoter, and downstream of a loxP flanked transcriptional STOP cassette (Figure 3A ERT2+/- fl/fl and 3B). Prox1Cre x IiMOG are called “MOG35-55-LEC mice” herein (Figure 3C). In WT mice, CLIP hides the MHC-II-binding groove after Ii proteolysis and can be exchanged with an antigenic peptide by H2-M (Figure 3A). In MOG35-55-LEC mice, MOG35-55 from the mutant IiMOG invariant chain is not exchanged with another peptide and is directly presented at the surface of LECs via MHCI-II, following Tx injection (Figure 3A and 3C). fl/fl 4 weeks after Tx treatment, MOG35-55-LEC mice and IiMOG control mice were immunized to induce EAE. Disease was significantly delayed and attenuated, and body weight loss and survival were significantly improved in MOG35-55-LEC mice compared to control mice (Figure 4A-C).

This suggests a tolerogenic role for the MHC-II-restricted presentation of MOG35-55 by LECs in the context of EAE. Correlating with EAE attenuation, the number of CD45+ cells (not shown) and the number of CD4+ T cell infiltrating the spinal cord (SC) at peak disease (D17 post-immunization) were dramatically reduced in MOG35-55-LEC mice compared with control mice (Figure 5A). Strikingly, Treg frequency was increased whereas the frequency of IFN-γ-producing CD4+ T cells was reduced in the SC of MOG35-55-LEC mice (Figure 5B and 5C). In contrast, the frequency of IL- 17-producing CD4+ T cells was not significantly affected (Figure 5C). As a result, the balance encephalitogenic T cells / Treg in the CNS was significantly shifted towards a suppressive phenotype in mice in which LECs presented MOG35-55 through MHC-II (Figure 5D). Despite these clear alterations in immune cells infiltrating the SC, the frequencies and absolute numbers of Tregs, IFN-γ+ and IL-17+ CD4+ T cells were not affected in the skin LNs draining the

- 10 - site of immunization, suggesting that LECs other than the ones located in skin dLNs may be implicated (Figure 5E and 5F). + Therefore, the presentation of MOG35-55 by LECs modulates encephalitogenic CD4 T cell responses in the CNS and delays EAE onset, as well as dampens its severity.

Conclusion

Altogether, our results reinforce the idea that MHCII-restricted antigen presentation by LECs has a tolerogenic effect on self-reactive T cell responses and autoimmune disease progression.

- 11 -

Acknowledgements

The authors thank J.P. Aubry-Lachainaye, C. Gameiro and G. Schneiter for excellent assistance in flow cytometry, D. Brighouse for help with experiments, and A. Waisman, and T. Mäkinen for providing IiMOGfl/fl and Prox-1-CreERT2, respectively. This work was supported by the Swiss National Science Foundation (PP00P3_152951 and 310030_166541) and by the Swiss Multiple Sclerosis Society.

- 12 -

References

1. D'Rozario J, Roberts D, Suliman M, Knoblich K, Fletcher A. Leukocyte-Stromal Interactions Within Lymph Nodes. Advances in experimental medicine and biology. 2018;1060:1-22. PubMed PMID: 30155619.

2. Turley SJ, Fletcher AL, Elpek KG. The stromal and haematopoietic antigen-presenting cells that reside in secondary lymphoid organs. Nature reviews Immunology. 2010 Dec;10(12):813-25. PubMed PMID: 21088682.

3. Malhotra D, Fletcher AL, Astarita J, Lukacs-Kornek V, Tayalia P, Gonzalez SF, et al. Transcriptional profiling of stroma from inflamed and resting lymph nodes defines immunological hallmarks. Nature immunology. 2012 Apr 1;13(5):499-510. PubMed PMID: 22466668. Pubmed Central PMCID: 3366863.

4. Mueller SN, Ahmed R. Lymphoid stroma in the initiation and control of immune responses. Immunological reviews. 2008 Aug;224:284-94. PubMed PMID: 18759934.

5. Randolph GJ, Ivanov S, Zinselmeyer BH, Scallan JP. The Lymphatic System: Integral Roles in Immunity. Annual review of immunology. 2017 Apr 26;35:31-52. PubMed PMID: 27860528. Pubmed Central PMCID: 5551392.

6. Card CM, Yu SS, Swartz MA. Emerging roles of lymphatic endothelium in regulating adaptive immunity. The Journal of clinical investigation. 2014 Mar;124(3):943-52. PubMed PMID: 24590280. Pubmed Central PMCID: 3938271.

7. Humbert M, Hugues S, Dubrot J. Shaping of Peripheral T Cell Responses by Lymphatic Endothelial Cells. Frontiers in immunology. 2016;7:684. PubMed PMID: 28127298. Pubmed Central PMCID: 5226940.

8. Hirosue S, Dubrot J. Modes of Antigen Presentation by Lymph Node Stromal Cells and Their Immunological Implications. Frontiers in immunology. 2015;6:446. PubMed PMID: 26441957. Pubmed Central PMCID: 4561840.

9. Dubrot J, Duraes FV, Potin L, Capotosti F, Brighouse D, Suter T, et al. Lymph node stromal cells acquire peptide-MHCII complexes from dendritic cells and induce antigen-specific CD4(+) T cell tolerance. The Journal of experimental medicine. 2014 Jun 2;211(6):1153-66. PubMed PMID: 24842370. Pubmed Central PMCID: 4042642.

10. Fletcher AL, Lukacs-Kornek V, Reynoso ED, Pinner SE, Bellemare-Pelletier A, Curry MS, et al. Lymph node fibroblastic reticular cells directly present peripheral tissue antigen under steady-state and inflammatory conditions. The Journal of experimental medicine. 2010 Apr 12;207(4):689-97. PubMed PMID: 20308362. Pubmed Central PMCID: 2856033.

11. Norder M, Gutierrez MG, Zicari S, Cervi E, Caruso A, Guzman CA. Lymph node-derived lymphatic endothelial cells express functional costimulatory molecules and impair dendritic cell- induced allogenic T-cell proliferation. FASEB journal : official publication of the Federation of American Societies for Experimental Biology. 2012 Jul;26(7):2835-46. PubMed PMID: 22459150.

12. Tewalt EF, Cohen JN, Rouhani SJ, Guidi CJ, Qiao H, Fahl SP, et al. Lymphatic endothelial cells induce tolerance via PD-L1 and lack of costimulation leading to high-level PD-1 expression on CD8 T cells. Blood. 2012 Dec 6;120(24):4772-82. PubMed PMID: 22993390. Pubmed Central PMCID: 3520619.

13. Sixt M, Kanazawa N, Selg M, Samson T, Roos G, Reinhardt DP, et al. The conduit system transports soluble antigens from the afferent lymph to resident dendritic cells in the T cell area of the lymph node. Immunity. 2005 Jan;22(1):19-29. PubMed PMID: 15664156.

- 13 -

14. Hirosue S, Vokali E, Raghavan VR, Rincon-Restrepo M, Lund AW, Corthesy-Henrioud P, et al. Steady-state antigen scavenging, cross-presentation, and CD8+ T cell priming: a new role for lymphatic endothelial cells. Journal of immunology. 2014 Jun 1;192(11):5002-11. PubMed PMID: 24795456. Pubmed Central PMCID: 4025611.

15. Lund AW, Duraes FV, Hirosue S, Raghavan VR, Nembrini C, Thomas SN, et al. VEGF-C promotes immune tolerance in B16 melanomas and cross-presentation of tumor antigen by lymph node lymphatics. Cell reports. 2012 Mar 29;1(3):191-9. PubMed PMID: 22832193.

16. Lee JW, Epardaud M, Sun J, Becker JE, Cheng AC, Yonekura AR, et al. Peripheral antigen display by lymph node stroma promotes T cell tolerance to intestinal self. Nature immunology. 2007 Feb;8(2):181-90. PubMed PMID: 17195844.

17. Magnusson FC, Liblau RS, von Boehmer H, Pittet MJ, Lee JW, Turley SJ, et al. Direct presentation of antigen by lymph node stromal cells protects against CD8 T-cell-mediated intestinal autoimmunity. Gastroenterology. 2008 Apr;134(4):1028-37. PubMed PMID: 18395084.

18. Cohen JN, Guidi CJ, Tewalt EF, Qiao H, Rouhani SJ, Ruddell A, et al. Lymph node-resident lymphatic endothelial cells mediate peripheral tolerance via Aire-independent direct antigen presentation. The Journal of experimental medicine. 2010 Apr 12;207(4):681-8. PubMed PMID: 20308365. Pubmed Central PMCID: 2856027.

19. Nichols LA, Chen Y, Colella TA, Bennett CL, Clausen BE, Engelhard VH. Deletional self- tolerance to a melanocyte/melanoma antigen derived from tyrosinase is mediated by a radio-resistant cell in peripheral and mesenteric lymph nodes. Journal of immunology. 2007 Jul 15;179(2):993-1003. PubMed PMID: 17617591.

20. Fletcher AL, Malhotra D, Turley SJ. Lymph node stroma broaden the peripheral tolerance paradigm. Trends in immunology. 2011 Jan;32(1):12-8. PubMed PMID: 21147035. Pubmed Central PMCID: 3163075.

21. Yip L, Su L, Sheng D, Chang P, Atkinson M, Czesak M, et al. Deaf1 isoforms control the expression of genes encoding peripheral tissue antigens in the pancreatic lymph nodes during type 1 diabetes. Nature immunology. 2009 Sep;10(9):1026-33. PubMed PMID: 19668219. Pubmed Central PMCID: 2752139.

22. Baptista AP, Roozendaal R, Reijmers RM, Koning JJ, Unger WW, Greuter M, et al. Lymph node stromal cells constrain immunity via MHC class II self-antigen presentation. eLife. 2014 Nov 19;3. PubMed PMID: 25407678. Pubmed Central PMCID: 4270074.

23. Cire S, Da Rocha S, Ferrand M, Collins MK, Galy A. In Vivo Gene Delivery to Lymph Node Stromal Cells Leads to Transgene-specific CD8+ T Cell Anergy in Mice. Molecular therapy : the journal of the American Society of Gene Therapy. 2016 Nov;24(11):1965-73. PubMed PMID: 27562586. Pubmed Central PMCID: 5154481.

24. Diehl L, Schurich A, Grochtmann R, Hegenbarth S, Chen L, Knolle PA. Tolerogenic maturation of liver sinusoidal endothelial cells promotes B7-homolog 1-dependent CD8+ T cell tolerance. Hepatology. 2008 Jan;47(1):296-305. PubMed PMID: 17975811.

25. Limmer A, Ohl J, Kurts C, Ljunggren HG, Reiss Y, Groettrup M, et al. Efficient presentation of exogenous antigen by liver endothelial cells to CD8+ T cells results in antigen-specific T-cell tolerance. Nature medicine. 2000 Dec;6(12):1348-54. PubMed PMID: 11100119.

26. von Oppen N, Schurich A, Hegenbarth S, Stabenow D, Tolba R, Weiskirchen R, et al. Systemic antigen cross-presented by liver sinusoidal endothelial cells induces liver-specific CD8 T- cell retention and tolerization. Hepatology. 2009 May;49(5):1664-72. PubMed PMID: 19205034.

- 14 -

27. Kruse N, Neumann K, Schrage A, Derkow K, Schott E, Erben U, et al. Priming of CD4+ T cells by liver sinusoidal endothelial cells induces CD25low forkhead box protein 3- regulatory T cells suppressing autoimmune hepatitis. Hepatology. 2009 Dec;50(6):1904-13. PubMed PMID: 19787806.

28. Wittlich M, Dudek M, Bottcher JP, Schanz O, Hegenbarth S, Bopp T, et al. Liver sinusoidal endothelial cell cross-priming is supported by CD4 T cell-derived IL-2. Journal of hepatology. 2017 May;66(5):978-86. PubMed PMID: 28025060.

29. Dieterich LC, Ikenberg K, Cetintas T, Kapaklikaya K, Hutmacher C, Detmar M. Tumor- Associated Lymphatic Vessels Upregulate PDL1 to Inhibit T-Cell Activation. Frontiers in immunology. 2017;8:66. PubMed PMID: 28217128. Pubmed Central PMCID: 5289955.

30. Dendrou CA, Fugger L, Friese MA. Immunopathology of multiple sclerosis. Nature reviews Immunology. 2015 Sep 15;15(9):545-58. PubMed PMID: 26250739.

31. Louveau A, Smirnov I, Keyes TJ, Eccles JD, Rouhani SJ, Peske JD, et al. Structural and functional features of central nervous system lymphatic vessels. Nature. 2015 Jul 16;523(7560):337- 41. PubMed PMID: 26030524. Pubmed Central PMCID: 4506234.

32. Absinta M, Ha SK, Nair G, Sati P, Luciano NJ, Palisoc M, et al. Human and nonhuman primate meninges harbor lymphatic vessels that can be visualized noninvasively by MRI. eLife. 2017 Oct 3;6. PubMed PMID: 28971799. Pubmed Central PMCID: 5626482.

33. Aspelund A, Antila S, Proulx ST, Karlsen TV, Karaman S, Detmar M, et al. A dural lymphatic vascular system that drains brain interstitial fluid and macromolecules. The Journal of experimental medicine. 2015 Jun 29;212(7):991-9. PubMed PMID: 26077718. Pubmed Central PMCID: 4493418.

34. Mezey E, Palkovits M. Neuroanatomy: Forgotten findings of brain lymphatics. Nature. 2015 Aug 27;524(7566):415. PubMed PMID: 26310754.

35. Engelhardt B, Vajkoczy P, Weller RO. The movers and shapers in immune privilege of the CNS. Nature immunology. 2017 Feb;18(2):123-31. PubMed PMID: 28092374.

36. Louveau A, Harris TH, Kipnis J. Revisiting the Mechanisms of CNS Immune Privilege. Trends in immunology. 2015 Oct;36(10):569-77. PubMed PMID: 26431936. Pubmed Central PMCID: 4593064.

37. Meyer C, Martin-Blondel G, Liblau RS. Endothelial cells and lymphatics at the interface between the immune and central nervous systems: implications for multiple sclerosis. Current opinion in neurology. 2017 Jun;30(3):222-30. PubMed PMID: 28323646.

38. Baxter AG. The origin and application of experimental autoimmune encephalomyelitis. Nature reviews Immunology. 2007 Nov;7(11):904-12. PubMed PMID: 17917672.

39. Frommer F, Heinen TJ, Wunderlich FT, Yogev N, Buch T, Roers A, et al. Tolerance without clonal expansion: self-antigen-expressing B cells program self-reactive T cells for future deletion. Journal of immunology. 2008 Oct 15;181(8):5748-59. PubMed PMID: 18832734.

40. Bazigou E, Lyons OT, Smith A, Venn GE, Cope C, Brown NA, et al. Genes regulating lymphangiogenesis control venous valve formation and maintenance in mice. The Journal of clinical investigation. 2011 Aug;121(8):2984-92. PubMed PMID: 21765212. Pubmed Central PMCID: 3223924.

41. Hashimoto K, Joshi SK, Koni PA. A conditional null allele of the major histocompatibility IA- beta chain gene. Genesis. 2002 Feb;32(2):152-3. PubMed PMID: 11857806.

- 15 -

Figure 1. MHC-II expression by LECs is upregulated during EAE. A. Dot plot representing the frequencies of fibroblastic reticular cells (FRCs; gp38+ CD31-), lymphatic endothelial cells (LECs; gp38+ CD31+), blood endothelial cells (BECs; gp38- CD31+) and double negative cells (DNCs; gp38- CD31-) among CD45-negative cells in skin LNs. B. Indicated time points represent days after EAE immunization. Fold change corresponds to MHC-II MFI at indicated time points for EAE mice over MHC-II MFI in naïve mice. MHC-II expression was measured by FACS (MFI) at indicated times for LECs, BECs and FRCs. Results show the mean ± SEM derived from 3 mice and are representative of two independent experiments. One-way ANOVA tests with Bonferroni post hoc tests were performed. *, p<0.05; **, p<0.01; ***, ns, non-significant.

- 16 -

Figure 2. MHC-II abrogation in LECs exacerbates EAE severity. A. Control mice (MHCIIfl/fl) or MHCII-/- LEC mice (Prox1CreERT2+/--MHCIIfl/fl), which had been injected with Tx, were treated or not with IFN-γ s.c. (2µg/mouse). One day later, cells were harvested from skin draining LNs. Histograms represent MHC-II MFI in lymphatic endothelial cells (LECs), blood endothelial cells (BECs) and fibroblastic reticular cells (FRCs), measured by FACS. B. EAE clinical scores after EAE immunization. Results show the mean ± SEM derived from 7 mice and are representative of two independent experiments. Two-way ANOVA tests with Bonferroni post hoc tests were performed. *, p<0.05.

- 17 -

Figure 3. Mouse model for the presentation of MOG35-55 via MHC-II at the surface of LECs. A. Upper panel, IiMOG is a mutant version of the invariant chain Ii, in which class II-associated Ii peptide (CLIP) fragment is replaced by MOG35-55 peptide. Intermediate panel, in mice expressing WT invariant chain Ii, after proteolysis of Ii, CLIP (purple) hides the MHC-II groove, until it is exchanged with an antigenic peptide (blue) by H2-M. The antigenic peptide is subsequently presented by MHC-II at the cell surface. Lower panel, in mice expressing mutant invariant chain IiMOG, MOG35-55 (green) is presented by MHC-II at the cell surface. B. IiMOG is expressed under the Rosa26 promoter, downstream of a floxed STOP cassette. Under Cre-mediated recombination, the STOP cassette is excised and IiMOG can be expressed. C. We crossed IiMOGfl/fl mice with Prox1CreERT2+/- mice, giving rise to Prox1CreERT2+/--IiMOGfl/fl, that we call “MOG35-55 LEC mice”, in which IiMOG is expressed under the Prox-1 promoter. Prox-1 is a transcription factor selectively expressed by lymphatic endothelial cells (LECs), therefore, in MOG35-

55 LEC mice, MOG35-55 peptide is presented through MHC-II molecules at the surface of LECs, after tamoxifen injection (the expression of Cre is inducible in this model).

- 18 -

Figure 4. MOG35-55 presentation by LECs dampens EAE severity. fl/fl ERT2+/- fl/fl Control mice (IiMOG ) or MOG35-55 LEC mice (Prox1Cre -IiMOG ) that had been treated with Tx were immunized for EAE. A. EAE clinical scores; B. Body weight change; C. Survival. Results show the mean ± SEM derived from 8 mice and are representative of three independent experiments. For A and B, two-way ANOVA tests with Bonferroni post hoc tests were performed. *, p<0.05; **, p<0.01.

- 19 -

Figure 5. MOG35-55 presentation by LECs modulates the T cell infiltrate towards a tolerogenic phenotype in the spinal cord during EAE. Cells were harvested from the spinal cord (SC) and skin draining LNs at D17 (peak disease) after EAE immunization. Frequencies of effector T cells and Tregs were measured by FACS for control mice fl/fl +/- fl/fl (IiMOG ) or MOG35-55 LEC mice (Prox1Cre -IiMOG ). A. Absolute numbers of infiltrating CD4+ T cells in the SC. B and C. Frequencies (upper panels) and absolute numbers (lower panels) of Tregs (Foxp3+ CD25+) among CD4+ T cells (B) and IFN-γ+ or IL-17+ among CD4+ T cells (C) in the SC. D. [Effector CD4+ T cell absolute number / Treg absolute number] ratios in the SC are depicted. E and F. Frequencies (upper panels) and absolute numbers (lower panels) of Tregs among CD4+ T cells (E) and IFN-γ+ or IL-17+ among CD4+ T cells (F) in skin draining LNs. Results show the mean ± SEM derived from 6 to 8 mice and are representative of two independent experiments. Two-tailed Mann-Whitney tests were performed. *, p<0.05; **, p<0.01; ***, p<0.001; ns, non-significant.

- 20 -

IV.3. Discussion

LN-LECs are implicated in CD8+ T cell tolerance through multiple pathways (Card et al., 2014; Humbert et al., 2016). In particular, dysfunctions in the ability of LN-LECs to present endogenously-expressed MHC-I-restricted PTAs to CD8+ T cells are associated with autoimmunity (Lee et al., 2007; Magnusson et al., 2008; Nichols et al., 2007). Emerging evidence is in favor of a role for LNSCs in CD4+ T cell tolerance to MHCII-restricted PTAs, although this phenomenon still needs to be fully elucidated and the specific contributions of each LNSC subsets to CD4+ T cell tolerance remain to be characterized (Baptista et al., 2014; Cire et al., 2016). For instance, the abrogation of MHC-II expression in all LNSC subsets (LECs, BECs and FRCs) in murine LNs alter the Treg compartment, leading to increased CD4+ and CD8+ T cell activation and effector functions, which correlates with signs of spontaneous autoimmunity in elderly mice (Dubrot et al., in press). Of note, LN-LEC MHCII-restricted antigen-presenting abilities seem to have a particularly important role in supporting the maintenance of Treg populations (Dubrot et al., in press). Here, we investigated whether LECs contribute as MHCII-restricted myelin antigen-presenting cells to autoreactive CD4+ T cell responses in EAE, a CD4+ T cell-mediated pathological mouse model of MS. We used genetically-engineered mice, in which MHCII-restricted antigen- presenting functions of LECs were selectively abrogated or enforced.

Result summary The expression of MHC-II was upregulated in LECs in LNs draining the site of immunization, during the early phase of EAE. Furthermore, the selective genetic abrogation of MHC-II expression in LECs led to an exacerbation of the disease severity.

The genetic enforcement of MHC-II-restricted myelin-derived MOG35-55 peptide presentation in LECs led, in contrast, to delayed onset and reduced EAE severity. At peak disease, numbers of cells infiltrating the spinal cord (SC), as well as the ratios [effector T cells (Th1 or Th17) / Treg], were drastically reduced in mice in which LECs presented MOG35-55 through MHC-II molecules. However, it did not have any effect on the frequency and numbers of encephalitogenic effector CD4+ T cells (Th1 and Th17) in the LNs draining the site of EAE immunization. Our results therefore suggest that this pathway might be involved in the modulation of effector T cells rather than the priming of naïve T cells in LNs draining EAE immunization sites. Consequently, by presenting myelin-derived peptides on MHCII, LECs contribute to the tolerization of antigen-specific CD4+ T cells. Overall, our results highlight a role for MHCII-restricted antigen

153

presentation by LECs, in impacting peripheral encephalitogenic T cell responses and EAE pathogenesis.

Potential caveats in our mouse models targeting LECs In adult mice, Prox-1 is not completely specifically expressed in LECs, as it is also found in neural stem cells and hepatocytes (Bunk et al., 2016; Sosa-Pineda et al., 2000; Stergiopoulos et al., 2014). The expression of Prox-1 in neural stem cells should not be an issue in our mouse models, as these cells do not express MHC-II (Bunk et al., 2016; Lavado and Oliver, 2007; Stergiopoulos et al., 2014). Hepatocytes can aberrantly express MHC-II under certain inflammatory conditions, such as viral infections, autoimmunity and hepatitis, and have been suggested to induce tolerance - Treg induction - to ectopically-expressed autoantigen (MBP) in EAE [reviewed in (Mehrfeld et al., 2018)] (Herkel et al., 2003; Luth et al., 2008). However, whether tolerance induction resulted from MHC-II-restricted antigen presentation by hepatocytes or whether the antigen was acquired by other liver-resident APCs remain to be elucidated and is under debate (Luth et al., 2008; Mehrfeld et al., 2018). In order to ensure that our results were not due to putative hepatocyte APC functions or to LSEC APC functions, these subtype of LECs also expressing Prox-1, it would be interesting to analyze the T cell content in the liver of MOG35-55 LEC mice after EAE immunization (Lukacs-Kornek, 2016; Mehrfeld et al., 2018; Shetty et al., 2018; Wohlleber and Knolle, 2016). Another alternative would be to confirm our results using Flt4CreERT2 mice (also called Vegfr3CreERT2) (Martinez-Corral et al., 2016). In these mice LECs are not the only cells that express Cre neither, with potential expression of Vegfr3 in neural stem cells, for instance (Calvo et al., 2011). However, similar results obtained in both Vegfr3CreERT2 and Prox1CreERT2 models should reinforce the implication of LECs as unconventional APCs in EAE.

Artificial model or physiological phenomenon?

We observed a delayed onset and attenuated EAE severity in MOG35-55 LEC mice, in which

MHC-II-restricted myelin-derived MOG35-55 peptide is enforced in LECs. Nonetheless, whether these findings reflect a physiological phenomenon or whether this model is artificial remains to be elucidated. For instance, it would be interesting to determine whether LECs from WT mice endogenously express MOG protein, as a PTA. MOG was shown to be expressed in minute amounts in WT C56BL/6 mice spleen and thymus (Delarasse et al., 2003). The weak thymic MOG expression correlated with an absence of thymus- derived tolerance to MOG in this mouse strain, assessed by comparing WT and MOG-deficient mice, which might explain the susceptibility of WT C56BL/6 mice to MOG-induced EAE,

154

although this is a matter of debate (Delarasse et al., 2003; Kieback et al., 2016; Rangachari and Kuchroo, 2013). MBP and PLP, other myelin-derived proteins, were found in the thymus of C57BL/6 mice, while these mice are resistant to EAE induced with MBP or PLP (Anderson et al., 2000; Klein et al., 2000; Rangachari and Kuchroo, 2013; Targoni and Lehmann, 1998). Nonetheless, it is possible that LECs express MOG (and/or MBP or PLP). In mTECs, the expression of the vast majority of PTAs is regulated by Aire, while PTAs expressed by LNSCs do not depend on Aire but on other transcription factors (Anderson et al., 2002; Derbinski et al., 2005) (Cohen et al., 2010).

If MOG was not physiologically expressed by LECs, this study would still be relevant, as LECs could also acquire exogenous MOG where this protein is normally expressed, i.e. in the CNS myelin sheaths. LN-LECs have the ability to acquire exogenous antigens, using mechanisms such as direct capture or cell-membrane transfer (Dubrot et al., 2014; Hirosue et al., 2014; Lund et al., 2012). In addition, meningeal LVs, composed of LECs, and which drain the CSF components towards deep cervical LNs, have recently been described (Aspelund et al., 2015; Louveau et al., 2015b; Mezey and Palkovits, 2015) . Whether LECs from the meningeal lymphatics could directly acquire myelin-derived antigens is a possibility that remains to be explored. In addition, LN- LECs could acquire myelin-derived peptide/MHC-II complexes from DCs migrating from the CNS to the deep cervical LNs (Louveau et al., 2015b). A putative implication of meningeal LECs and/or of LECs from deep cervical LNs, as tolerogenic APCs, is discussed later.

Role of MHC-II+ LECs in priming vs effector EAE phase? We show that LN-LECs upregulate MHC-II expression a few days after EAE induction. However, the CD4+ T cell numbers and phenotype in the LNs draining the site of EAE immunization were not affected. This suggests that the effect we observed would be rather due to a modulation of effector T cell responses rather than to the priming of naïve T cells in the draining LNs. In addition, the genetic enforcement of MHC-II-restricted myelin-derived MOG35-55 peptide presentation in LECs led to decreased numbers of cells infiltrating the spinal cord (SC) and to a shift in the CD4+ T cell infiltrate towards a suppressive phenotype. Therefore, whether the findings we obtained result from modulations in the priming phase or the effector phase remains opened. Passive EAE induction bypasses the priming phase, which occurs in the LNs draining the site of EAE immunization (Bettelli et al., 2003; Hohlfeld and Steinman, 2017; + Peters et al., 2015). Therefore, adoptive transfer of MOG-specific effector CD4 T cells into MOG35- -/- 55 LEC mice or MHCII LEC mice, compared with control mice, may help answering this question.

155

T cell responses At EAE peak disease, we observed a decrease in the frequency of IFN-γ-producing cells but no difference in IL-17-producing cell frequency, in the SC of MOG35-55 LEC mice, compared with floxed control mice. Whether the reduction in Th1 cell frequency was due to anergy or deletion of these cells by LECs or to a suppression of these cells by Tregs remains to be elucidated. Regarding Th17 cells, it is possible that the encephalitogenicity of Th17 cells may be modified, even if the frequency of total Th17 cells was not affected. The secretion of GM-CSF by Th17 cells is a way to distinguish encephalitogenic from non-encephalogenic Th17 cells (Codarri et al., 2011; El-Behi et al., 2011). This cytokine leads to the recruitment of myeloid cells in the CNS and further inflammation (Croxford et al., 2015a; King et al., 2009; King et al., 2010; Spath et al., 2017). Therefore, it would be interesting to assess the production of GM-CSF, among IL-17-producing CD4+ T cells. + -/- The adoptive transfer of effector CD4 T cells from MHCII LEC mice, MOG35-55 LEC mice, or floxed control mice, into naïve WT recipient mice may also shed light on whether the encephalogenicity of CD4+ T cells is modified by antigen-presenting LECs. In addition, whether LECs, as tolerogenic APCs, affect naïve or effector CD4+ T cells, by modulating/inducing their activation, proliferation, differentiation, dysfunction or apoptosis remains to be elucidated. This might be assessed by transferring naïve MOG-specific CD4+ T cells into

MOG35-55 LEC mice, and analyzing their phenotype, as well as by performing tamoxifen (Tx) injection after EAE induction, in order to investigate the effect of antigen-presenting LECs on already ongoing T cell responses.

Moreover, we observed an increase in Treg frequency in the SC at peak disease in MOG35-55 LEC mice. DEREG mice, in which Tregs can be depleted upon diphtheria toxin injection, could be used in order to determine whether the results we obtained were Treg-mediated (Lahl et al., 2007). Furthermore, whether antigen-presenting LECs support the proliferation of tTregs or promote the induction of pTregs remains to be elucidated (Abbas et al., 2013; Kanamori et al., 2016; Lu et al., 2017). In addition, Tregs could differentiate from Th17 cells (Kleinewietfeld and Hafler, 2013). The transfer of Th17 cells isolated from Foxp3-RFP+ RORγt-GFP+ (RORγt is a Th17-specific transcription factor) mice into MOG35-55 LEC mice may help to determine whether there is a Th17 to Treg conversion (Yang et al., 2016). In addition to the frequency of Tregs, their suppressive functions could be impacted in our two -/- mouse models; Treg exposed to MOG35-55 presenting LECs, or in contrast to MHCII LECs in vivo may acquire, before or after EAE induction, distinct suppressive functions.

156

LN-LECs vs meningeal LECs LECs other than LN-LECs, e.g. LSECs and tumors-associated LECs, also have the ability to function as tolerogenic APCs (Dieterich et al., 2017; Lukacs-Kornek, 2016; Mehrfeld et al., 2018; Shetty et al., 2018; Wohlleber and Knolle, 2016). In addition, as above-mentioned, meningeal lymphatic vessels (LVs), which drain the CNS towards the deep cervical LNs, have recently been characterized (Aspelund et al., 2015; Louveau et al., 2015b; Mezey and Palkovits, 2015). Therefore, it is possible that our findings result from antigen presentation by LECs from the meningeal LVs or from the deep cervical LNs, rather than LECs from the LNs draining the site of EAE immunization. Meninges are an essential immunological site that allows CNS immune surveillance to function correctly. A system of LVs that express the traditional LEC markers (Prox-1, CD31, gp38, Lyve- 1) and that runs along the perisinusal space has been recently identified in rodents and later visualized in human and non-human primates (Absinta et al., 2017; Aspelund et al., 2015; Louveau et al., 2015b; Mezey and Palkovits, 2015). Meningeal LVs carry numerous immune cells under physiological conditions, suggesting their role in the immune surveillance of the brain (Louveau et al., 2015b). So far, mainly DCs have been detected in meningeal LVs, but it is possible that T cells use intralymphatic routes to travel in and out the CNS, and recirculate in cervical LNs (Louveau et al., 2015b). Crawling T cells, including Tregs, in afferent LVs, have been described [reviewed in (Hunter et al., 2016)]. Tertiary lymphoid structures (TLSs) are sites that resemble SLOs, which organize by the accumulation of lymphoid and stromal cells, in response to chronic inflammation, such as in the case of autoimmunity (Neyt et al., 2012; Pipi et al., 2018; Ruddle, 2016). LVs from TLSs seem to function and be regulated in a similar way to LN LVs (Ruddle, 2016). Interestingly, FRC-like stromal cell networks develop in the meninges during Th17-induced passive EAE in a pathway dependent on LT-βR (Pikor et al., 2015). Louveau and colleagues recently investigated the role of meningeal LVs in EAE (Louveau et al., 2018). Meningeal LVs did not expand during CNS inflammation, but were implicated in the drainage of CSF components to the cervical LNs. The ablation of lymphatic drainage using a photosensitizing agent and laser at D8 post-immunization led to decreased EAE severity (Louveau et al., 2018). Whether CD4+ T cell responses can be modulated by antigen-presenting LECs in meningeal LVs and/or cervical LNs, which could subsequently affect EAE development, remains to be investigated. For instance, it is not known whether LECs in meningeal LVs express MHC-II upon inflammation. In addition, meningeal LECs were shown to have a transcriptional signature very different from that of skin or diaphragm during EAE (Louveau et al., 2018).

157

Passive EAE induction, as described above, which allows a distinction between the priming phase and the effector phase, may shed light on the relative contributions of LECs from immunization site draining LNs in one hand, and meningeal LECs and/or LECs from the deep cervical LNs in the other hand (Bettelli et al., 2003; Hohlfeld and Steinman, 2017; Peters et al., 2015). We observed a drastic decrease in the CD4+ T cell infiltrate in the SC at peak disease, therefore, an analysis of the expression of VLA-4, or of other molecules implicated in the entry of T cells into the CNS, by CD4+ T cells in LECs from LNs draining the immunization site or from cervical LNs could be interesting (Baron et al., 1993; Glatigny et al., 2011). In addition, LN transplantation experiments, by transplanting either skin LNs or cervical LNs, from -/- either MOG35-55 LEC mice or MHCII LEC mice into WT mice, before Tx injection and before active EAE induction, may also help determining the relative contribution of different LECs, although this approach may be challenging. Ex vivo antigen presentation assays using EAE WT mice may also help to elucidate which LEC types are implicated as tolerogenic APCs, however, this does not necessarily reflect what occurs in vivo as cultured LECs may lose many properties, and is also technically challenging. Another possibility would be to culture LECs from Tx-treated MOG35-55-LEC mice in the presence of IFN-γ in order to promote their MHC-II expression. Co-culture with 2D2 cells may shed light on the expansion and polarization of these MOG-specific CD4+ T cells. Furthermore, the respective contributions of meningeal LECs vs LECs from the deep cervical LNs may be determined by using an S1P1R inhibitor, which blocks T cell egress from the LNs (Cyster and Schwab, 2012). The analysis (isolation of cervical LNs or meningeal LVs) of transferred effector CD4+ T cells into -/- MOG35-55-LEC mice or MHCII LEC mice may inform on whether meningeal LECs and/or LECs from the deep cervical LNs modulate the phenotype of already primed CD4+ T cells in vivo.

Moreover, the pathways involved in the induction of tolerance by the LEC type that act as tolerogenic APC in our model remain to be elucidated. For instance, LN-LECs do not express co-stimulatory molecules and they upregulate the expression of PD-L1 under inflammatory conditions (Dubrot et al., 2014; Fletcher et al., 2010; Norder et al., 2012; Tewalt et al., 2012) (Dubrot et al., in press). Clonal deletion of CD8+ T cells after presentation of PTAs through MHC-I by LN-LECs was demonstrated to be mediated by PD-L1 (Rouhani et al., 2015). Therefore, an analysis of the phenotype of LECs implicated in tolerance induction in our model, such as measuring the expression of co-stimulatory molecules, PD-L1, and other co-inhibitory molecules, would be of interest. In case meningeal LECs

158

would be implicated in our model, their phenotype might be completely different from that of LN- LECs, as suggested by Louveau et al. with distinct transcriptional signature in EAE (Louveau et al., 2018). Putative tolerance induction by meningeal LECs might therefore rely on pathways distinct from LN-LECs.

Role of IFN-γ As previously mentioned, the expression of pIV, the promoter of CIITA expressed in LECs, is IFN-γ-inducible, LECs therefore expressing MHC-II only under inflammatory conditions (Reith et al., 2005). We noticed, in WT mice, that the expression of MHC-II was upregulated in LECs in LNs draining the site of EAE immunization, during the course of the disease, as soon as D5 post-immunization, and gradually increasing until reaching a peak at D15. Whether the frequencies and numbers of IFN-γ-producing cells in the draining LNs correlate with the kinetic of LN-LEC MHC-II expression remains to be elucidated. In addition, whether circulating IFN-γ levels or IFN-γ receptor (IFN-γR) expression in LECs are increased during the course of EAE, which might explain the increase in MHC-II expression in LECs, and whether those levels differ between mutant and control mice, remain to be investigated. For instance, our group observed an increase in IFN-γR expression in LECs from elderly compared to young adult naïve mice, suggesting an enhanced sensitivity of LECs to IFN-γ upon aging (Dubrot et al., in press). Accordingly, MHC-II expression levels were higher in LECs from aging compared to young mice (Dubrot et al., in press). If there was a modification of IFN-γR expression in LECs, which may modulate their MHC-II- restricted antigen-presenting functions, using Prox1creERT2xPIG (overexpressing IFN-γR) mice or Prox1creERT2xIFN-γRfl/fl (lacking IFN-γR), in the context of EAE, might shed light on this phenomenon (Kammertoens et al., 2017).

Potential for translation into the clinic? Is it relevant in human? Most of the work performed so far by different groups showing that LNSCs, among which LECs, function as unconventional APCs, for MHC-I- or MHC-II-restricted presentation, and for the presentation of exogenous peptides or of endogenously-expressed PTA-derived peptides, has been performed in murine models. Two studies were recently published, regarding a direct role of human FRCs in the modulation of peripheral T cell responses, but there is no published work, to our knowledge, investigating the role of human LECs as putative APCs (Costa et al., 2018; Knoblich et al., 2018). Knoblich and colleagues showed in vitro and in situ that FRCs from LNs or

159

tonsils inhibit CD4+ and CD8+ T cell proliferation and differentiation into effector cells, without demonstrating a role for human FRCs as APCs (Knoblich et al., 2018). T cell inhibition was due to four distinct mechanisms occurring simultaneously, involving the expression of IDO, A2AR, PGE2 and TGF-βR. In addition, Costa et al. showed that a subset of cancer-associated fibroblasts from human breast cancer has the ability to induce the differentiation of CD4+ CD25+ T cells into Tregs (CD25hi Foxp3hi), for which the expression of dipeptyl peptidase-4 (DPP-4), CD73 and the co-inhibitory molecule B7H3 was required (Costa et al., 2018). This subset of cancer- associated fibroblasts also enhanced the suppressive functions of Tregs. Whether human LECs function as unconventional APCs and are implicated in peripheral T cell tolerance remains to be elucidated, as well as whether human LECs could be implicated in the pathogenesis of MS and/or could be targeted for potential MS therapies. For instance, if LECs functioned as tolerogenic APCs in human in inflammatory situations, it would be interesting to determine whether defects in putative PTA expression and/or in the presentation of these PTAs, or in the acquisition of exogenous self-antigens, could play a role in the immunopathophysiology of MS. In addition, investigating the effects of actual MS treatments on the ability of LECs to modulate peripheral T cell responses might also be of interest. Altogether, a better understanding of the implication of LECs as unconventional APCs in autoimmunity, in mouse models as well as in human, may shed light on new targets for the design of therapeutic drugs.

Targeting LECs to treat diseases? Anti-cancer treatments targeting stromal cells, including some targeting LECs, are currently used in the clinic (Chan et al., 2016; Dieterich and Detmar, 2016; Pure and Lo, 2016; Szot et al., 2018). Nonetheless, in the anti-tumor context, these therapies are used to inhibit stromal activation that promotes tumor growth, to restrain lymphangiogenesis, or to utilize stromal cells as a way of delivery for cytotoxic drugs (antibody-drug conjugate) (Chan et al., 2016; Dieterich and Detmar, 2016; Pure and Lo, 2016; Szot et al., 2018). In addition, lymphatic functions can be targeted for the treatment of rheumatoid arthritis, an autoimmune disease [reviewed in (Bouta et al., 2018)]. If human LECs functioned as tolerogenic APCs upon inflammation, an interesting goal would be to harness the power of LECs for the induction of tolerance. Since LECs play a tolerogenic role towards autoreactive CD4+ T cell responses in EAE as unconventional MHC-II-restricted APCs, myelin-derived peptide targeting to LECs through a

160

surface receptor specific for LECs, such as VEGFR-3 or Lyve-1, might induce tolerance towards myelin and be beneficial in MS, if relevant in human, as above-mentioned. For instance, in EAE mice, MOG targeting via Siglec-H to pDCs, whose antigen-presenting functions play a tolerogenic role in EAE, was shown to dampen disease severity (Loschko et al., 2011a). Furthermore, as LECs constitute the LVs, the lymphatic route could be used for the delivery of MOG to LECs (Maisel et al., 2017).

161

V. CONCLUDING REMARKS

Modulation of peripheral T cell responses by unconventional APCs

Besides professional APCs, several cell types from the hematopoietic and non-hematopoietic compartments have the ability to cross-present antigens via MHC-I to CD8+ T cells and/or to present antigens through MHC-II to CD4+ T cells, as unconventional APCs (Kambayashi and Laufer, 2014). Distinct immunological contexts and unconventional APC types lead to various outcomes on T cell responses. The aim of this thesis was to characterize the role of pDCs and LECs in the modulation of peripheral T cell responses, as unconventional APCs, with a particular focus on MHC-II- restricted antigen presentation.

Role of pDCs in anti-tumor immunity

Plasmacytoid DCs are professional IFN-I-producing cells, which also have the ability to (cross)- present antigens via MHC-I and MHC-II molecules (Alculumbre et al., 2018a; Reizis et al., 2011a; Reizis et al., 2011b; Swiecki and Colonna, 2015). These cells are extremely plastic and, depending on the immunological context, their MHCII-mediated antigen-presenting functions can be either tolerogenic or immunogenic (Guery and Hugues, 2013). TA-pDCs harbor a tolerogenic phenotype, due to their immersion in the TME. Nonetheless, the potential of “naïve” pDCs to mount potent anti-tumor responses can be harnessed for anticancer therapies (Aspord et al., 2012; Liu et al., 2008a; Loschko et al., 2011b; Tel et al., 2013a; Tel et al., 2012b). In tumor-bearing mice, pDCs in the distal LNs can be activated by a contralateral vaccination with the TLR9 ligand CpG-B along with a model MHC-II-restricted tumor antigenic peptide (Guery et al., 2014). This induces an enhancement of their MHC-II-mediated antigen- presenting functions, leading to anti-tumor immunity through the priming of Th17 cells (Guery et al., 2014). In the first part of this thesis, we asked the question whether the tolerogenic phenotype of TA- pDCs could be reversed after i.t. administration of CpG-B±MHC-II-restricted peptide. This local treatment led to tumor growth inhibition. However, TA-pDCs did not undergo a tolerogenic-to-immunogenic reprogramming. These cells were refractory to the treatment and did not contribute to its efficacy. The tumor-derived factors responsible for the induction of a tolerogenic pDC state that is refractory to i.t. CpG-B remain to be elucidated. On the contrary, the working model we propose involves neutrophil, cDC and T cell cooperation, subsequently leading to tumor growth control.

162

Role of LECs in autoimmunity

LECs, a LNSC subtype, were long thought to function as simple scaffolds (Turley et al., 2010). However, new roles have emerged for these cells in the regulation of T cell responses, including implication in antigen presentation, as unconventional APCs (Humbert et al., 2016; Turley et al., 2010). Whether LNSCs are implicated in the presentation of MHC-II-restricted PTAs and modulate peripheral CD4+ T cell responses is not fully understood (Baptista et al., 2014; Cire et al., 2016). We recently showed that elderly naïve mice, in which MHC-II expression was abrogated in LNSCs, presented signs of spontaneous autoimmunity (Dubrot et al., in press). The aim of the second part of this thesis was to determine whether LECs contribute to the modulation of autoreactive CD4+ T cell responses, as MHC-II-restricted APCs, in EAE. The selective genetic ablation of MHC-II in LECs exacerbated disease severity, whereas the genetic enforcement of MOG35-55 presentation via MHC-II by LECs led to delayed onset and dampened EAE severity. MOG35-55 presentation in LECs induced an important decrease in the spinal cord CD4+ T cell infiltrate and a modulation of these cells towards a tolerogenic phenotype. CD4+ T cell numbers and phenotype in the LNs draining the site of EAE immunization were unaffected, suggesting that this pathway may inhibit effector T cell responses, rather than impact the priming of naïve T cells in LNs. Altogether, our findings highlight a role for LECs as unconventional tolerogenic APCs in EAE.

Remarks

Immunological context Our studies highlight the importance of the immunological context for the function of unconventional APCs. Plasmacytoid DCs are extremely plastic cells and their antigen-presenting functions can be immunogenic or tolerogenic (Guery and Hugues, 2013). However, our results suggest that they may be refractory to reprogramming when they are deeply imprinted in certain environments, such as the TME. Furthermore, LECs function as tolerogenic APCs, but they upregulate MHC-II expression only in the presence of IFN-γ (Dubrot et al., 2014; Reith et al., 2005) (Dubrot et al., in press). In the steady state, neither pDCs nor LECs function as MHC-II-restricted APCs, as both of these cell types need a stimulus to upregulate the expression of MHC-II (Alculumbre et al., 2018a; Dubrot et al., 2014; Reith et al., 2005). This idea might have been overlooked in studies,

163

performed in vitro without stimulus or in vivo in the steady-state, concluding that pDCs or LECs cannot function as MHC-II-restricted APCs (Rouhani et al., 2015; See et al., 2017; Villani et al., 2017).

Professional vs unconventional APCs Our studies shed light on the unique characteristics of unconventional APCs, compared with professional APCs (Kambayashi and Laufer, 2014). In the first part of this thesis, we observed that tolerogenic TA-pDCs cannot be reprogrammed by i.t. administration of CpG-B, while TA- cDCs were activated following this treatment, suggesting differences in their intrinsic features. LN-LECs, studied in the second part of this thesis, have been shown to harbor specific characteristics as APCs. Indeed, these cells do not upregulate the expression of co-stimulatory molecules upon inflammation, and rather upregulate the expression of PD-L1 and other co- inhibitory molecules (Dubrot et al., 2014; Fletcher et al., 2010; Norder et al., 2012; Tewalt et al., 2012) (Dubrot et al., in press).

Non-hematopoietic cells in immunity The second part of this thesis shows the importance of the contribution of non-hematopoietic cells to immune responses. LNSCs, including LECs, as well as other cell types in tissues, can affect immune responses in many ways in indirect manners, such as playing structural roles, or in the draining of immune cells and molecules (Card et al., 2014; Humbert et al., 2016). However, increasing evidence suggest that several non-hematopoietic cells also have the ability to present antigens and/or to express or secrete inflammatory molecules (Duraes et al., 2013; Kambayashi and Laufer, 2014; Reynoso and Turley, 2009; Turley et al., 2010). Therefore, the hematopoietic innate and adaptive immune cells are far from being the only players of immune responses.

Mouse vs human Caveats in our studies for potential applications in human involve differences in mouse and human. Indeed, in the first part of this thesis, we used a TLR9 ligand, this TLR being expressed exclusively in pDCs and B cells in human, while being expressed by more cell types in mouse (Jarrossay et al., 2001; Takeuchi and Akira, 2010) (Lindau et al., 2013; Puttur et al., 2016; Schneberger et al., 2013; Xu et al., 2015). In addition, in the second part of this thesis, we investiagted the MHC-II-restricted APC functions of LECs, as previously published work in mouse demonstrated the APC functions of LECs (Cohen et al., 2010; Fletcher et al., 2010;

164

Humbert et al., 2016; Turley et al., 2010). However, no evidence, to our knowledge, has demonstrated the ability of human LECs to present MHC-I- or MHC-II-restricted antigens.

Anti-tumor treatments and autoimmunity A tightly regulated balance between immune activation and tolerance is required in order for the immune system to mount efficient immune responses, for instance required to fight cancers, without inducing adverse effects, such as autoimmunity. Malignancy treatments, which include immunotherapies, are most of the time administered systemically. Therefore, they are often linked with damaging effects, including autoimmunity. The first part of this thesis highlights, in accordance with other studies in mouse and human, the potential of local immunotherapy administration as a way that may solve many aspects that limit these treatments (Fransen et al., 2013; Marabelle et al., 2014).

Conclusion

Overall, the two parts of this thesis highlight important contributions of unconventional APCs in shaping peripheral T cell responses. Therefore, these pathways should be considered in the development of therapies aiming at modulating immune responses in disease development.

165

VI. APPENDICES

VI.1. Appendix 1 Shaping of peripheral T cell responses by lymphatic endothelial cells

Marion Humbert 1, Stéphanie Hugues 1* and Juan Dubrot 1*

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

Published in Frontiers in Immunology (DOI: 10.3389/fimmu.2016.00684)

Objective: Increasing evidence demonstrates a role for lymph node stromal cells (LNSCs) in the modulation of T cell responses. These cells were previously thought to be a simple scaffold allowing the encounter of lymphocytes with their cognate antigen in the lymph nodes (LNs). It is nowadays accepted that LNSCs, among other roles in immune modulation, affect both dendritic cell (DC) and T cell functions. Of particular interest, lymphatic endothelial cells (LECs), a subset of LNSCs that compose the efferent and afferent lymphatic vessels, have been described to express peripheral tissue-restricted antigens (PTAs). In addition, they present peptides derived from these PTAs to CD8+ T cells, via major histocompatibility class I (MHC-I), leading to their clonal deletion. Moreover, LECs, in homeostatic conditions or in the tumor, are able to acquire exogenous antigens and cross-present them to CD8+ T cells. Whether CD4+ T cell responses can also be impacted by LEC MHC-II-restricted antigen presentation is still debated. The aim of this review was to discuss our current knowledge on the impact of LECs on the regulation of T cell responses, focusing on their antigen-presenting ability, particularly in the context of autoimmunity and anti-tumor immunity.

Personal contribution: For this review, I wrote the section related to the presentation of endogenously-expressed PTAs to T cells by LECs and the molecular pathways involved in this process and prepared the Figure 1, under the supervision of Prof. Stéphanie Hugues and Dr. Juan Dubrot, who wrote the other sections and prepared the Figure 2. I participated in the development of the concept of the manuscript, and critically read and revised it.

166

Review published: 12 January 2017 doi: 10.3389/fimmu.2016.00684

Shaping of Peripheral T Cell Responses by Lymphatic Endothelial Cells

Marion Humbert, Stéphanie Hugues* and Juan Dubrot*

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

Lymph node stromal cells (LNSCs) have newly been promoted to the rank of new modulators of T cell responses. The different non-hematopoietic cell subsets in lymph node (LN) were considered for years as a simple scaffold, forming routes and proper environment for antigen (Ag)-lymphocyte encountering. Deeper characterization of those cells has recently clearly shown their impact on both dendritic cell and T cell functions. In particular, lymphatic endothelial cells (LECs) control lymphocyte trafficking and homeostasis in LNs and limit adaptive immune responses. Therefore, the new role of Edited by: LECs in shaping immune responses has drawn the attention of immunologists. Striking Sonia Elhadad, Weill Cornell Medical College, USA is the discovery that LECs, among other LNSCs, ectopically express a large range

Reviewed by: of peripheral tissue-restricted Ags (PTAs), and further present PTA-derived peptides Hideki Nakano, through major histocompatibility class I molecules to induce self-reactive CD8+ T cell National Institute of Environmental deletional tolerance. In addition, both steady-state and tumor-associated LECs were Health Sciences, USA Ingrid E. Dumitriu, described to be capable of exogenous Ag cross-presentation. Whether LECs can St. George’s University similarly impact CD4+ T cell responses through major histocompatibility class II restricted of London, UK Beth Ann Tamburini, Ag presentation is still a matter of debate. Here, we review and discuss our current University of Colorado Denver, USA knowledge on the contribution of Ag-presenting LECs as regulators of peripheral T cell *Correspondence: responses in different immunological contexts, including autoimmunity and cancer. Stéphanie Hugues [email protected]; Keywords: lymphatic endothelial cells, peripheral tissue antigens, antigen presentation, immunomodulation, Juan Dubrot tolerance [email protected]

Specialty section: INTRODUCTION This article was submitted to Inflammation, The lymphatic system comprises a network of vessels together with lymphoid tissues all over the body a section of the journal that drain the extracellular compartment from most of the tissues. It transports lymph fluid, which Frontiers in Immunology is composed of immune cells and proteins drained from interstitial tissues, and helps to dispose of Received: 21 October 2016 toxins and other unwanted components from the body. Lymphocytes follow the lymphatic system to Accepted: 22 December 2016 migrate to infection sites, which supports and facilitates immune responses against potential harms. Published: 12 January 2017 Frequently underestimated by scientists, the importance of lymphatics in controlling the immune Citation: system beyond the regulation of leukocyte trafficking has reached a new level with recent discoveries. Humbert M, Hugues S and Dubrot J The initial observations of the lymphatic system date back to the Ancient Greece, referred to (2017) Shaping of Peripheral T Cell Responses by Lymphatic Endothelial as “white blood.” However, it was in the seventeenth century that Asellius formally discovered the Cells. lymphatic vessels or, what he called, the “milky veins” from mesenteries in dogs (1). Several diseases Front. Immunol. 7:684. have been described to result from failures in the lymphatic system, some of them having life- doi: 10.3389/fimmu.2016.00684 threatening consequences, such as lymphedema (2). Even more strikingly, the role of lymphatics in

Frontiers in Immunology | www.frontiersin.org 1 January 2017 | Volume 7 | Article 684 Humbert et al. Role of LECs in Peripheral T Cell Responses tumor spreading is known since the eighteenth century. Despite by Prox-1-induced reprograming when overexpressed in BECs the ancient knowledge in the lymphatic system organization, (16). After development, functional Prox-1 is required to main- our understanding in its multiple functions has rapidly evolved tain the lymphatic phenotype (19). The molecular mechanisms thanks to the unveiling of lymphatic endothelial cell (LEC) of Prox-1-driven lymphatic differentiation have been reviewed specific markers, such as the surface protein Lyve-1 or the tran- recently (4). In addition, recent studies in zebrafish validated scription factor Prox-1, which are lacking in other endothelial the molecular mechanisms governing lymphatic development, cells. Several studies have subsequently demonstrated that LECs further demonstrating that the vast majority of cells contributing impact immune responses in many ways, including the modula- to LECs in thoracic ducts of zebrafish raised from primitive veins tion of immune cell migration and encounter, effector functions, (3, 20). Later in development, however, the origin of organ- and survival. In this review, we discuss our current understand- specific lymphatic vasculature might be slightly different. Using ing of the imunoregulatory properties of LECs. We specifically cell-fate mapping technologies, a recent publication suggested a discuss the ability of LECs to directly impact T cell responses by combination of venous- and non-venous-derived LECs in the presenting endogenous or exogenous antigens (Ags) to T cells in developing cardiac lymphatics (21). This spatiotemporal discrep- lymph nodes (LNs), and therefore to shape Ag-specific peripheral ancy may explain the difficulties experienced in obtaining a fully T cell responses in the context of autoimmunity and cancers. convincing explanation in the origin of LECs. The specification of LECs during development entails structural and functional differences between blood and lymphatic ORIGIN AND TYPES OF LYMPHATICS systems. In sharp contrast to the circular and closed blood vasculature, lymphatic circulation appears as a linear- and LEC Development blind-ended circuit. Capillaries of the lymphatic system drain Nowadays, it is well accepted and documented that, during interstitial fluids from peripheral organs and tissues thanks to embryogenesis, LECs differentiate from specialized angioblasts the particular organization of LECs in the terminal lymphatics. in the developing veins (3, 4). Nevertheless, this has been con- The uptake of interstitial fluid, macromolecules, and cells is pos- troversial for long until just few decades ago due to, in particular, sible due to the highly permeable thin-walled capillary vessels the lack of knowledge on lymphatic-specific markers. Two dif- composed of a single layer of LECs, which are not covered by ferent hypotheses raised in early twentieth century debated the pericytes or smooth muscle cells and have little or no basement possible origin of the lymphatic system. On one hand, studies on membrane (22). Lymphatic capillaries exhibit discontinuous or embryonic cats suggested that primary lymph sacs arised from “button-like” junctions where the interjunctional gaps act as sites mesenchymal progenitors (5). On the other hand, intravenous of leukocyte entry into the vessels (23, 24). Terminal lymphatic injection of ink in pig embryos revealed that lymph sacs devel- capillaries are linked to the surrounding extracellular matrix by oped from budding of embryonic veins (6, 7). The identification anchoring filaments that sense changes in interstitial pressures of the vascular endothelial growth factor receptor-3 (VEGFR-3) during inflammation. This results in vessel lumen and junction (8) reinforced the latter hypothesis of a common origin for both aperture, therefore facilitating the uptake of tissue-derived fluids. lymphatic and blood endothelial cells (BECs). In adulthood, Deeper, lymphatics change from a drainage-prone phenotype to VEGFR-3 expression is restricted to LECs (8, 9). However, it a collector vessel morphology specialized in lymph transport. is also expressed by angioblasts and developing veins during Collecting lymphatics are surrounded by pericytes and smooth embryonic development (8, 10, 11). Impaired development of muscle cells and possess a basement membrane, displaying con- both lymphatic and blood endothelium in VEGFR-3-deficient tinuous “zipper-like” junctions. The presence of valves (22, 23) mice suggested a common progenitor for LECs and BECs (11). ensures the lymph circulation while preventing retrograde flow. Further ratification of VEGFR-3 requirement for lymphatic development was provided by studies modulating the expres- Main LEC Types sion of its main ligand, the vascular endothelial growth factor Lymphatic vessels are present in almost all the vascularized C (VEGF-C). Overexpression of VEGF-C induced lymphatic organs, with the exception of the bone marrow. LEC immune sprouting and lymphangiogenesis (12–14). modulatory properties represent a growing research area. LN The identification of the homeobox gene Prox-1 in 1993 led LECs being the most characterized subset and representing the few years later to the final confirmation of the theory propos- objective of this review is not discussed in this section. ing the venous origin of lymphatics. Deletion of Prox-1 in mice However, lessons taken from studies performed during the results in the absence of early lymphatic endothelial differentia- last decade clearly establish different functions and possible roles tion and, as a consequence, Prox-1 knockout mice totally lack the for LECs from different anatomic locations. Deeper and careful lymphatic system (10, 15). Prox-1 expression in particular cells future analyses will identify specific immunoregulatory features of the embryonic veins at E9.5 starts the lymphatic polarization of distinct LEC populations. and imprints the LEC signature (10, 15, 16). Transcriptome For decades, lymphatic drainage was suggested to be involved studies showed high proximity in LECs and BECs gene expres- in local immune responses (25). Dendritic cells (DCs) draw all sion profiles. However, Prox-1 acts as the specific regulator the attention in initiating and eliciting tolerance or activation of genes that are inversely regulated in a type-specific manner of the immune system. However, the role of lymph drainage in (17, 18). Indeed, potentially all venous endothelial cells may modulating adaptive immunity and tolerance remained largely give rise to blood or lymphatic endothelium as demonstrated unexplored. K14-VEGFR-3-Ig mice express soluble VEGFR-3-Ig

Frontiers in Immunology | www.frontiersin.org 2 January 2017 | Volume 7 | Article 684 Humbert et al. Role of LECs in Peripheral T Cell Responses via the keratin 14 promoter, resulting in a lack of lymphatic this challenging task is strictly located into highly organized growth, which is restricted to the skin, and in a drop in fluid secondary lymphoid organs (SLOs), such as LNs, Peyer’s patches clearance (26). In these mice, local lymphatic drainage appeared (PPs), and the spleen. These SLOs contain both tissue-derived and to be critical for humoral immunity and acquired tolerance, while blood-borne Ags, therefore facilitating naïve T cell-Ag encounter, T cell responses remained delayed but mostly unaffected. There and subsequent T cell activation and differentiation into T cell is no doubt that additional mechanisms and functions of dermal effectors. This part summarizes the different pathways by which LECs will be discovered in the future. LECs will impact T cell outcome inside and after exiting LNs. LSECs could be seen as LEC counterparts in the liver. First described in 1970 (27), LSECs possess a high ability to filter fluids, Ag Delivery to LNs solutes, and particles from hepatic circulation, occupy a large As described before, LNs are connected to lymphatics, which surface area exposed to blood that carries external food and com- drain peripheral tissue-derived fluids. By connecting tissues to mensal bacterial Ag, and are known to cross-present exogenous draining LNs, LECs facilitate the passive entry of tissue-derived Ag to T cells (28). Ags that can thereby be captured, processed, and presented by A traditional dogma states the immune privilege and lack resident DCs to T cells entering LNs through high endothelial of lymphatic system in the central nervous system (CNS). This venules (37, 38). Soluble Ags are immediately sampled by LN idea has persisted despite the notion of immune surveillance of DCs, whereas particles carrying Ags, such as exosomes, apop- T cells in the brain (29). A recent and elegant study identified totic bodies or microvesicles, which have not been captured by for the first time the lymphatic vasculature in a specific area of subcapsular sinus macrophages, flow to LN medullary sinuses the meninges lining the dural sinuses (30). The vessels express where they can be sampled by DCs (39). LECs also support the LEC-specific markers such as Lyve-1, Prox-1, or Podoplanin and active migration of tissue-resident DCs into LNs. DC migration drain the cerebrospinal fluid to deep cervical LNs. These findings from tissues to draining LNs via lymphatic vessels is an important provide new insights in the establishment and progression of way to present Ags and activate naïve T cells. DCs enter affer- some neurological diseases involving immune cell contribution, ent lymphatics through preformed portals (40), independent such as multiple sclerosis or Alzheimer’s. Moreover, CNS-resident of integrin-mediated adhesion (41). However, LECs upregulate stromal fibroblastic and endothelial cells were shown to guide adhesion molecules upon inflammation, further favoring DC antiviral CD8+ T cell responses in a model of virus-induced neu- access to lymphatic vessels (42). In addition, expression of CLEC2 roinflammation (31). The production of CCR7 ligands CCL19 (a C-type lectin receptor) by DCs promotes their migration to and CCL21 by CNS stromal cells was found critical for the induc- LNs via lymphatics through interaction with its ligand gp38 tion of viral-specific T cell recruitment and the support of local T (Podoplanin), which is expressed by both LECs and fibroblastic cell reactivation. Whether newly discovered CNS lymphatics (30) reticular cells (FRCs) (43). similarly contribute to neuroinflammatory immunopathologies remains to be determined. Modulation of DC Functions Lymphatic development in the tumor microenvironment, Tissue-resident DCs having acquired peripheral Ags subse- known as tumor lymphangiogenesis, has been extensively stud- quently migrate through afferent lymphatics into LNs in a ied. The participation of tumor lymphatics in the spread of the CCR7-dependent manner. However, the lymphatic system does disease, or metastasis, has been studied for many years. In fact, not only support DC migration from tissues to LNs. Indeed, most human melanomas and carcinomas metastasize through the close interactions between migrating DCs and LECs induce lymphatic system (32). The presence of tumor-associated LECs phenotypic and functional changes in DCs. First, contacts correlates with bad clinical outcome in several types of cancer between TNF-α-stimulated LECs and DCs lead to decreased (33) and therapies aiming the blockade of tumor lymphangiogen- expression of costimulatory molecules by DCs in vitro, thus esis are being considered for treatment of such malignancies (34). impairing DC ability to induce T cell proliferation (44). LEC- Growing evidence highlight the impact of tumor-associated LECs mediated regulation of DC functions is dependent on interac- in dampening antitumor immunity. How interactions between tions between CD11b (Mac-1) on DCs and ICAM-1 on LECs lymphatics and T cells in the context of tumor development will (44). Interestingly, LECs are able to inhibit the function of further alter T cell responses is discussed below. LPS-activated DCs, suggesting once again a regulatory role for LECs in the resolution phase of inflammation. A recent report demonstrated that LECs function as reservoirs of peripheral Ag PRESENTATION INDEPENDENT tissue-restricted Ags (PTAs), which are subsequently acquired IMPACT OF LECs ON PERIPHERAL and presented by DCs to induce T cell anergy, therefore contrib- T CELL RESPONSES uting to peripheral CD4+ T cell tolerance (45).

Hallmarks of T cell immunity include the generation of pathogen- T Cell Homeostasis specific effector responses to confer protection against a large While T cell migration inside LNs is mainly driven by CCL19 and range of invaders, without causing unwanted self-tissue damage. CCL21 produced by FRCs (46), naive and memory T lymphocyte Naïve T cells constantly scan for their cognate Ag. However, given maintenance in SLOs is highly dependent on IL-7. Together with the extremely low frequency of T cells being specific for a particu- FRCs (47), LECs represent an important source of IL-7 in vivo, lar peptide–major histocompatibility (MHC) complex (35, 36), regulating lymphocyte homeostasis and access to SLOs. IL-7-GFP

Frontiers in Immunology | www.frontiersin.org 3 January 2017 | Volume 7 | Article 684 Humbert et al. Role of LECs in Peripheral T Cell Responses knock-in mice exhibit moderate GFP expression in LN-FRCs, Likewise, expression of CCL21 in the tumor promotes immune whereas high levels were detected in both LN LECs and tissue escape and tumor progression (61), which may be explained, at LECs (48, 49). Similarly, LECs were shown to be the major source least in part, by the enhancement of naïve T cell recruitment. of IL-7 in both human and murine LNs (50). Furthermore, Although T cell receptor (TCR)-transgenic tumor-infiltrating LECs not only produce IL-7 but also express the IL-7 receptor naïve T cells may be activated in situ (62), it is unlikely, given chains IL-7Rα and CD132, suggesting a possible role for IL-7 as the immunosuppressive tumor-related environment, that this an autocrine mediator of lymphatic drainage. IL-7-stimulated will lead to fully competent effector T cell differentiation. In this LECs induced lymphangiogenesis in the cornea of mice in vitro, regard, it is still to be demonstrated whether CCL21-producing whereas in IL-7Rα−/− mice, lymphatic drainage was compromised LECs contribute to this effect. How LECs contribute to the overall (51). In addition, IL-7 upregulation by both FRCs and LECs is tolerogenic properties of the tumor microenvironment is still an essential for LN reconstruction and remodeling following viral open question. infection or avascular transplantation (50). This suggests that We have demonstrated that the lymphangiogenic growth factor IL-7 production in LN after resolution of an infection could VEGF-C produced in the tumor promoted immunological toler- be involved in memory T cell homeostasis. Accordingly, IL-7 ance in murine melanoma (63). VEGF-C protected tumors against promotes the development, the proliferation, and the survival of preexisting antitumor immunity and promoted local deletion of memory CD8+ T cells (52, 53). tumor-specific CD8+ T cells (63, 64). Our findings introduce a new role for lymphatics in promoting tumor development and T Cell Egress from LNs suggest that lymphatic endothelium in the local microenviron- T cell egress from LNs is dependent on their expression of the ment may be a novel target for immunomodulation. Supporting sphingosine-1-phosphate (S1P) receptor (S1PR1). Using mice those hypotheses there is a recent publication demonstrating lacking S1P selectively in LECs while maintaining normal blood that following exposure to tumor-derived factors, FRCs of the S1P, Cyster and collaborators have shown that LECs are an tumor-draining LNs adapt on multiple levels to exhibit features in vivo source of S1P in LNs, allowing T cell egress from LNs associated with immunosuppression, such as decreased produc- and PPs (54). S1PR1 expression is downregulated by blood tion of IL-7 and CCL19/21 (65). Whether a similar profound circulating lymphocytes, and upregulated in LNs. Interactions reprograming occurs to LECs in tumor-draining LNs remains to between S1P-producing LECs and S1P1R-expressing T cells be determined. promote LN egress by overcoming retention signals mediated by CCR7 (55, 56). Although steady-state LECs express low levels Ag PRESENTATION-DEPENDENT IMPACT of S1P, its production is upregulated in medullary sinus LECs OF LECs ON PERIPHERAL T CELL upon PAMP/DAMP-mediated inflammation, suggesting that RESPONSES high S1P-expressing LECs can promote effector T cell egress from LNs in pathogenic situations. In contrast, in non-infectious In addition to their ability to modulate T cell responses by sterile inflammatory contexts, low S1P-producing LECs would impacting immune cell migration, interactions, and homeostasis, rather dampen T cell effector functions by favoring T cell reten- LECs can also function as Ag-presenting cells through several tion in LNs. mechanisms and directly influence peripheral T cell outcome.

T Cell Migration in Tumor-Associated Presentation of Endogenously Expressed Lymphatics PTAs to T Cells by LECs Increasing evidence suggest that tumor-associated lymphatics In order to prevent autoimmunity, thymocytes go through a pro- not only simply function as tumor cell transporters but also cess of negative selection, part of the so-called central tolerance, play additional important roles impacting tumor development. allowing the deletion of autoreactive T cell clones before they exit Accordingly, not only metastatic but also primary tumor progres- from the thymus to enter into the periphery [reviewed in Ref. (66, sion can be affected by modulating tumor-associated lymphatic 67)]. In the thymus, medullary thymic epithelial cells (mTECs) expansion. In the context of solid tumors, lymph flow from tumors promiscuously express PTAs, Ag that are normally expressed in is elevated, driving intense interstitial flow in the tumor stroma the periphery (68, 69). The expression of a vast majority of PTAs and increasing lymphatic drainage from the tumor to the drain- in mTECs is regulated by transcription factors (70), including ing LN (57). Combined with a suppressive cytokine environment, the autoimmune regulator (Aire), mutations in Aire leading to it is therefore possible that increased tumor Ags drainage could severe autoimmune disorders (71, 72). PTAs can be either directly promote tumor-specific T cell dysfunction, including anergy and presented by mTECs to the thymocytes, acquired from mTECs apoptosis. In addition, the lymph supports cells migrating from by thymus-resident DCs or acquired in tissues by migrating DCs tissues, in particular CCR7+ DCs, a phenomenon shown to be or plasmacytoid DCs (pDCs), and cross-presented to the thymo- critical for initiating antitumor immune responses (58). cytes (73–76) (Figure 1A). Thymocytes expressing a TCR with Tumor infiltration by T cells is one of the key steps in antitumor a too high affinity for self-Ag/MHC complexes undergo clonal immunity. While cytotoxic T lymphocyte infiltration correlates deletion (73–75). A fraction of the CD4+ thymocytes having a with good prognosis, accumulation of T regulatory cells (Treg) TCR with a high affinity differentiates into thymus-derived Tregs or naïve T cells is detrimental for the clinical outcome (59, 60). (tTregs), previously called natural Tregs (nTregs), and expresses

Frontiers in Immunology | www.frontiersin.org 4 January 2017 | Volume 7 | Article 684 Humbert et al. Role of LECs in Peripheral T Cell Responses

FIGURE 1 | Maintenance of T cell tolerance. (A) Schematic view of thymic central tolerance, reviewed in Ref. (67). After positive selection (not depicted), simple positive (SP) thymocytes undergo a process of negative selection. Thymus-resident conventional dendritic cells (cDCs) and peripheral tissue-restricted antigens (Ags) (PTA) (green)-expressing medullary thymic epithelial cells, as well as peripheral plasmacytoid DCs (pDCs) and cDCs, that have acquired Ag (yellow) in the periphery and migrate to the thymus, present self-peptide major histocompatibility complex (MHC) complexes to SP thymocytes. Thymocytes expressing a T cell receptor (TCR) with high affinity for self (dark colors) are clonally deleted. SP expressing a TCR with intermediate affinity differentiate into thymus-derived T regulatory cell (tTreg) (medium colors). Low-affinity TCR-expressing SP (light colors) exit from the thymus and enter the periphery, however comprising some self-reactive T cells (dark colors) that escaped central tolerance. (B) Peripheral T cell tolerance in the lymph nodes (LNs). References related to lymph node stromal cell contributions are indicated (numbers). Self-Ag-specific T cell tolerance is further maintained in the periphery in LNs. cDCs and pDCs acquire Ag from peripheral tissues (yellow) and migrate to LNs to present Ag to autoreactive T cells. cDCs also acquire Ag expressed by lymphatic endothelial cells (LECs). LECs, fibroblastic reticular cells, and blood endothelial cells present endogenously expressed PTAs (pink), as well as peptide–MHC-II complexes acquired from cDCs, therefore contributing to peripheral T cell tolerance via distinct mechanisms. Extrathymic autoimmune regulator (Aire)-expressing cells (eTACs) present endogenously expressed PTAs. The outcome of Ag presentation by each cell subtype is depicted in the figure. Cell migration and Ag transfer are represented by dotted and dashed arrows, respectively. exo Ags, exogenous antigens; migr. cDC, migratory cDC; pTreg, peripherally induced Treg; thym. cDC, thymus-resident cDC.

Frontiers in Immunology | www.frontiersin.org 5 January 2017 | Volume 7 | Article 684 Humbert et al. Role of LECs in Peripheral T Cell Responses the transcription factor Foxp3 (77). A population of CD8+ Foxp3+ In mTECs, the expression of most, but not all, PTAs is tTregs has also been described (78–81). However, some autore- regulated by Aire (70, 71). In the LN, a rare bone marrow- active—non-Treg—T cells do escape thymic central tolerance derived population was described to express Aire and was mechanisms and reach the periphery (82, 83), as a result from called extrathymic Aire-expressing cells (eTACs). Consequently, either an absence of specific self-Ag presentation in the thymus, eTACs express various PTAs in an Aire-dependent manner, and or a lack of deletion due to a TCR exhibiting an affinity for self- present them through major histocompatibility complex class I Ag/MHC complexes below the negative selection threshold (84) (MHC-I) and MHC-II molecules to induce CD8+ T cell deletion (Figure 1A). (105), and CD4+ T cell anergy (106), respectively (Figure 1B). Therefore, additional mechanisms, called peripheral tolerance On the contrary, PTAs expressed by non-hematopoietic LNSCs, mechanisms, have evolved to maintain T cell tolerance apart including LECs, are not dependent on Aire (103). The regulation from the thymus [reviewed in Ref. (66, 85)]. Cross-tolerance of the expression of the pancreatic self-Ag Ppy by LECs in induction by peripheral DCs has been extensively studied and pancreatic LNs depends on the transcriptional regulator Deaf1, reviewed over the past two decades (86); immature DCs acquire which, together with Aire, belongs to the SAND gene family Ag through the phagocytosis of apoptotic cells in peripheral tis- (107, 108). Interestingly, variant isoforms of Deaf1 in mice and sues to present them to T cells in SLOs (87–89). In the absence of human display an impaired Ppy expression, and were linked costimulatory signals, Ag presentation leads to CD4+ and CD8+ to autoimmune type I diabetes (107). The fact that LNSCs T cell clonal deletion (physical elimination) or anergy (functional do not express Aire may explain the low overlapping PTA inactivation) and/or to the induction of peripherally induced expression in mTECs and LNSCs (109), therefore suggesting Tregs (pTregs), previously called induced Tregs (iTregs) in the a complementary contribution of mTECs and LNSCs in T cell presence of anti-inflammatory factors (77, 90–92). Both resident tolerance induction and maintenance. Future investigations will and migratory DCs, including pDCs, contribute to this process identify other transcription factors, selectively or commonly in the LNs (93–96) (Figure 1B). Nevertheless, emerging evidence expressed by LNSC subsets, which promote different PTA demonstrates that peripheral tolerance does not exclusively rely expression. on DCs. Lymph node stromal cells (LNSCs), and in particular LECs, also play an important role in the maintenance of periph- PTA Presentation by LECs to T Cells eral tolerance (Figure 1B). LNSCs not only endogenously express PTAs but also the direct presentation of PTA-derived peptides in the context of MHC-I PTA-specific Expression by LECs molecules to CD8+ T cells leads to their clonal deletion and The discovery of the ectopic PTA expression by mTECs in the subsequent tolerance induction (97, 98, 101) (Figure 1B). In thymus was the first example that cells of non-hematopoietic the GFAP-HA or iFABP-tOVA models mentioned above, the origin present endogenously expressed self-Ag to T cells (68, 69). lack of presentation of HA or tOVA by enteric stromal cells to Using GFAP-HA or iFABP-tOVA transgenic mouse models, in HA- or tOVA-specific CD8+ T cells was associated with enteric which hemagglutinin (HA) or a truncated form of ovalbumin autoimmunity. Among other LNSC subsets, LECs are involved (tOVA) are expressed as self-Ag in enteric glial cells (EGCs) or in this CD8+ T cell deletional tolerance and are necessary and mature intestinal epithelial cells (IECs), respectively, it was shown sufficient for the induction of peripheral tolerance to some self- few years ago that the EGC-associated HA or IEC-associated Ag, like Tyr, an autoantigen associated with autoimmune vitiligo tOVA proteins were unexpectedly expressed not only by EGCs or (102, 103, 107). These studies show a crucial role for LECs in the IECs but also by CD45-negative stromal cells, in all LNs and not maintenance of peripheral tolerance. exclusively in mesenteric LNs. Those LNSCs were able to process Nevertheless, the ability of LNSCs, and in particular LECs, to endogenously expressed self-proteins into antigenic peptides directly present endogenously expressed PTAs in the context of to directly present these Ag to CD8+ T cells in SLOs, making MHC-II molecules to CD4+ T cells is still a matter of debate, as them functionally similar to mTECs in the thymus (97–100). well as the subsequent impact on CD4+ T cell outcome. We have Moreover, it was shown in non-transgenic mouse models that previously shown that the endogenous expression of MHC-II LNSCs naturally express PTAs and directly present them to molecules is regulated in LECs, BECs, and FRCs by the promoter CD8+ T cells. Among other examples, LNSCs ectopically express IV (pIV) of the master regulator CIITA (110). One study has tyrosinase (tyr), while its expression is normally confined to however demonstrated that the adoptive transfer of HA-specific melanocytes (101). It was later shown that LECs are the only cells TCR transgenic CD4+ T cells (6.5) in GFAP-HA transgenic ectopically expressing this Ag in the LN (102, 103). Indeed, using mice, in which HA is expressed as an autoantigen by EGCs, CD31 and gp38 (Podoplanin) as markers to distinguish the LNSC did not dampen lethal enteric autoimmunity (98). However, as subtypes, it was observed that each subtype expresses a distinct mentioned by the authors, the absence of direct presentation of set of PTAs, with some PTAs exclusively expressed in one specific HA peptide by LNSCs to HA-specific CD4+ T cells in their model LNSC subset and some others redundantly expressed (102, 103) does not rule out a possible upregulation of MHC-II molecules (Figure 1B). This suggests a non-redundant role for the different in LNSCs and a direct presentation under pro-inflammatory LNSC subtypes in the tolerization of various self-specific T cells. conditions (98). Indeed, several studies that will be discussed In addition, the expression of PTAs by LECs is subanatomically later in this review have suggested that LNSCs, among which compartmentalized, with a high expression of PTAs observed LECs, upregulate MHC-II molecules at their surface upon only in LN medullary sinus LECs (104). inflammation (110, 111).

Frontiers in Immunology | www.frontiersin.org 6 January 2017 | Volume 7 | Article 684 Humbert et al. Role of LECs in Peripheral T Cell Responses

For their part, Engelhard and colleagues claim that LECs are were deleted only when PD-L1 was expressed by the non- unable to present endogenously expressed PTAs (β-galactosidase, hematopoietic LN compartment (116). Moreover, among the membrane-bound HA or I-Eα in their models) to CD4+ T cells, LNSC subsets, LECs were the ones expressing the highest level of not related to Ag localization but due to a lack of H2-M expres- PD-L1, with medullary sinuses LECs being the highest express- sion in LECs, which would prevent the loading of peptides onto ers. In addition, LECs do not express costimulatory molecules MHC-II molecules (45). However, this study was carried out in at their surface. The administration of agonistic anti-4-1BB the steady state, whereas LECs, BECs, and FRCs, that express antibodies prevented the deletion of FH CD8+ T cells. The lack IFN-γ inducible-CIITA pIV, might require IFN-γ to upregulate of costimulation through 4-1BB by LECs would lead to PD-1 H-2M molecules, as they do for MHC-II expression, these two upregulation by FH T cells, as Tyr presentation by LECs led to genes being co-regulated by CIITA (112). Moreover, Mebius and a higher expression of PD-1 by FH T cells, an effect that was colleagues observed the presence of mRNA transcripts for H2-M suppressed upon agonistic anti-4-1BB antibody administration. in LECs, among other MHC-II-related molecules (113). This would, in turn, prevent CD25 upregulation, which is neces- Mebius and colleagues identified that in transgenic mice sary for CD8+ T cells survival. Indeed, CD25 expression on FH expressing OVA under the control of the keratin 14 promoter T cells was upregulated only in the presence of agonistic anti-4- (K14mOVA mice), OVA was unexpectedly expressed in LECs. In 1BB or blocking anti-PD-L1 antibodies after Tyr presentation by addition, OVA+ LEC were able to present OVA peptides through LECs (116). Hence, in this model, LECs are responsible for the MHC-II to OTII cells in vitro, leading to an increased Foxp3+ presentation of the endogenously expressed Tyr, which, together OT-II cells Treg homeostasis (113). Using LN transplantation with a combination of a lack of costimulation and a provision of experiments, the authors further suggested that the presentation co-inhibitory signal, leads to Tyr-specific CD8+ T cell deletion of endogenously expressed self-Ag by LNSCs, and especially by (116). The high expression of PD-L1 in LECs is likely regulated LECs, contribute in vivo to the maintenance of Foxp3+ CD4+ by lymphotoxin β receptor (Ltβr), as the treatment of mice with Tregs in the periphery (Figure 1B) (113). Finally, lentiviral anti-Ltβr antibodies led to decreased PD-L1 expression in LECs vectors allowing the selective transduction of MHC-II+ non- (104). Using μMT−/−, CD3ε−/−, and Rag1−/− mice, it was further hematopoietic cells with MHC-II- and MHC-I-restricted HY shown that B cells are required for the expression of the adhe- male-derived epitopes induced T cell hyporesponsiveness/ sion molecule MadCAM-1 at the surface of LECs in the medulla, anergy of HY-specific CD4+ and CD8+ T cells in female mice itself necessary for the expression of PD-L1. On the contrary, (114). Moreover, in Marilyn TCR transgenic mice expressing T cells seemed to suppress PD-L1 expression in LECs through HY-specific CD4+ T cells, increased conversion of effector mechanisms that have not been deciphered yet (104). Finally, CD4+ T cells into CD25+ Foxp3+ pTregs was observed (114). it was suggested that the expression of MHC-II on LECs would Whether these effects were due to a direct Ag presentation of be involved in the induction of CD8+ T cells tolerance to endog- endogenously expressed HY to CD4+ T cells by gp38+ stromal enously expressed self-Ag in LECs by engaging the inhibitory cells, i.e., LECs and FRCs in the LN, remains to be determined. molecule LAG-3. Indeed, after adoptive transfer ofβ -gal-specific Indeed, as stated by the authors, they cannot rule out that other, TCR transgenic CD8+ T cells (Bg1 cells) into Prox-1xβgal mice, non-DC, hematopoietic cell types could contribute to the presen- in which β-gal is selectively expressed by LECs, the proliferation tation of HY Ags, due to undesired transduction and subsequent of Bg1 cells was increased following administration of blocking direct Ag presentation and/or Ag transfer to stromal cells (110, anti-LAG-3 antibodies, which was acting in synergy with anti- 114). Despite a lack of demonstration of direct Ag presentation PD-L1 blocking antibodies (45). by gp38+ stromal cells and the lack of distinction between the We previously showed that high PD-L1 expression by LECs contribution of the different stromal cell subtypes in this model, correlate with their unique ability, compared to other LNSC these data are in accordance with the results of Baptista et al., as subsets, to induce CD4+ T cell apoptosis after presentation of mentioned above (113). DC-acquired peptide–MHC-II complexes (110). Although the molecular mechanisms accounting for the induction of tolerance to MHC-II-restricted self-Ag endogenously expressed Molecular Pathways Involved in and directly presented by LECs to CD4+ T cells have not been LEC-Mediated Peripheral T Cell Tolerance elucidated so far, they are thus likely to involve PD-L1 expression The molecular pathways involved in the clonal deletion of by LECs, as in the case of CD8+ T cells. CD8+ T cells by LNSCs, and in particular by LECs, are not fully elucidated. Using the iFABP-tOVA transgenic mouse model described above, in which tOVA is expressed as a self-Ag in Ag Acquisition and Presentation by the intestinal epithelium, it was shown that the induction of LECs to T Cells CD8+ T cell tolerance requires PD-1:PD-L1 interaction, as the The lymphatic system, by controlling Ag availability, consti- disruption of this pathway leads to severe intestinal enteric tutes one of the first checkpoints for immune responses (100). autoimmune disorder (115). More specifically, in a model of It is not surprising then that LECs, which have early access to adoptive transfer of Tyr-specific TCR transgenic CD8+ T cells any given Ag, display different mechanisms for Ag uptake and (FH T cells) into Tyr-expressing bone marrow chimeric mice, processing (Figure 2). Indeed, recent work revealed that Ag in which either radiosensitive hematopoietic or radioresistant trafficking can be observed at more levels than the classical con- non-hematopoietic cells lacked PD-L1 expression, FH T cells cept of LECs as lymph carriers. Complex interactions between

Frontiers in Immunology | www.frontiersin.org 7 January 2017 | Volume 7 | Article 684 Humbert et al. Role of LECs in Peripheral T Cell Responses

FIGURE 2 | Pathways of Ag acquisition and presentation by LECs. Several pathways of antigen (Ag) acquisition and loading coexist in lymphatic endothelial cells (LECs). Interactions with dendritic cells (DCs) underlie complex mechanisms of Ag transfer in both directions. On one hand, LECs act as Ag reservoirs for DCs which can uptake LEC-derived Ag. The mechanisms accounting for this phenomenon remain however unclear. On the other hand, LECs acquire peptide–MHC-II complexes from DCs in a cell–cell contact dependent manner (DC-derived Ag is depicted in yellow). DC-derived exosomes might also be implicated. Peripheral tissue-restricted Ags (PTA) (in pink) expressed by LEC can be loaded into MHC-I molecules. Intracellular pathways of degradation of such PTAs have however been not investigated. Moreover, whether PTA can be incorporated in MHC-II compartments is still a matter of debate. Alternatively, LECs possess the ability to uptake exogenous lymph-borne and tumor-derived Ag that can be incorporated in MHC-I pathway in a TAP-1-dependent manner. Related references are indicated in numbers.

LECs and DCs (45, 110, 117) depict an exciting picture of Ag LECs upregulate the immunostimulatory molecules HVEM, bidirectional exchange that ultimately may serve to modulate CD48, and MHC-II under such conditions (116), they also the overall magnitude of the immune response (Figure 2). upregulate PD-L1 (102, 110, 119). Pointing at the same direction, Ag cross-presentation by LSECs induces tolerized CD8+ T cells Uptake of Exogenous Ag in the liver. In this context, PD-L1 expression was also relevant It has been extensively demonstrated in several mouse and for such outcome (120). Interestingly, in the absence of inflam- human models that LECs exhibit an active endocytotic capac- mation, surviving LSEC-educated T cells had an Ag-experienced ity (38, 118). They are able to uptake exogenous molecules and, central memory-like phenotype in SLOs (121). Furthermore, depending on their location, process Ag for cross-presentation LSEC-primed memory T cells could be reactivated in vitro and and cross-priming of Ag-specific CD8+ T cells (63, 64) (Figure 2). in vivo in an Ag-specific manner, and they could contribute to a Interestingly, Ag-loaded primary LN LECs were shown to be viral challenge (121). capable of cross-priming Ag-specific CD8+ T cells in a TAP1- The direct contribution of Ag presentation by LECs to dependent manner (64). As described above for endogenous CD4+ viral immunity is still a matter of debate. As mentioned PTA presentation, Ag-loaded LECs induced T cell apoptosis, the above, LECs serve as Ag reservoir during viral infections (117) lack of expression of costimulatory molecules being the most (Figure 2). Nonetheless, genetic ablation of MHC-II in radiore- extended explanation. LECs neither express nor upregulate the sistant stromal cells in LNs resulted in longer maintenance of costimulatory molecules CD40, CD80, and CD86 following TLR Ag-specific CD4+ T cells (122). Specific impact of LN LECs and engagement or in presence of IFN-γ or TNF-α (110, 116). While mechanisms accounting for such effects should be yet clarified.

Frontiers in Immunology | www.frontiersin.org 8 January 2017 | Volume 7 | Article 684 Humbert et al. Role of LECs in Peripheral T Cell Responses

Cellular Ag Transfer dependent on recognition of apoptotic cells or DC phagocytosis. The hallmark of professional APCs is the constitutive cell surface These complementary bidirectional observations highlight the presence of MHC-II and their ability for Ag processing and close relationship and communication between professional presentation (123). Constitutive MHC-II expression is restricted APCs and LECs to enable MHC-II presentation. to a small number of cells of the immune system. Nonetheless, there are many different cell types from both hematopoietic and CONCLUDING REMARKS non-hematopoietic origins that can indeed express MHC-II and interact with CD4+ T cells in the periphery (100, 124, 125). Increasing evidence suggest that lymphatics are much more than As mentioned above, LECs constitute such non-professional simple pipes that drain tissue-derived fluids containing proteins, APC cell types that express MHC-II in an IFN-γ-dependent man- particles, and cells. Through the expression of different surface ner. Indeed, MHC-II expression in LN LECs has been reported at molecules and the production of soluble factors, LECs indeed both transcriptional and protein expression levels (102, 110, 111). modulate immune responses in many ways, including the active By using transgenic mouse models lacking the different CIITA regulation of cellular migration, interactions, and functions. promoters, we have previously demonstrated that steady-state Recent studies have highlighted a possible role for LECs as levels of MHC-II molecules on the surface of LECs and other direct instructors of T cell immunity. Indeed, the discovery that stromal subsets in LNs reflect a combination of IFN-γ-inducible LNSCs, including LECs, ectopically express tissue-derived Ags, basal activity and acquired peptide:MHC-II complexes from a feature thought to be restricted to mTECs and thymic central DCs (110). The acquired MHC-II molecules were loaded with T cell tolerance, has pushed forward LECs to potentially func- DC-derived Ags, licensing LECs to induce anergy and increased tion as Ag-presenting cells. Accordingly, the selective expression cell death Ag-specific CD4+ T cells (Figures 1B and 2). Lack of model Ags in LECs leads to an Ag-specific recognition by of measurable productive T cell responses has been one of the T cells, which, after an early step of activation and proliferation, major difficulties preventing the clarification of the impact of are either inactivated or deleted. Therefore, the presentation of Ag presentation by LECs on CD4+ T cell outcome. As for CD8+ endogenously expressed Ags by LECs seems to contribute to T cell responses, the absence of costimulatory signals, such as peripheral T cell tolerance. Studies have also suggested that LECs CD80 or CD86 and the constitutive expression of PD-L1 by LECs, acquire exogenous Ags by distinct pathways, including direct preclude the possibility of functional effector CD4+ T cell prim- uptake, or cell-membrane transfer, and present them to induce ing. In this regard, it has been shown that human LN-derived T cell dysfunction. The molecular mechanisms contributing LECs fail to induce allogeneic CD4+ T cell proliferation even after to LEC ability to inactivate T cells are still not fully elucidated. IFN-γ stimulation (119). In these particular in vitro settings, LECs However, a consensus candidate, PD-L1, the ligand for program- were unable to induce proliferation of either naïve or memory cell death 1 receptor expressed by T cells, emerged from several CD4+ T cells. recent studies to be highly expressed by LECs, and important to Membrane exchange between cells is not uncommon in immu- mediate T cell tolerance. Although pioneering studies suggest nology (126). Peptide:MHC-I and MHC-II complexes have been that Ag-presenting LNSCs are sufficient to maintain peripheral shown to be transferred between DC and tumor cells (127) or T cell tolerance, the specific contribution of LECs remains to be infected cells (128), as well as between DCs (129). Ag transfer can addressed. Likewise, substantial differences among LECs from occur as peptide exchange on cell surfaces. Peptide epitopes can distinct anatomical locations entail different functions. Specific bind directly on cell surface or early endosomal MHC molecules roles of local LECs should be carefully dissected in order to fully (130), where both MHC-I and MHC-II are receptive for lymph- understand how they differentially impact T cell responses. In borne peptide binding. This might be particularly relevant in the addition, most studies so far have been performed in steady state, context of self-tolerance, since recent analyses showed that the and the contribution of Ag presentation by LECs under different human lymph peptidome contains predominantly self-peptides, pathological conditions in shaping of peripheral T cell responses including products derived from extracellular processing of remains to be determined. In addition, future studies will assess proteins (131). Exosomes were also implicated in the transfer of how current therapies for cancer or autoimmune diseases aiming peptide:MHC-II complexes from DCs to LNSCs (110), and they at modulating immune cell functions, specifically alter the ability cannot be excluded to contribute to alternative Ag trafficking of LECs to impact T cell responses. (Figure 2). Antigen transfer between LECs and DCs is, however, not AUTHOR CONTRIBUTIONS restricted to one direction. Indeed, the transfer of PTAs specifically expressed in LECs to hematopoietic cells has been described SH, JD, and MH have developed the concept, wrote the (45) (Figure 2). Neither membrane-bound nor cytoplasmic PTAs manuscript, prepared the figures, and critically read, revised, and were directly presented by LECs to prime Ag-specific CD4+ T cell approved the manuscript. responses. As mentioned above, this was attributed to the lower expression of H2-M in LECs compared to professional APCs, FUNDING which is required for peptide binding into the MHC-II groove. Instead, peptides derived from PTAs expressed by LECs were SH’s laboratory is supported by the Swiss National Science found to be loaded onto MHC-II in DCs (45). While the exchange Foundation (310030_166541), the European Research Council mechanism is still open to examination, it was reported not to be (281365), and the Swiss MS Society.

Frontiers in Immunology | www.frontiersin.org 9 January 2017 | Volume 7 | Article 684 Humbert et al. Role of LECs in Peripheral T Cell Responses

REFERENCES 23. Baluk P, Fuxe J, Hashizume H, Romano T, Lashnits E, Butz S, et al. Functionally specialized junctions between endothelial cells of lymphatic vessels. J Exp 1. Asellius G. De Lactibus Sive Lacteis Venis. Milan: Mediolani (1627). Med (2007) 204(10):2349–62. doi:10.1084/jem.20062596 2. Witte MH, Bernas MJ, Martin CP, Witte CL. Lymphangiogenesis and 24. Tammela T, Saaristo A, Holopainen T, Lyytikkä J, Kotronen A, Pitkonen M, lymphangiodysplasia: from molecular to clinical lymphology. Microsc Res et al. Therapeutic differentiation and maturation of lymphatic vessels after Tech (2001) 55(2):122–45. doi:10.1002/jemt.1163.abs lymph node dissection and transplantation. Nat Med (2007) 13(12):1458–66. 3. Nicenboim J, Malkinson G, Lupo T, Asaf L, Sela Y, Mayseless O, et al. doi:10.1038/nm1689 Lymphatic vessels arise from specialized angioblasts within a venous niche. 25. Friedlaender MH, Chisari FV, Baer H. The role of the inflammatory response Nature (2015) 522(7554):56–61. doi:10.1038/nature14425 of skin and lymph nodes in the induction of sensitization to simple chemicals. 4. Choi I, Lee S, Hong YK. The new era of the lymphatic system: no longer J Immunol (1973) 111(1):164–70. secondary to the blood vascular system. Cold Spring Harb Perspect Med 26. Thomas SN, Rutkowski JM, Pasquier M, Kuan EL, Alitalo K, Randolph GJ, (2012) 2(4):a006445. doi:10.1101/cshperspect.a006445 et al. Impaired humoral immunity and tolerance in K14-VEGFR-3-Ig mice 5. Huntington GSM, McClure CFW. The anatomy and development of the that lack dermal lymphatic drainage. J Immunol (2012) 189(5):2181–90. jugular lymph sac in the domestic cat (Felis domestica). Am J Anat (1910) doi:10.4049/jimmunol.1103545 (10):177–311. 27. Wisse E. An electron microscopic study of the fenestrated endothelial lining 6. Sabin FR. On the origin of the lymphatic system from the veins and the of rat liver sinusoids. J Ultrastruct Res (1970) 31(1):125–50. doi:10.1016/ development of the lymph hearts and thoracic duct in the pig. Am J Anat S0022-5320(70)90150-4 (1902) 1:367–691. doi:10.1002/aja.1000010310 28. Limmer A, Ohl J, Kurts C, Ljunggren HG, Reiss Y, Groettrup M, et al. Efficient 7. Sabin FR. On the development of the superficial lymphatics in the skin of the presentation of exogenous antigen by liver endothelial cells to CD8+ T cells pig. Am J Anat (1904) 3(2):183–95. doi:10.1002/aja.1000030205 results in antigen-specific T-cell tolerance. Nat Med (2000) 6(12):1348–54. 8. Kaipainen A, Korhonen J, Mustonen T, van Hinsbergh VW, Fang GH, doi:10.1038/82161 Dumont D, et al. Expression of the FMS-like tyrosine kinase 4 gene becomes 29. Mohammad MG, Tsai VW, Ruitenberg MJ, Hassanpour M, Li H, Hart PH, restricted to lymphatic endothelium during development. Proc Natl Acad Sci et al. Immune cell trafficking from the brain maintains CNS immune toler- U S A (1995) 92(8):3566–70. doi:10.1073/pnas.92.8.3566 ance. J Clin Invest (2014) 124(3):1228–41. doi:10.1172/JCI71544 9. Oliver G. Lymphatic vasculature development. Nat Rev Immunol (2004) 30. Louveau A, Smirnov I, Keyes TJ, Eccles JD, Rouhani SJ, Peske JD, et al. 4(1):35–45. doi:10.1038/nri1258 Structural and functional features of central nervous system lymphatic 10. Wigle JT, Harvey N, Detmar M, Lagutina I, Grosveld G, Gunn MD, et al. vessels. Nature (2015) 523(7560):337–41. doi:10.1038/nature14432 An essential role for Prox1 in the induction of the lymphatic endothelial cell 31. Cupovic J, Onder L, Gil-Cruz C, Weiler E, Caviezel-Firner S, Perez-Shibayama phenotype. EMBO J (2002) 21(7):1505–13. doi:10.1093/emboj/21.7.1505 C, et al. Central Nervous system stromal cells control local CD8(+) T cell 11. Dumont DJ, Jussila L, Taipale J, Lymboussaki A, Mustonen T, Pajusola K, responses during virus-induced neuroinflammation. Immunity (2016) et al. Cardiovascular failure in mouse embryos deficient in VEGF receptor-3. 44(3):622–33. doi:10.1016/j.immuni.2015.12.022 Science (1998) 282(5390):946–9. doi:10.1126/science.282.5390.946 32. Leong SP, Nakakura EK, Pollock R, Choti MA, Morton DL, Henner WD, 12. Kukk E, Lymboussaki A, Taira S, Kaipainen A, Jeltsch M, Joukov V, et al. et al. Unique patterns of metastases in common and rare types of malignancy. VEGF-C receptor binding and pattern of expression with VEGFR-3 J Surg Oncol (2011) 103(6):607–14. doi:10.1002/jso.21841 suggests a role in lymphatic vascular development. Development (1996) 33. Stacker SA, Williams SP, Karnezis T, Shayan R, Fox SB, Achen MG. 122(12):3829–37. Lymphangiogenesis and lymphatic vessel remodelling in cancer. Nat Rev 13. Saaristo A, Veikkola T, Tammela T, Enholm B, Karkkainen MJ, Pajusola Cancer (2014) 14(3):159–72. doi:10.1038/nrc3677 K, et al. Lymphangiogenic gene therapy with minimal blood vascular side 34. Dieterich LC, Detmar M. Tumor lymphangiogenesis and new drug effects. J Exp Med (2002) 196(6):719–30. doi:10.1084/jem.20020587 development. Adv Drug Deliv Rev (2016) 99(Pt B):148–60. doi:10.1016/ 14. Jeltsch M, Kaipainen A, Joukov V, Meng X, Lakso M, Rauvala H, et al. j.addr.2015.12.011 Hyperplasia of lymphatic vessels in VEGF-C transgenic mice. Science (1997) 35. Moon JJ, Chu HH, Pepper M, McSorley SJ, Jameson SC, Kedl RM, et al. 276(5317):1423–5. doi:10.1126/science.276.5317.1423 Naive CD4(+) T cell frequency varies for different epitopes and predicts 15. Wigle JT, Oliver G. Prox1 function is required for the development of repertoire diversity and response magnitude. Immunity (2007) 27(2):203–13. the murine lymphatic system. Cell (1999) 98(6):769–78. doi:10.1016/ doi:10.1016/j.immuni.2007.07.007 S0092-8674(00)81511-1 36. Obar JJ, Khanna KM, Lefrancois L. Endogenous naive CD8+ T cell precursor 16. Petrova TV, Mäkinen T, Mäkelä TP, Saarela J, Virtanen I, Ferrell RE, et al. frequency regulates primary and memory responses to infection. Immunity Lymphatic endothelial reprogramming of vascular endothelial cells by the (2008) 28(6):859–69. doi:10.1016/j.immuni.2008.04.010 Prox-1 homeobox transcription factor. EMBO J (2002) 21(17):4593–9. 37. Roozendaal R, Mempel TR, Pitcher LA, Gonzalez SF, Verschoor A, doi:10.1093/emboj/cdf470 Mebius RE, et al. Conduits mediate transport of low-molecular-weight 17. Hirakawa S, Hong YK, Harvey N, Schacht V, Matsuda K, Libermann T, et al. antigen to lymph node follicles. Immunity (2009) 30(2):264–76. doi:10.1016/ Identification of vascular lineage-specific genes by transcriptional profiling j.immuni.2008.12.014 of isolated blood vascular and lymphatic endothelial cells. Am J Pathol (2003) 38. Sixt M, Kanazawa N, Selg M, Samson T, Roos G, Reinhardt DP, et al. The 162(2):575–86. doi:10.1016/S0002-9440(10)63851-5 conduit system transports soluble antigens from the afferent lymph to 18. Podgrabinska S, Braun P, Velasco P, Kloos B, Pepper MS, Skobe M. Molecular resident dendritic cells in the T cell area of the lymph node. Immunity (2005) characterization of lymphatic endothelial cells. Proc Natl Acad Sci U S A 22(1):19–29. doi:10.1016/j.immuni.2004.11.013 (2002) 99(25):16069–74. doi:10.1073/pnas.242401399 39. Gerner MY, Torabi-Parizi P, Germain RN. Strategically localized 19. Johnson NC, Dillard ME, Baluk P, McDonald DM, Harvey NL, Frase SL, dendritic cells promote rapid T cell responses to lymph-borne partic- et al. Lymphatic endothelial cell identity is reversible and its maintenance ulate antigens. Immunity (2015) 42(1):172–85. doi:10.1016/j.immuni. requires Prox1 activity. Genes Dev (2008) 22(23):3282–91. doi:10.1101/ 2014.12.024 gad.1727208 40. Pflicke H, Sixt M. Preformed portals facilitate dendritic cell entry into 20. Yaniv K, Isogai S, Castranova D, Dye L, Hitomi J, Weinstein BM, et al. afferent lymphatic vessels. J Exp Med (2009) 206(13):2925–35. doi:10.1084/ Live imaging of lymphatic development in the zebrafish. Nat Med (2006) jem.20091739 12(6):711–6. doi:10.1038/nm1427 41. Lämmermann T, Bader BL, Monkley SJ, Worbs T, Wedlich-Söldner R, 21. Klotz L, Norman S, Vieira JM, Masters M, Rohling M, Dubé KN, et al. Cardiac Hirsch K, et al. Rapid leukocyte migration by integrin-independent lymphatics are heterogeneous in origin and respond to injury. Nature (2015) flowing and squeezing. Nature (2008) 453(7191):51–5. doi:10.1038/ 522(7554):62–7. doi:10.1038/nature14483 nature06887 22. Alitalo K, Tammela T, Petrova TV. Lymphangiogenesis in development and 42. Johnson LA, Clasper S, Holt AP, Lalor PF, Baban D, Jackson DG. An inflam- human disease. Nature (2005) 438(7070):946–53. doi:10.1038/nature04480 mation-induced mechanism for leukocyte transmigration across lymphatic

Frontiers in Immunology | www.frontiersin.org 10 January 2017 | Volume 7 | Article 684 Humbert et al. Role of LECs in Peripheral T Cell Responses

vessel endothelium. J Exp Med (2006) 203(12):2763–77. doi:10.1084/ 61. Shields JD, Kourtis IC, Tomei AA, Roberts JM, Swartz MA. Induction jem.20051759 of lymphoidlike stroma and immune escape by tumors that express 43. Acton SE, Astarita JL, Malhotra D, Lukacs-Kornek V, Franz B, Hess PR, et al. the chemokine CCL21. Science (2010) 328(5979):749–52. doi:10.1126/ Podoplanin-rich stromal networks induce dendritic cell motility via activa- science.1185837 tion of the C-type lectin receptor CLEC-2. Immunity (2012) 37(2):276–89. 62. Thompson ED, Enriquez HL, Fu YX, Engelhard VH. Tumor masses support doi:10.1016/j.immuni.2012.05.022 naive T cell infiltration, activation, and differentiation into effectors. J Exp 44. Podgrabinska S, Kamalu O, Mayer L, Shimaoka M, Snoeck H, Randolph GJ, Med (2010) 207(8):1791–804. doi:10.1084/jem.20092454 et al. Inflamed lymphatic endothelium suppresses dendritic cell maturation 63. Lund AW, Duraes FV, Hirosue S, Raghavan VR, Nembrini C, and function via Mac-1/ICAM-1-dependent mechanism. J Immunol (2009) Thomas SN, et al. VEGF-C promotes immune tolerance in B16 melanomas 183(3):1767–79. doi:10.4049/jimmunol.0802167 and cross-presentation of tumor antigen by lymph node lymphatics. Cell Rep 45. Rouhani SJ, Eccles JD, Riccardi P, Peske JD, Tewalt EF, Cohen JN, et al. (2012) 1(3):191–9. doi:10.1016/j.celrep.2012.01.005 Roles of lymphatic endothelial cells expressing peripheral tissue antigens in 64. Hirosue S, Vokali E, Raghavan VR, Rincon-Restrepo M, Lund AW, CD4 T-cell tolerance induction. Nat Commun (2015) 6:6771. doi:10.1038/ Corthésy-Henrioud P, et al. Steady-state antigen scavenging, cross- ncomms7771 presentation, and CD8+ T cell priming: a new role for lymphatic endothelial 46. Luther SA, Bidgol A, Hargreaves DC, Schmidt A, Xu Y, Paniyadi J, et al. cells. J Immunol (2014) 192(11):5002–11. doi:10.4049/jimmunol.1302492 Differing activities of homeostatic chemokines CCL19, CCL21, and 65. Riedel A, Shorthouse D, Haas L, Hall BA, Shields J. Tumor-induced stromal CXCL12 in lymphocyte and dendritic cell recruitment and lymphoid reprogramming drives lymph node transformation. Nat Immunol (2016) neogenesis. J Immunol (2002) 169(1):424–33. doi:10.4049/jimmunol. 17(9):1118–27. doi:10.1038/ni.3492 169.1.424 66. Xing Y, Hogquist KA. T-cell tolerance: central and peripheral. Cold Spring 47. Link A, Vogt TK, Favre S, Britschgi MR, Acha-Orbea H, Hinz B, et al. Harb Perspect Biol (2012) 4(6). doi:10.1101/cshperspect.a006957 Fibroblastic reticular cells in lymph nodes regulate the homeostasis of naive 67. Klein L, Kyewski B, Allen PM, Hogquist KA. Positive and negative selection T cells. Nat Immunol (2007) 8(11):1255–65. doi:10.1038/ni1513 of the T cell repertoire: what thymocytes see (and don’t see). Nat Rev Immunol 48. Hara T, Shitara S, Imai K, Miyachi H, Kitano S, Yao H, et al. Identification (2014) 14(6):377–91. doi:10.1038/nri3667 of IL-7-producing cells in primary and secondary lymphoid organs using 68. Kyewski B, Rottinger B, Klein L. Making central T-cell tolerance efficient: IL-7-GFP knock-in mice. J Immunol (2012) 189(4):1577–84. doi:10.4049/ thymic stromal cells sample distinct self-antigen pools. Curr Top Microbiol jimmunol.1200586 Immunol (2000) 251:139–45. 49. Miller CN, Hartigan-O’Connor DJ, Lee MS, Laidlaw G, Cornelissen IP, 69. Derbinski J, Schulte A, Kyewski B, Klein L. Promiscuous gene expression in Matloubian M, et al. IL-7 production in murine lymphatic endothelial cells medullary thymic epithelial cells mirrors the peripheral self. Nat Immunol and induction in the setting of peripheral lymphopenia. Int Immunol (2013) (2001) 2(11):1032–9. doi:10.1038/ni723 25(8):471–83. doi:10.1093/intimm/dxt012 70. Derbinski J, Gäbler J, Brors B, Tierling S, Jonnakuty S, Hergenhahn M, 50. Onder L, Narang P, Scandella E, Chai Q, Iolyeva M, Hoorweg K, et al. IL-7- et al. Promiscuous gene expression in thymic epithelial cells is regulated at producing stromal cells are critical for lymph node remodeling. Blood (2012) multiple levels. J Exp Med (2005) 202(1):33–45. doi:10.1084/jem.20050471 120(24):4675–83. doi:10.1182/blood-2012-03-416859 71. Anderson MS, Venanzi ES, Klein L, Chen Z, Berzins SP, Turley SJ, et al. 51. Iolyeva M, Aebischer D, Proulx ST, Willrodt AH, Ecoiffier T, Häner S, Projection of an immunological self shadow within the thymus by the Aire et al. Interleukin-7 is produced by afferent lymphatic vessels and supports protein. Science (2002) 298(5597):1395–401. doi:10.1126/science.1075958 lymphatic drainage. Blood (2013) 122(13):2271–81. doi:10.1182/blood- 72. Liston A, Gray DH, Lesage S, Fletcher AL, Wilson J, Webster KE, et al. 2013-01-478073 Gene dosage – limiting role of Aire in thymic expression, clonal deletion, 52. Schluns KS, Kieper WC, Jameson SC, Lefrancois L. Interleukin-7 mediates and organ-specific autoimmunity. J Exp Med (2004) 200(8):1015–26. the homeostasis of naive and memory CD8 T cells in vivo. Nat Immunol doi:10.1084/jem.20040581 (2000) 1(5):426–32. doi:10.1038/80868 73. Gallegos AM, Bevan MJ. Central tolerance to tissue-specific antigens 53. Schluns KS, Lefrancois L. Cytokine control of memory T-cell develop- mediated by direct and indirect antigen presentation. J Exp Med (2004) ment and survival. Nat Rev Immunol (2003) 3(4):269–79. doi:10.1038/ 200(8):1039–49. doi:10.1084/jem.20041457 nri1052 74. Koble C, Kyewski B. The thymic medulla: a unique microenvironment 54. Pham TH, Baluk P, Xu Y, Grigorova I, Bankovich AJ, Pappu R, et al. Lymphatic for intercellular self-antigen transfer. J Exp Med (2009) 206(7):1505–13. endothelial cell sphingosine kinase activity is required for lymphocyte egress doi:10.1084/jem.20082449 and lymphatic patterning. J Exp Med (2010) 207(1):17–27. doi:10.1084/ 75. Bonasio R, Scimone ML, Schaerli P, Grabie N, Lichtman AH, jem.20091619 von Andrian UH. Clonal deletion of thymocytes by circulating dendritic 55. Grigorova IL, Schwab SR, Phan TG, Pham TH, Okada T, Cyster JG. Cortical cells homing to the thymus. Nat Immunol (2006) 7(10):1092–100. sinus probing, S1P1-dependent entry and flow-based capture of egressing doi:10.1038/ni1106-1234b T cells. Nat Immunol (2009) 10(1):58–65. doi:10.1038/ni.1682 76. Hadeiba H, Lahl K, Edalati A, Oderup C, Habtezion A, Pachynski R, et al. 56. Pham TH, Okada T, Matloubian M, Lo CG, Cyster JG. S1P1 receptor Plasmacytoid dendritic cells transport peripheral antigens to the thymus signaling overrides retention mediated by G alpha I-coupled receptors to promote central tolerance. Immunity (2012) 36(3):438–50. doi:10.1016/ to promote T cell egress. Immunity (2008) 28(1):122–33. doi:10.1016/ j.immuni.2012.01.017 j.immuni.2007.11.017 77. Abbas AK, Benoist C, Bluestone JA, Campbell DJ, Ghosh S, Hori S, et al. 57. Swartz MA, Lund AW. Lymphatic and interstitial flow in the tumour Regulatory T cells: recommendations to simplify the nomenclature. Nat microenvironment: linking mechanobiology with immunity. Nat Rev Cancer Immunol (2013) 14(4):307–8. doi:10.1038/ni.2554 (2012) 12(3):210–9. doi:10.1038/nrc3186 78. Cosmi L, Liotta F, Lazzeri E, Francalanci M, Angeli R, Mazzinghi B, et al. 58. Roberts EW, Broz ML, Binnewies M, Headley MB, Nelson AE, Wolf DM, Human CD8+CD25+ thymocytes share phenotypic and functional features et al. Critical role for CD103(+)/CD141(+) dendritic cells bearing CCR7 with CD4+CD25+ regulatory thymocytes. Blood (2003) 102(12):4107–14. for tumor antigen trafficking and priming of T cell immunity in melanoma. doi:10.1182/blood-2003-04-1320 Cancer Cell (2016) 30(2):324–36. doi:10.1016/j.ccell.2016.06.003 79. Ménager-Marcq I, Pomié C, Romagnoli P, van Meerwijk JP. CD8+CD28- 59. Galon J, Costes A, Sanchez-Cabo F, Kirilovsky A, Mlecnik B, Lagorce- regulatory T lymphocytes prevent experimental inflammatory bowel Pages C, et al. Type, density, and location of immune cells within human disease in mice. Gastroenterology (2006) 131(6):1775–85. doi:10.1053/ colorectal tumors predict clinical outcome. Science (2006) 313(5795):1960–4. j.gastro.2006.09.008 doi:10.1126/science.1129139 80. Vuddamalay Y, Attia M, Vicente R, Pomié C, Enault G, Leobon B, et al. 60. Fridman WH, Pagès F, Sautès-Fridman C, Galon J. The immune contexture Mouse and human CD8(+) CD28(low) regulatory T lymphocytes differ- in human tumours: impact on clinical outcome. Nat Rev Cancer (2012) entiate in the thymus. Immunology (2016) 148(2):187–96. doi:10.1111/ 12(4):298–306. doi:10.1038/nrc3245 imm.12600

Frontiers in Immunology | www.frontiersin.org 11 January 2017 | Volume 7 | Article 684 Humbert et al. Role of LECs in Peripheral T Cell Responses

81. Pomie C, Menager-Marcq I, van Meerwijk JP. Murine CD8+ regulatory 101. Nichols LA, Chen Y, Colella TA, Bennett CL, Clausen BE, Engelhard VH. T lymphocytes: the new era. Hum Immunol (2008) 69(11):708–14. Deletional self-tolerance to a melanocyte/melanoma antigen derived doi:10.1016/j.humimm.2008.08.288 from tyrosinase is mediated by a radio-resistant cell in peripheral and 82. Lohse AW, Dinkelmann M, Kimmig M, Herkel J, Meyer zum Büschenfelde mesenteric lymph nodes. J Immunol (2007) 179(2):993–1003. doi:10.4049/ KH. Estimation of the frequency of self-reactive T cells in health and jimmunol.179.2.993 inflammatory diseases by limiting dilution analysis and single cell cloning. 102. Fletcher AL, Lukacs-Kornek V, Reynoso ED, Pinner SE, Bellemare-Pelletier J Autoimmun (1996) 9(5):667–75. doi:10.1006/jaut.1996.0087 A, Curry MS, et al. Lymph node fibroblastic reticular cells directly present 83. Anderson AC, Nicholson LB, Legge KL, Turchin V, Zaghouani H, Kuchroo peripheral tissue antigen under steady-state and inflammatory conditions. VK. High frequency of autoreactive myelin proteolipid protein-specific J Exp Med (2010) 207(4):689–97. doi:10.1084/jem.20092642 T cells in the periphery of naive mice: mechanisms of selection of the 103. Cohen JN, Guidi CJ, Tewalt EF, Qiao H, Rouhani SJ, Ruddell A, et al. Lymph self-reactive repertoire. J Exp Med (2000) 191(5):761–70. doi:10.1084/ node-resident lymphatic endothelial cells mediate peripheral tolerance via jem.191.5.761 Aire-independent direct antigen presentation. J Exp Med (2010) 207(4): 84. Enouz S, Carrié L, Merkler D, Bevan MJ, Zehn D. Autoreactive T cells 681–8. doi:10.1084/jem.20092465 bypass negative selection and respond to self-antigen stimulation during 104. Cohen JN, Tewalt EF, Rouhani SJ, Buonomo EL, Bruce AN, Xu X, et al. infection. J Exp Med (2012) 209(10):1769–79. doi:10.1084/jem.20120905 Tolerogenic properties of lymphatic endothelial cells are controlled by the 85. Walker LS, Abbas AK. The enemy within: keeping self-reactive T cells at lymph node microenvironment. PLoS One (2014) 9(2):e87740. doi:10.1371/ bay in the periphery. Nat Rev Immunol (2002) 2(1):11–9. doi:10.1038/ journal.pone.0087740 nri701 105. Gardner JM, Devoss JJ, Friedman RS, Wong DJ, Tan YX, Zhou X, et al. 86. Steinman RM, Turley S, Mellman I, Inaba K. The induction of tolerance Deletional tolerance mediated by extrathymic Aire-expressing cells. Science by dendritic cells that have captured apoptotic cells. J Exp Med (2000) (2008) 321(5890):843–7. doi:10.1126/science.1159407 191(3):411–6. doi:10.1084/jem.191.3.411 106. Gardner JM, Metzger TC, McMahon EJ, Au-Yeung BB, Krawisz AK, Lu W, 87. Kurts C. Cross-presentation: inducing CD8 T cell immunity and tolerance. et al. Extrathymic Aire-expressing cells are a distinct bone marrow-derived J Mol Med (2000) 78(6):326–32. doi:10.1007/s001090000108 population that induce functional inactivation of CD4(+) T cells. Immunity 88. Adler AJ, Marsh DW, Yochum GS, Guzzo JL, Nigam A, Nelson WG, et al. (2013) 39(3):560–72. doi:10.1016/j.immuni.2013.08.005 CD4+ T cell tolerance to parenchymal self-antigens requires presentation 107. Yip L, Su L, Sheng D, Chang P, Atkinson M, Czesak M, et al. Deaf1 iso- by bone marrow-derived antigen-presenting cells. J Exp Med (1998) forms control the expression of genes encoding peripheral tissue antigens 187(10):1555–64. doi:10.1084/jem.187.10.1555 in the pancreatic lymph nodes during type 1 diabetes. Nat Immunol (2009) 89. Heath WR, Kurts C, Miller JF, Carbone FR. Cross-tolerance: a pathway 10(9):1026–33. doi:10.1038/ni.1773 for inducing tolerance to peripheral tissue antigens. J Exp Med (1998) 108. Gibson TJ, Ramu C, Gemund C, Aasland R. The APECED polyglandular 187(10):1549–53. doi:10.1084/jem.187.10.1549 autoimmune syndrome protein, AIRE-1, contains the SAND domain and 90. Steinman RM, Hawiger D, Nussenzweig MC. Tolerogenic dendritic is probably a transcription factor. Trends Biochem Sci (1998) 23(7):242–4. cells. Annu Rev Immunol (2003) 21:685–711. doi:10.1146/annurev. doi:10.1016/S0968-0004(98)01231-6 immunol.21.120601.141040 109. Metzger TC, Anderson MS. Control of central and peripheral tolerance 91. Blander JM, Medzhitov R. Toll-dependent selection of microbial antigens for by Aire. Immunol Rev (2011) 241(1):89–103. doi:10.1111/j.1600-065X. presentation by dendritic cells. Nature (2006) 440(7085):808–12. doi:10.1038/ 2011.01008.x nature04596 110. Dubrot J, Duraes FV, Potin L, Capotosti F, Brighouse D, Suter T, et al. Lymph 92. Wilson NS, El-Sukkari D, Belz GT, Smith CM, Steptoe RJ, Heath WR, node stromal cells acquire peptide-MHCII complexes from dendritic cells et al. Most lymphoid organ dendritic cell types are phenotypically and and induce antigen-specific CD4(+) T cell tolerance. J Exp Med (2014) functionally immature. Blood (2003) 102(6):2187–94. doi:10.1182/blood- 211(6):1153–66. doi:10.1084/jem.20132000 2003-02-0513 111. Malhotra D, Fletcher AL, Astarita J, Lukacs-Kornek V, Tayalia P, Gonzalez 93. Allan RS, Smith CM, Belz GT, van Lint AL, Wakim LM, Heath WR, et al. SF, et al. Transcriptional profiling of stroma from inflamed and resting lymph Epidermal viral immunity induced by CD8alpha+ dendritic cells but nodes defines immunological hallmarks.Nat Immunol (2012) 13(5):499–510. not by Langerhans cells. Science (2003) 301(5641):1925–8. doi:10.1126/ doi:10.1038/ni.2262 science.1087576 112. Reith W, LeibundGut-Landmann S, Waldburger JM. Regulation of MHC 94. Allan RS, Waithman J, Bedoui S, Jones CM, Villadangos JA, Zhan Y, et al. class II gene expression by the class II transactivator. Nat Rev Immunol (2005) Migratory dendritic cells transfer antigen to a lymph node-resident dendritic 5(10):793–806. doi:10.1038/nri1708 cell population for efficient CTL priming. Immunity (2006) 25(1):153–62. 113. Baptista AP, Roozendaal R, Reijmers RM, Koning JJ, Unger WW, Greuter doi:10.1016/j.immuni.2006.04.017 M, et al. Lymph node stromal cells constrain immunity via MHC class II 95. Lukacs-Kornek V, Burgdorf S, Diehl L, Specht S, Kornek M, Kurts C. The self-antigen presentation. Elife (2014) 3. doi:10.7554/eLife.04433 kidney-renal lymph node-system contributes to cross-tolerance against 114. Ciré S, Da Rocha S, Ferrand M, Collins MK, Galy A. In vivo gene delivery to innocuous circulating antigen. J Immunol (2008) 180(2):706–15. doi:10.4049/ lymph node stromal cells leads to transgene-specific CD8+ T cell anergy in jimmunol.180.2.706 mice. Mol Ther (2016) 24(11):1965–73. doi:10.1038/mt.2016.168 96. Guery L, Hugues S. Tolerogenic and activatory plasmacytoid dendritic 115. Reynoso ED, Elpek KG, Francisco L, Bronson R, Bellemare-Pelletier A, cells in autoimmunity. Front Immunol (2013) 4:59. doi:10.3389/fimmu. Sharpe AH, et al. Intestinal tolerance is converted to autoimmune enteritis 2013.00059 upon PD-1 ligand blockade. J Immunol (2009) 182(4):2102–12. doi:10.4049/ 97. Lee JW, Epardaud M, Sun J, Becker JE, Cheng AC, Yonekura AR, et al. jimmunol.0802769 Peripheral antigen display by lymph node stroma promotes T cell tolerance 116. Tewalt EF, Cohen JN, Rouhani SJ, Guidi CJ, Qiao H, Fahl SP, et al. Lymphatic to intestinal self. Nat Immunol (2007) 8(2):181–90. doi:10.1038/ni1427 endothelial cells induce tolerance via PD-L1 and lack of costimulation leading 98. Magnusson FC, Liblau RS, von Boehmer H, Pittet MJ, Lee JW, Turley SJ, et al. to high-level PD-1 expression on CD8 T cells. Blood (2012) 120(24):4772–82. Direct presentation of antigen by lymph node stromal cells protects against doi:10.1182/blood-2012-04-427013 CD8 T-cell-mediated intestinal autoimmunity. Gastroenterology (2008) 117. Tamburini BA, Burchill MA, Kedl RM. Antigen capture and archiving by 134(4):1028–37. doi:10.1053/j.gastro.2008.01.070 lymphatic endothelial cells following vaccination or viral infection. Nat 99. Collier AY, Lee JW, Turley SJ. Self-encounters of the third kind: lymph node Commun (2014) 5:3989. doi:10.1038/ncomms4989 stroma promotes tolerance to peripheral tissue antigens. Mucosal Immunol 118. Fruhwürth S, Pavelka M, Bittman R, Kovacs WJ, Walter KM, Röhrl C, et al. (2008) 1(4):248–51. doi:10.1038/mi.2008.19 High-density lipoprotein endocytosis in endothelial cells. World J Biol Chem 100. Hirosue S, Dubrot J. Modes of antigen presentation by lymph node stromal (2013) 4(4):131–40. doi:10.4331/wjbc.v4.i4.131 cells and their immunological implications. Front Immunol (2015) 6:446. 119. Nörder M, Gutierrez MG, Zicari S, Cervi E, Caruso A, Guzmán CA. Lymph doi:10.3389/fimmu.2015.00446 node-derived lymphatic endothelial cells express functional costimulatory

Frontiers in Immunology | www.frontiersin.org 12 January 2017 | Volume 7 | Article 684 Humbert et al. Role of LECs in Peripheral T Cell Responses

molecules and impair dendritic cell-induced allogenic T-cell proliferation. 127. Zhang QJ, Li XL, Wang D, Huang XC, Mathis JM, Duan WM, et al. FASEB J (2012) 26(7):2835–46. doi:10.1096/fj.12-205278 Trogocytosis of MHC-I/peptide complexes derived from tumors and infected 120. Diehl L, Schurich A, Grochtmann R, Hegenbarth S, Chen L, Knolle PA. cells enhances dendritic cell cross-priming and promotes adaptive T cell Tolerogenic maturation of liver sinusoidal endothelial cells promotes responses. PLoS One (2008) 3(8):e3097. doi:10.1371/journal.pone.0003097 B7-homolog 1-dependent CD8+ T cell tolerance. Hepatology (2008) 128. Wakim LM, Bevan MJ. Cross-dressed dendritic cells drive memory CD8+ 47(1):296–305. doi:10.1002/hep.21965 T-cell activation after viral infection. Nature (2011) 471(7340):629–32. 121. Böttcher JP, Schanz O, Wohlleber D, Abdullah Z, Debey-Pascher S, doi:10.1038/nature09863 Staratschek-Jox A, et al. Liver-primed memory T cells generated under non- 129. de Heusch M, Blocklet D, Egrise D, Hauquier B, Vermeersch M, Goldman inflammatory conditions provide anti-infectious immunity. Cell Rep (2013) S, et al. Bidirectional MHC molecule exchange between migratory and 3(3):779–95. doi:10.1016/j.celrep.2013.02.008 resident dendritic cells. J Leukoc Biol (2007) 82(4):861–8. doi:10.1189/ 122. Abe J, Shichino S, Ueha S, Hashimoto S, Tomura M, Inagaki Y, et al. jlb.0307167 Lymph node stromal cells negatively regulate antigen-specific CD4+ 130. Griffin JP, Chu R, Harding CV. Early endosomes and a late endocytic com- T cell responses. J Immunol (2014) 193(4):1636–44. doi:10.4049/jimmunol. partment generate different peptide-class II MHC complexes via distinct 1302946 processing mechanisms. J Immunol (1997) 158(4):1523–32. 123. Roche PA, Furuta K. The ins and outs of MHC class II-mediated antigen 131. Clement CC, Santambrogio L. The lymph self-antigen repertoire. Front processing and presentation. Nat Rev Immunol (2015) 15(4):203–16. Immunol (2013) 4:424. doi:10.3389/fimmu.2013.00424 doi:10.1038/nri3818 124. Duraes FV, Thelemann C, Sarter K, Acha-Orbea H, Hugues S, Reith W. Conflict of Interest Statement: The authors declare that the research was con- Role of major histocompatibility complex class II expression by non- ducted in the absence of any commercial or financial relationships that could be hematopoietic cells in autoimmune and inflammatory disorders: facts and construed as a potential conflict of interest. fiction. Tissue Antigens (2013) 82(1):1–15. doi:10.1111/tan.12136 125. Kambayashi T, Laufer TM. Atypical MHC class II-expressing antigen- Copyright © 2017 Humbert, Hugues and Dubrot. This is an open-access article presenting cells: can anything replace a dendritic cell? Nat Rev Immunol distributed under the terms of the Creative Commons Attribution License (CC BY). (2014) 14(11):719–30. doi:10.1038/nri3754 The use, distribution or reproduction in other forums is permitted, provided the 126. Davis DM. Intercellular transfer of cell-surface proteins is common and original author(s) or licensor are credited and that the original publication in this can affect many stages of an immune response. Nat Rev Immunol (2007) journal is cited, in accordance with accepted academic practice. No use, distribution 7(3):238–43. doi:10.1038/nri2020 or reproduction is permitted which does not comply with these terms.

Frontiers in Immunology | www.frontiersin.org 13 January 2017 | Volume 7 | Article 684

VI.2. Appendix 2

Warming up the tumor microenvironment in order to enhance immunogenicity

Marion Humbert 1 and Stéphanie Hugues 1

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

Published in Oncoimmunology (DOI: 10.1080/2162402X.2018.1510710)

Objective: In our article “Intratumoral CpG-B promotes antitumoral neutrophil, cDC and T cell cooperation without reprogramming tolerogenic pDC” (Chapter A: Humbert et al., Cancer Res, 2018), we demonstrated that intratumoral delivery of CpG-B enhances anti-tumor immunity and leads to tumor growth control. In addition, we showed that the administration of this TLR9 agonist induces a tolerogenic-to-immunogenic conversion of the tumor microenvironment, which could be beneficial for cancer immunotherapies currently used in the clinic. For instance, this strategy enhances innate and adaptive anti-tumor immune responses. The aim of this author’s view was to discuss our findings and put them in the context of currently used therapies, with a particular focus on local treatment administration (as opposed to systemic), the use of TLR ligand agonists, and the potential of combining different immunotherapeutic anti-cancer approaches.

Personal contribution: For this review, I participated in the development of the concept, in the manuscript redaction, and I prepared Figure 1, under the supervision of Prof. Stéphanie Hugues.

180

ONCOIMMUNOLOGY https://doi.org/10.1080/2162402X.2018.1510710

AUTHOR’S VIEW Warming up the tumor microenvironment in order to enhance immunogenicity Marion Humbert and Stephanie Hugues Department of Pathology and Immunology, University of Geneva Medical School, Gemeva, Switzerland

ABSTRACT ARTICLE HISTORY We have recently demonstrated that intratumoral CpG-B vaccination enhances anti-tumor immunity and Received 27 July 2018 tumor regression in mice. We further show that the local delivery of TLR9 agonists converts the Accepted 28 July 2018 tolerogenic tumor microenvironment into an immunopermissive one, which may benefit current immu- KEYWORDS notherapeutic anticancer strategies by enhancing innate and adaptive tumor-associated immune cell tumor microenvironment; responses. local immunotherapy; CpG-B

Treatments for malignancies, including immunotherapies, are recruited to the tumor, and cytotoxic T cells (CTL) induce generally provided systemically, therefore frequently asso- tumor cell death.4 Therefore, pDCs represent good candidates ciated with severe toxic effects. Locally administered immu- to boost the antitumor immune response. However, whether notherapy – topical, intradermal, intranodal or intratumoral – the tolerogenic phenotype of TA-pDCs can be reverted by is nowadays considered as a promising approach. It may intratumoral administration of TLR9 ligands, and how the actually solve several aspects limiting immunotherapeutic TME is impacted by this approach, is still unclear. strategies: local treatments necessitate lower doses to reach We recently explored how intratumoral delivery of CpG-B the required concentration, leading to reduced systemic drug alone or together with a tumor antigen affects tumor-infiltrat- levels, therefore limiting the risks of autoimmune adverse ing leukocytes in mice having developed solid tumors.5 Using effects while enhancing the robustness of the anti-tumor melanoma or thymoma tumors, we observed a significant immune response.1 reduction of the tumor growth following intratumoral The aim of immunotherapy is to induce a vigorous immune CpG-B administration. However, we did not observe any response against tumor antigens. The tumor microenvironment tolerogenic-to-immunogenic conversion of TA-pDC pheno- (TME) however negatively impacts the immune system through type after treatment, and the genetic depletion of pDCs did three main mechanisms, namely immunosuppression, immu- not affect the efficacy of the treatment. Therefore, in our noevasion and immunoediting.2 These mechanisms involve can- model, TA-pDCs are refractory to TLR9 stimulation and are cer and mesenchymal cells, as well as immune cells, such as not involved in the mechanism of action leading to tumor myeloid-derived suppressor cells, regulatory T cells (Tregs), regression. In contrast, upon intratumoral CpG-B delivery, tumor-associated (TA-) macrophages and TA-neutrophils, some changes occur in the TME, inducing an increased pro- which produce anti-inflammatory cytokines.2 In this regard, duction of neutrophil attractants, by either tumor cells or local immunotherapy aims at overcoming the inhibitory proper- tumor-infiltrating cells. As a consequence, neutrophils are ties of the TME in order to “warm up” the tumor, enhance its rapidly recruited into the tumor following local CpG-B injec- immunogenicity and generate effector immune cells. tion, and promote the activation of conventional DCs (cDCs) One conceivable local immunotherapeutic approach is the (Figure 1). The role of tumor-associated neutrophils is administration of TLR ligand agonists, in order to reactivate debated in the literature; these cells can have either pro- or immune cells at the injection site. In many types of human anti-tumoral properties depending on different parameters, cancers, the recruitment of plasmacytoid dendritic cells (pDCs) such as the type or the stage of the tumor.6 to the TME contributes to the induction of immune tolerance. Following intratumoral CpG-B injection, activated cDCs Indeed, pDCs infiltrating tumors are characterized by reduced migrate from the tumor bed to the draining lymph nodes to expression of costimulatory molecules, a decreased ability to activate tumor-specific cytotoxic T lymphocytes (CTLs) and produce type I IFN (IFN-I), as well as an enhanced ability to Th1 cells (Figure 1). Effector cells subsequently infiltrate solid induce Tregs.3 We have previously shown that, following con- tumors, which also contain reduced numbers of Tregs. tralateral vaccination of mice with CpG oligonucleotides-B Therefore, the balance effector T cells/Treg is modulated in (CpG-B), a TLR9 agonist, together with a tumor antigen, favour of effector T cells following local immunotherapy and pDCs can be activated at a site distal from the tumor and results in tumor cell elimination and tumor regression used as immunogenic antigen-presenting cells (APCs) that (Figure 1). Importantly, all these events are dependent on promote tumor-specific Th17 cells. Immune cells are further the primary recruitment of neutrophils after CpG-B

CONTACT Stephanie Hugues [email protected] Color versions of one or more of the figures in the article can be found online at www.tandfonline.com/koni. © 2018 The Author(s). Published by Taylor & Francis This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License (http://creativecommons.org/licenses/by-nc-nd/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited, and is not altered, transformed, or built upon in any way. e1510710-2 M. HUMBERT AND S. HUGUES

Figure 1. Intratumoral delivery of tumor antigenic peptide and CpG-B induces tumor rejection: working model. 1) Intratumoral injection of tumor antigenic peptide and CpG-B 2) Activated neutrophils are recruited into the tumor 3) Neutrophils promote the activation of conventional dendritic cells (cDCs) 4) Activated cDCs migrate to the tumor-draining lymph nodes (TdLNs) 5) Activated cDCs prime cytotoxic T cells (CTLs) and Th1 cells in the TdLNs 6) Effector Th1 and CTLs migrate to the tumor and secrete IFN-γ 7) Treg numbers are decreased in the tumor 8) Tumor cells are killed administration, since both cDC activation and tumor-specific volume), currently restrain the development of localized T cell priming are abrogated upon neutrophil depletion.5 immunotherapeutic approaches. However, in situ vaccination Interestingly, the efficacy of CpG-B immunotherapy was strategies are achievable and can induce an immune response potentialized by the addition of a tumor antigen, suggesting and tumor regression while reducing systemic toxic events. that when an immunogenic tumor antigen is available for Recently, in melanoma patients, a single local injection of presentation, in situ TLR9 vaccination may induce the conver- low-dose CpG-B, at the tumor excision site resulted in sion of dysfunctional into immunogenic antigen-presenting lymph node cDC activation and protection against relapse.10 cells and generate efficient antitumor T-cell responses. Future work deciphering the mechanisms involved in the Local TLR9 delivery did not induce a robust systemic protection observed after intratumoral CpG-B treatment will antitumoral immune response in our models, with a rather improve anticancer therapies, for example by combining dif- modest inhibition of contralateral tumor growth. This is con- ferent approaches such as checkpoint inhibitors or techniques sistent with a recent study showing that intratumoral CpG-C inducing the release of tumor antigens. does not impact the growth of distal tumors, unless it is combined with anti-OX40 agonistic antibodies.7 In human, systemic tumor-specific CD8+ T cell responses have been Funding observed in tumor patients treated with CpG-B combined to This work was supported by the European Research Council [281365]; 8 low dose radiotherapy. Therefore, preclinical and early clin- Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen ical studies propose that intratumoral therapies (in situ vacci- Forschung [PP00P3_152951]; Schweizerischer Nationalfonds zur nation or local radiation), could convert “cold” TME into Förderung der Wissenschaftlichen Forschung [310030_166541]. “hot” TME, thus enhancing the potential for a response after its combination with systemic immunotherapies, such as References immune checkpoint inhibitors. For instance, sialic acid sugars, which promote an immunosuppressive TME, can be blocked 1. Fransen MF, Ossendorp F, Arens R, Melief CJ. Local immuno- to induce an immune-permissive TME, suitable for combina- modulation for cancer therapy: providing treatment where tion with immune-based interventions.9 needed. Oncoimmunology. 2013;2:e26493. 2. Zitvogel L, Tesniere A, Kroemer G. Cancer despite immunosur- Some limitations, such as tumor accessibility or the need veillance: immunoselection and immunosubversion. Nat Reviews for an optimization of the delivery methods (route, dose, Immunol. 2006;6:715–727. ONCOIMMUNOLOGY e1510710-3

3. Le Mercier I, Poujol D, Sanlaville A, Sisirak V, Gobert M, Durand 8.BrodyJD,AiWZ,CzerwinskiDK,TorchiaJA,LevyM, I, Dubois, B, Treilleux, I, Marvel, J, Vlach, J, et al. Tumor promo- AdvaniRH,Kim,YH,Hoppe,RT,Knox,SJ,Shin,LK,etal. tion by intratumoral plasmacytoid dendritic cells is reversed by In situ vaccination with a TLR9 agonist induces systemic TLR7 ligand treatment. Cancer Res. 2013;73:4629–4640. lymphoma regression: a phase I/II study. J Clinical 4. Guery L, Dubrot J, Lippens C, Brighouse D, Malinge P, Irla M, Oncology: Official Journal Am Soc Clin Oncol. Pot, C, Reith, W, Waldburger, J-M, Hugues, S. Ag-presenting 2010;28:4324–4332. CpG-activated pDCs prime Th17 cells that induce tumor regres- 9. Bull C, Boltje TJ, Balneger N, Weischer SM, Wassink M, van sion. Cancer Res. 2014;74:6430–6440. Gemst JJ, Bloemendal, VR, Boon, L, van der Vlag, J, Heise, T, 5. Humbert M, Guery L, Brighouse D, Lemeille S, Hugues S. et al. Sialic acid blockade suppresses tumor growth by enhan- Intratumoral CpG-B promotes antitumoral neutrophil, cDC, and cing T-cell-mediated tumor immunity. Cancer Res. T-cell cooperation without reprograming tolerogenic pDC. 2018;78:3574–3588. Cancer Res. 2018;78:3280–3292. 10. Koster BD, van den Hout M, Sluijter BJR, Molenkamp BG, 6. Nicolas-Avila JA, Adrover JM, Hidalgo A. Neutrophils Vuylsteke R, Baars A, van Leeuwen, Paul A M, Scheper, RJ, in homeostasis, immunity, and cancer. Immunity. 2017;46:15–28. Petrousjka van den Tol, M, van den Eertwegh, Alfons JM, et al. 7. Sagiv-Barfi I, Czerwinski DK, Levy S, Alam IS, Mayer AT, Local adjuvant treatment with low-dose CpG-B offers durable Gambhir SS, Levy, R. Eradication of spontaneous malignancy by protection against disease recurrence in clinical stage I-II local immunotherapy. Sci Transl Med. 2018;10(426). eaan4488. Melanoma: data from two randomized phase II trials. Clin doi: 10.1126/scitranslmed.aan4488. Cancer Res. 2017;23:5679–5686.

VII. REFERENCES

Abbas, A.K., Benoist, C., Bluestone, J.A., Campbell, D.J., Ghosh, S., Hori, S., Jiang, S., Kuchroo, V.K., Mathis, D., Roncarolo, M.G., et al. (2013). Regulatory T cells: recommendations to simplify the nomenclature. Nature immunology 14, 307-308.

Abe, J., Shichino, S., Ueha, S., Hashimoto, S., Tomura, M., Inagaki, Y., Stein, J.V., and Matsushima, K. (2014). Lymph node stromal cells negatively regulate antigen-specific CD4+ T cell responses. Journal of immunology 193, 1636- 1644.

Abi Abdallah, D.S., Egan, C.E., Butcher, B.A., and Denkers, E.Y. (2011). Mouse neutrophils are professional antigen- presenting cells programmed to instruct Th1 and Th17 T-cell differentiation. International immunology 23, 317-326.

Absinta, M., Ha, S.K., Nair, G., Sati, P., Luciano, N.J., Palisoc, M., Louveau, A., Zaghloul, K.A., Pittaluga, S., Kipnis, J., et al. (2017). Human and nonhuman primate meninges harbor lymphatic vessels that can be visualized noninvasively by MRI. eLife 6.

Acton, S.E., Astarita, J.L., Malhotra, D., Lukacs-Kornek, V., Franz, B., Hess, P.R., Jakus, Z., Kuligowski, M., Fletcher, A.L., Elpek, K.G., et al. (2012). Podoplanin-rich stromal networks induce dendritic cell motility via activation of the C-type lectin receptor CLEC-2. Immunity 37, 276-289.

Acton, S.E., and Reis e Sousa, C. (2016). Dendritic cells in remodeling of lymph nodes during immune responses. Immunological reviews 271, 221-229.

Adiko, A.C., Babdor, J., Gutierrez-Martinez, E., Guermonprez, P., and Saveanu, L. (2015). Intracellular Transport Routes for MHC I and Their Relevance for Antigen Cross-Presentation. Frontiers in immunology 6, 335.

Adler, A.J., Marsh, D.W., Yochum, G.S., Guzzo, J.L., Nigam, A., Nelson, W.G., and Pardoll, D.M. (1998). CD4+ T cell tolerance to parenchymal self-antigens requires presentation by bone marrow-derived antigen-presenting cells. The Journal of experimental medicine 187, 1555-1564.

Adorini, L. (2003). Tolerogenic dendritic cells induced by vitamin D receptor ligands enhance regulatory T cells inhibiting autoimmune diabetes. Annals of the New York Academy of Sciences 987, 258-261.

Adorini, L., and Penna, G. (2009). Dendritic cell tolerogenicity: a key mechanism in immunomodulation by vitamin D receptor agonists. Human immunology 70, 345-352.

Al-Soudi, A., Kaaij, M.H., and Tas, S.W. (2017). Endothelial cells: From innocent bystanders to active participants in immune responses. Autoimmunity reviews 16, 951-962.

Alcantara-Hernandez, M., Leylek, R., Wagar, L.E., Engleman, E.G., Keler, T., Marinkovich, M.P., Davis, M.M., Nolan, G.P., and Idoyaga, J. (2017). High-Dimensional Phenotypic Mapping of Human Dendritic Cells Reveals Interindividual Variation and Tissue Specialization. Immunity 47, 1037-1050 e1036.

Alculumbre, S., Raieli, S., Hoffmann, C., Chelbi, R., Danlos, F.X., and Soumelis, V. (2018a). Plasmacytoid pre-dendritic cells (pDC): from molecular pathways to function and disease association. Seminars in cell & developmental biology.

Alculumbre, S.G., Saint-Andre, V., Di Domizio, J., Vargas, P., Sirven, P., Bost, P., Maurin, M., Maiuri, P., Wery, M., Roman, M.S., et al. (2018b). Diversification of human plasmacytoid predendritic cells in response to a single stimulus. Nature immunology 19, 63-75.

Alitalo, K., Tammela, T., and Petrova, T.V. (2005). Lymphangiogenesis in development and human disease. Nature 438, 946-953.

Allan, R.S., Smith, C.M., Belz, G.T., van Lint, A.L., Wakim, L.M., Heath, W.R., and Carbone, F.R. (2003). Epidermal viral immunity induced by CD8alpha+ dendritic cells but not by Langerhans cells. Science 301, 1925-1928.

Allan, R.S., Waithman, J., Bedoui, S., Jones, C.M., Villadangos, J.A., Zhan, Y., Lew, A.M., Shortman, K., Heath, W.R., and Carbone, F.R. (2006). Migratory dendritic cells transfer antigen to a lymph node-resident dendritic cell population for efficient CTL priming. Immunity 25, 153-162.

Alonso, A., Cook, S.D., Maghzi, A.H., and Divani, A.A. (2011). A case-control study of risk factors for multiple sclerosis in Iran. Multiple sclerosis 17, 550-555.

184

Anderson, A.C., Nicholson, L.B., Legge, K.L., Turchin, V., Zaghouani, H., and Kuchroo, V.K. (2000). High frequency of autoreactive myelin proteolipid protein-specific T cells in the periphery of naive mice: mechanisms of selection of the self-reactive repertoire. The Journal of experimental medicine 191, 761-770.

Anderson, D.A., 3rd, Murphy, K.M., and Briseno, C.G. (2018). Development, Diversity, and Function of Dendritic Cells in Mouse and Human. Cold Spring Harbor perspectives in biology 10.

Anderson, G., and Takahama, Y. (2012). Thymic epithelial cells: working class heroes for T cell development and repertoire selection. Trends in immunology 33, 256-263.

Anderson, M.S., and Su, M.A. (2016). AIRE expands: new roles in immune tolerance and beyond. Nature reviews Immunology 16, 247-258.

Anderson, M.S., Venanzi, E.S., Klein, L., Chen, Z., Berzins, S.P., Turley, S.J., von Boehmer, H., Bronson, R., Dierich, A., Benoist, C., et al. (2002). Projection of an immunological self shadow within the thymus by the aire protein. Science 298, 1395-1401.

Antel, J., Antel, S., Caramanos, Z., Arnold, D.L., and Kuhlmann, T. (2012). Primary progressive multiple sclerosis: part of the MS disease spectrum or separate disease entity? Acta neuropathologica 123, 627-638.

Antila, S., Karaman, S., Nurmi, H., Airavaara, M., Voutilainen, M.H., Mathivet, T., Chilov, D., Li, Z., Koppinen, T., Park, J.H., et al. (2017). Development and plasticity of meningeal lymphatic vessels. The Journal of experimental medicine 214, 3645-3667.

Appelbe, O.K., Moynihan, K.D., Flor, A., Rymut, N., Irvine, D.J., and Kron, S.J. (2017). Radiation-enhanced delivery of systemically administered amphiphilic-CpG oligodeoxynucleotide. Journal of controlled release : official journal of the Controlled Release Society 266, 248-255.

Appelberg, R. (1992). CD4+ T cells are required for antigen-specific recruitment of neutrophils by BCG-immune spleen cells. Immunology 75, 414-419.

Araujo, E.F., Medeiros, D.H., Galdino, N.A., Condino-Neto, A., Calich, V.L., and Loures, F.V. (2016). Tolerogenic Plasmacytoid Dendritic Cells Control Paracoccidioides brasiliensis Infection by Inducting Regulatory T Cells in an IDO-Dependent Manner. PLoS pathogens 12, e1006115.

Ascherio, A., and Munger, K.L. (2007). Environmental risk factors for multiple sclerosis. Part I: the role of infection. Annals of neurology 61, 288-299.

Ascherio, A., Munger, K.L., and Lunemann, J.D. (2012). The initiation and prevention of multiple sclerosis. Nature reviews Neurology 8, 602-612.

Aspelund, A., Antila, S., Proulx, S.T., Karlsen, T.V., Karaman, S., Detmar, M., Wiig, H., and Alitalo, K. (2015). A dural lymphatic vascular system that drains brain interstitial fluid and macromolecules. The Journal of experimental medicine 212, 991-999.

Aspord, C., Charles, J., Leccia, M.T., Laurin, D., Richard, M.J., Chaperot, L., and Plumas, J. (2010). A novel cancer vaccine strategy based on HLA-A*0201 matched allogeneic plasmacytoid dendritic cells. PloS one 5, e10458.

Aspord, C., Leccia, M.T., Charles, J., and Plumas, J. (2013). Plasmacytoid dendritic cells support melanoma progression by promoting Th2 and regulatory immunity through OX40L and ICOSL. Cancer immunology research 1, 402-415.

Aspord, C., Leccia, M.T., Charles, J., and Plumas, J. (2014a). Melanoma hijacks plasmacytoid dendritic cells to promote its own progression. Oncoimmunology 3, e27402.

Aspord, C., Leccia, M.T., Salameire, D., Laurin, D., Chaperot, L., Charles, J., and Plumas, J. (2012). HLA-A(*)0201(+) plasmacytoid dendritic cells provide a cell-based immunotherapy for melanoma patients. The Journal of investigative dermatology 132, 2395-2406.

Aspord, C., Leloup, C., Reche, S., and Plumas, J. (2014b). pDCs efficiently process synthetic long peptides to induce functional virus- and tumour-specific T-cell responses. European journal of immunology 44, 2880-2892.

Aspord, C., Tramcourt, L., Leloup, C., Molens, J.P., Leccia, M.T., Charles, J., and Plumas, J. (2014c). Imiquimod inhibits melanoma development by promoting pDC cytotoxic functions and impeding tumor vascularization. The Journal of investigative dermatology 134, 2551-2561.

185

Asselin-Paturel, C., Boonstra, A., Dalod, M., Durand, I., Yessaad, N., Dezutter-Dambuyant, C., Vicari, A., O'Garra, A., Biron, C., Briere, F., et al. (2001). Mouse type I IFN-producing cells are immature APCs with plasmacytoid morphology. Nature immunology 2, 1144-1150.

Audiger, C., Rahman, M.J., Yun, T.J., Tarbell, K.V., and Lesage, S. (2017). The Importance of Dendritic Cells in Maintaining Immune Tolerance. Journal of immunology 198, 2223-2231.

Auf, G., Carpentier, A.F., Chen, L., Le Clanche, C., and Delattre, J.Y. (2001). Implication of macrophages in tumor rejection induced by CpG-oligodeoxynucleotides without antigen. Clinical cancer research : an official journal of the American Association for Cancer Research 7, 3540-3543.

Axisa, P.P., and Hafler, D.A. (2016). Multiple sclerosis: genetics, biomarkers, treatments. Current opinion in neurology 29, 345-353.

Bacchetta, R., Gregori, S., and Roncarolo, M.G. (2005). CD4+ regulatory T cells: mechanisms of induction and effector function. Autoimmunity reviews 4, 491-496.

Bach, J.F. (2018). The hygiene hypothesis in autoimmunity: the role of pathogens and commensals. Nature reviews Immunology 18, 105-120.

Bahmani, B., Uehara, M., Jiang, L., Ordikhani, F., Banouni, N., Ichimura, T., Solhjou, Z., Furtmuller, G.J., Brandacher, G., Alvarez, D., et al. (2018). Targeted delivery of immune therapeutics to lymph nodes prolongs cardiac allograft survival. The Journal of clinical investigation 128, 4770-4786.

Baldwin, T.A., and Hogquist, K.A. (2007). Transcriptional analysis of clonal deletion in vivo. Journal of immunology 179, 837-844.

Balkwill, F.R., Capasso, M., and Hagemann, T. (2012). The tumor microenvironment at a glance. Journal of cell science 125, 5591-5596.

Baluk, P., Fuxe, J., Hashizume, H., Romano, T., Lashnits, E., Butz, S., Vestweber, D., Corada, M., Molendini, C., Dejana, E., et al. (2007). Functionally specialized junctions between endothelial cells of lymphatic vessels. The Journal of experimental medicine 204, 2349-2362.

Baptista, A.P., Roozendaal, R., Reijmers, R.M., Koning, J.J., Unger, W.W., Greuter, M., Keuning, E.D., Molenaar, R., Goverse, G., Sneeboer, M.M., et al. (2014). Lymph node stromal cells constrain immunity via MHC class II self- antigen presentation. eLife 3.

Bar-Or, A. (2016). Multiple sclerosis and related disorders: evolving pathophysiologic insights. The Lancet Neurology 15, 9-11.

Bar-Or, A., Pachner, A., Menguy-Vacheron, F., Kaplan, J., and Wiendl, H. (2014). Teriflunomide and its mechanism of action in multiple sclerosis. Drugs 74, 659-674.

Baron, J.L., Madri, J.A., Ruddle, N.H., Hashim, G., and Janeway, C.A., Jr. (1993). Surface expression of alpha 4 integrin by CD4 T cells is required for their entry into brain parenchyma. The Journal of experimental medicine 177, 57-68.

Barthlott, T., Bosch, A.J., Berkemeier, C., Nogales-Cadenas, R., Jeker, L.T., Keller, M.P., Pascual-Montano, A., and Hollander, G.A. (2015). A subpopulation of CD103(pos) ICOS(pos) Treg cells occurs at high frequency in lymphopenic mice and represents a lymph node specific differentiation stage. European journal of immunology 45, 1760-1771.

Bartholomaus, I., Kawakami, N., Odoardi, F., Schlager, C., Miljkovic, D., Ellwart, J.W., Klinkert, W.E., Flugel-Koch, C., Issekutz, T.B., Wekerle, H., et al. (2009). Effector T cell interactions with meningeal vascular structures in nascent autoimmune CNS lesions. Nature 462, 94-98.

Battaglia, M., Gregori, S., Bacchetta, R., and Roncarolo, M.G. (2006). Tr1 cells: from discovery to their clinical application. Seminars in immunology 18, 120-127.

Bauer, S., and Wagner, H. (2002). Bacterial CpG-DNA licenses TLR9. Current topics in microbiology and immunology 270, 145-154.

Baxter, A.G. (2007). The origin and application of experimental autoimmune encephalomyelitis. Nature reviews Immunology 7, 904-912.

186

Beauvillain, C., Delneste, Y., Scotet, M., Peres, A., Gascan, H., Guermonprez, P., Barnaba, V., and Jeannin, P. (2007). Neutrophils efficiently cross-prime naive T cells in vivo. Blood 110, 2965-2973.

Becher, B. (2015). The Good, the Bad, or the Pretty: IL-17 Builds Lymphoid Tissues in the Brain. Immunity 43, 1033- 1034.

Becher, B., and Greter, M. (2012). Acquitting an APC: DCs found "not guilty" after trial by ablation. European journal of immunology 42, 2551-2554.

Becher, B., Spath, S., and Goverman, J. (2017). Cytokine networks in neuroinflammation. Nature reviews Immunology 17, 49-59.

Beckebaum, S., Zhang, X., Chen, X., Yu, Z., Frilling, A., Dworacki, G., Grosse-Wilde, H., Broelsch, C.E., Gerken, G., and Cicinnati, V.R. (2004). Increased levels of interleukin-10 in serum from patients with hepatocellular carcinoma correlate with profound numerical deficiencies and immature phenotype of circulating dendritic cell subsets. Clinical cancer research : an official journal of the American Association for Cancer Research 10, 7260-7269.

Benitez-Ribas, D., Adema, G.J., Winkels, G., Klasen, I.S., Punt, C.J., Figdor, C.G., and de Vries, I.J. (2006). Plasmacytoid dendritic cells of melanoma patients present exogenous proteins to CD4+ T cells after Fc gamma RII-mediated uptake. The Journal of experimental medicine 203, 1629-1635.

Bennouna, S., Bliss, S.K., Curiel, T.J., and Denkers, E.Y. (2003). Cross-talk in the innate immune system: neutrophils instruct recruitment and activation of dendritic cells during microbial infection. Journal of immunology 171, 6052- 6058.

Benvenuti, F. (2016). The Dendritic Cell Synapse: A Life Dedicated to T Cell Activation. Frontiers in immunology 7, 70.

Bettelli, E., Pagany, M., Weiner, H.L., Linington, C., Sobel, R.A., and Kuchroo, V.K. (2003). Myelin oligodendrocyte glycoprotein-specific T cell receptor transgenic mice develop spontaneous autoimmune optic neuritis. The Journal of experimental medicine 197, 1073-1081.

Bettelli, E., Sullivan, B., Szabo, S.J., Sobel, R.A., Glimcher, L.H., and Kuchroo, V.K. (2004). Loss of T-bet, but not STAT1, prevents the development of experimental autoimmune encephalomyelitis. The Journal of experimental medicine 200, 79-87.

Binnewies, M., Roberts, E.W., Kersten, K., Chan, V., Fearon, D.F., Merad, M., Coussens, L.M., Gabrilovich, D.I., Ostrand-Rosenberg, S., Hedrick, C.C., et al. (2018). Understanding the tumor immune microenvironment (TIME) for effective therapy. Nature medicine 24, 541-550.

Birnberg, T., Bar-On, L., Sapoznikov, A., Caton, M.L., Cervantes-Barragan, L., Makia, D., Krauthgamer, R., Brenner, O., Ludewig, B., Brockschnieder, D., et al. (2008). Lack of conventional dendritic cells is compatible with normal development and T cell homeostasis, but causes myeloid proliferative syndrome. Immunity 29, 986-997.

Blander, J.M., and Medzhitov, R. (2006). Toll-dependent selection of microbial antigens for presentation by dendritic cells. Nature 440, 808-812.

Blasius, A.L., Giurisato, E., Cella, M., Schreiber, R.D., Shaw, A.S., and Colonna, M. (2006). Bone marrow stromal cell antigen 2 is a specific marker of type I IFN-producing cells in the naive mouse, but a promiscuous cell surface antigen following IFN stimulation. Journal of immunology 177, 3260-3265.

Boasso, A., Herbeuval, J.P., Hardy, A.W., Anderson, S.A., Dolan, M.J., Fuchs, D., and Shearer, G.M. (2007). HIV inhibits CD4+ T-cell proliferation by inducing indoleamine 2,3-dioxygenase in plasmacytoid dendritic cells. Blood 109, 3351- 3359.

Bonaccorsi, I., Morandi, B., Antsiferova, O., Costa, G., Oliveri, D., Conte, R., Pezzino, G., Vermiglio, G., Anastasi, G.P., Navarra, G., et al. (2014). Membrane transfer from tumor cells overcomes deficient phagocytic ability of plasmacytoid dendritic cells for the acquisition and presentation of tumor antigens. Journal of immunology 192, 824- 832.

Bonasio, R., Scimone, M.L., Schaerli, P., Grabie, N., Lichtman, A.H., and von Andrian, U.H. (2006). Clonal deletion of thymocytes by circulating dendritic cells homing to the thymus. Nature immunology 7, 1092-1100.

187

Bonnefoy, F., Couturier, M., Clauzon, A., Remy-Martin, J.P., Gaugler, B., Tiberghien, P., Chen, W., Saas, P., and Perruche, S. (2011). TGF-beta-exposed plasmacytoid dendritic cells participate in Th17 commitment. Journal of immunology 186, 6157-6164.

Bookstaver, M.L., Tsai, S.J., Bromberg, J.S., and Jewell, C.M. (2018). Improving Vaccine and Immunotherapy Design Using Biomaterials. Trends in immunology 39, 135-150.

Boonstra, A., Asselin-Paturel, C., Gilliet, M., Crain, C., Trinchieri, G., Liu, Y.J., and O'Garra, A. (2003). Flexibility of mouse classical and plasmacytoid-derived dendritic cells in directing T helper type 1 and 2 cell development: dependency on antigen dose and differential toll-like receptor ligation. The Journal of experimental medicine 197, 101-109.

Bossard, C., Malard, F., Arbez, J., Chevallier, P., Guillaume, T., Delaunay, J., Mosnier, J.F., Tiberghien, P., Saas, P., Mohty, M., et al. (2012). Plasmacytoid dendritic cells and Th17 immune response contribution in gastrointestinal acute graft- versus-host disease. Leukemia 26, 1471-1474.

Bottcher, J.P., Bonavita, E., Chakravarty, P., Blees, H., Cabeza-Cabrerizo, M., Sammicheli, S., Rogers, N.C., Sahai, E., Zelenay, S., and Reis e Sousa, C. (2018). NK Cells Stimulate Recruitment of cDC1 into the Tumor Microenvironment Promoting Cancer Immune Control. Cell 172, 1022-1037 e1014.

Bottcher, J.P., Schanz, O., Wohlleber, D., Abdullah, Z., Debey-Pascher, S., Staratschek-Jox, A., Hochst, B., Hegenbarth, S., Grell, J., Limmer, A., et al. (2013). Liver-primed memory T cells generated under noninflammatory conditions provide anti-infectious immunity. Cell reports 3, 779-795.

Bouta, E.M., Bell, R.D., Rahimi, H., Xing, L., Wood, R.W., Bingham, C.O., 3rd, Ritchlin, C.T., and Schwarz, E.M. (2018). Targeting lymphatic function as a novel therapeutic intervention for rheumatoid arthritis. Nature reviews Rheumatology 14, 94-106.

Bower, N.I., and Hogan, B.M. (2018). Brain drains: new insights into brain clearance pathways from lymphatic biology. Journal of molecular medicine 96, 383-390.

Bower, N.I., Koltowska, K., Pichol-Thievend, C., Virshup, I., Paterson, S., Lagendijk, A.K., Wang, W., Lindsey, B.W., Bent, S.J., Baek, S., et al. (2017). Mural lymphatic endothelial cells regulate meningeal angiogenesis in the zebrafish. Nature neuroscience 20, 774-783.

Brewitz, A., Eickhoff, S., Dahling, S., Quast, T., Bedoui, S., Kroczek, R.A., Kurts, C., Garbi, N., Barchet, W., Iannacone, M., et al. (2017). CD8(+) T Cells Orchestrate pDC-XCR1(+) Dendritic Cell Spatial and Functional Cooperativity to Optimize Priming. Immunity 46, 205-219.

Brody, J.D., Ai, W.Z., Czerwinski, D.K., Torchia, J.A., Levy, M., Advani, R.H., Kim, Y.H., Hoppe, R.T., Knox, S.J., Shin, L.K., et al. (2010). In situ vaccination with a TLR9 agonist induces systemic lymphoma regression: a phase I/II study. Journal of clinical oncology : official journal of the American Society of Clinical Oncology 28, 4324-4332.

Brown, C. (2016). Aetiology: Neighbourhood watch. Nature 540, S4-S6.

Brown, F.D., and Turley, S.J. (2015). Fibroblastic reticular cells: organization and regulation of the T lymphocyte life cycle. Journal of immunology 194, 1389-1394.

Bruck, W., and Wegner, C. (2011). Insight into the mechanism of laquinimod action. Journal of the neurological sciences 306, 173-179.

Bruckener, K.E., el Baya, A., Galla, H.J., and Schmidt, M.A. (2003). Permeabilization in a cerebral endothelial barrier model by pertussis toxin involves the PKC effector pathway and is abolished by elevated levels of cAMP. Journal of cell science 116, 1837-1846.

Bull, C., Boltje, T.J., Balneger, N., Weischer, S.M., Wassink, M., van Gemst, J.J., Bloemendal, V.R., Boon, L., van der Vlag, J., Heise, T., et al. (2018). Sialic Acid Blockade Suppresses Tumor Growth by Enhancing T-cell-Mediated Tumor Immunity. Cancer research 78, 3574-3588.

Bunin, A., Sisirak, V., Ghosh, H.S., Grajkowska, L.T., Hou, Z.E., Miron, M., Yang, C., Ceribelli, M., Uetani, N., Chaperot, L., et al. (2015). Protein Tyrosine Phosphatase PTPRS Is an Inhibitory Receptor on Human and Murine Plasmacytoid Dendritic Cells. Immunity 43, 277-288.

188

Bunk, E.C., Ertaylan, G., Ortega, F., Pavlou, M.A., Gonzalez Cano, L., Stergiopoulos, A., Safaiyan, S., Vols, S., van Cann, M., Politis, P.K., et al. (2016). Prox1 Is Required for Oligodendrocyte Cell Identity in Adult Neural Stem Cells of the Subventricular Zone. Stem cells 34, 2115-2129.

Caielli, S., Athale, S., Domic, B., Murat, E., Chandra, M., Banchereau, R., Baisch, J., Phelps, K., Clayton, S., Gong, M., et al. (2016). Oxidized mitochondrial nucleoids released by neutrophils drive type I interferon production in human lupus. The Journal of experimental medicine 213, 697-713.

Calvo, C.F., Fontaine, R.H., Soueid, J., Tammela, T., Makinen, T., Alfaro-Cervello, C., Bonnaud, F., Miguez, A., Benhaim, L., Xu, Y., et al. (2011). Vascular endothelial growth factor receptor 3 directly regulates murine neurogenesis. Genes & development 25, 831-844.

Camisaschi, C., De Filippo, A., Beretta, V., Vergani, B., Villa, A., Vergani, E., Santinami, M., Cabras, A.D., Arienti, F., Triebel, F., et al. (2014). Alternative activation of human plasmacytoid DCs in vitro and in melanoma lesions: involvement of LAG-3. The Journal of investigative dermatology 134, 1893-1902.

Cao, X., Cai, S.F., Fehniger, T.A., Song, J., Collins, L.I., Piwnica-Worms, D.R., and Ley, T.J. (2007). Granzyme B and perforin are important for regulatory T cell-mediated suppression of tumor clearance. Immunity 27, 635-646.

Cao, Y., Goods, B.A., Raddassi, K., Nepom, G.T., Kwok, W.W., Love, J.C., and Hafler, D.A. (2015). Functional inflammatory profiles distinguish myelin-reactive T cells from patients with multiple sclerosis. Science translational medicine 7, 287ra274.

Carbo, A., Hontecillas, R., Andrew, T., Eden, K., Mei, Y., Hoops, S., and Bassaganya-Riera, J. (2014). Computational modeling of heterogeneity and function of CD4+ T cells. Frontiers in cell and developmental biology 2, 31.

Card, C.M., Yu, S.S., and Swartz, M.A. (2014). Emerging roles of lymphatic endothelium in regulating adaptive immunity. The Journal of clinical investigation 124, 943-952.

Carpentier, A., Metellus, P., Ursu, R., Zohar, S., Lafitte, F., Barrie, M., Meng, Y., Richard, M., Parizot, C., Laigle-Donadey, F., et al. (2010). Intracerebral administration of CpG oligonucleotide for patients with recurrent glioblastoma: a phase II study. Neuro-oncology 12, 401-408.

Carpentier, A.F., Xie, J., Mokhtari, K., and Delattre, J.Y. (2000). Successful treatment of intracranial gliomas in rat by oligodeoxynucleotides containing CpG motifs. Clinical cancer research : an official journal of the American Association for Cancer Research 6, 2469-2473.

Cassan, C., Piaggio, E., Zappulla, J.P., Mars, L.T., Couturier, N., Bucciarelli, F., Desbois, S., Bauer, J., Gonzalez-Dunia, D., and Liblau, R.S. (2006). Pertussis toxin reduces the number of splenic Foxp3+ regulatory T cells. Journal of immunology 177, 1552-1560.

Cella, M., Facchetti, F., Lanzavecchia, A., and Colonna, M. (2000). Plasmacytoid dendritic cells activated by influenza virus and CD40L drive a potent TH1 polarization. Nature immunology 1, 305-310.

Chan, T.S., Hsu, C.C., Pai, V.C., Liao, W.Y., Huang, S.S., Tan, K.T., Yen, C.J., Hsu, S.C., Chen, W.Y., Shan, Y.S., et al. (2016). Metronomic chemotherapy prevents therapy-induced stromal activation and induction of tumor-initiating cells. The Journal of experimental medicine 213, 2967-2988.

Chan, V.S., Nie, Y.J., Shen, N., Yan, S., Mok, M.Y., and Lau, C.S. (2012). Distinct roles of myeloid and plasmacytoid dendritic cells in systemic lupus erythematosus. Autoimmunity reviews 11, 890-897.

Chappell, C.P., Giltiay, N.V., Draves, K.E., Chen, C., Hayden-Ledbetter, M.S., Shlomchik, M.J., Kaplan, D.H., and Clark, E.A. (2014). Targeting antigens through blood dendritic cell antigen 2 on plasmacytoid dendritic cells promotes immunologic tolerance. Journal of immunology 192, 5789-5801.

Charles, J., Chaperot, L., Revol, B., Baudin, M., Mouret, S., Hamon, A., Leccia, M.T., Plumas, J., and Aspord, C. (2018). The avidity of tumor-specific T cells amplified by a plasmacytoid dendritic cell-based assay can predict the clinical evolution of melanoma patients. Pigment cell & melanoma research 31, 82-94.

Charles, J., Di Domizio, J., Salameire, D., Bendriss-Vermare, N., Aspord, C., Muhammad, R., Lefebvre, C., Plumas, J., Leccia, M.T., and Chaperot, L. (2010). Characterization of circulating dendritic cells in melanoma: role of CCR6 in plasmacytoid dendritic cell recruitment to the tumor. The Journal of investigative dermatology 130, 1646-1656.

189

Charmoy, M., Brunner-Agten, S., Aebischer, D., Auderset, F., Launois, P., Milon, G., Proudfoot, A.E., and Tacchini- Cottier, F. (2010). Neutrophil-derived CCL3 is essential for the rapid recruitment of dendritic cells to the site of Leishmania major inoculation in resistant mice. PLoS pathogens 6, e1000755.

Chauhan, D., Singh, A.V., Brahmandam, M., Carrasco, R., Bandi, M., Hideshima, T., Bianchi, G., Podar, K., Tai, Y.T., Mitsiades, C., et al. (2009). Functional interaction of plasmacytoid dendritic cells with multiple myeloma cells: a therapeutic target. Cancer cell 16, 309-323.

Chen, D.S., and Mellman, I. (2013). Oncology meets immunology: the cancer-immunity cycle. Immunity 39, 1-10.

Chen, D.S., and Mellman, I. (2017). Elements of cancer immunity and the cancer-immune set point. Nature 541, 321-330.

Chen, Q., Kim, Y.C., Laurence, A., Punkosdy, G.A., and Shevach, E.M. (2011). IL-2 controls the stability of Foxp3 expression in TGF-beta-induced Foxp3+ T cells in vivo. Journal of immunology 186, 6329-6337.

Chen, W., Jin, W., Hardegen, N., Lei, K.J., Li, L., Marinos, N., McGrady, G., and Wahl, S.M. (2003). Conversion of peripheral CD4+CD25- naive T cells to CD4+CD25+ regulatory T cells by TGF-beta induction of transcription factor Foxp3. The Journal of experimental medicine 198, 1875-1886.

Chen, W., Liang, X., Peterson, A.J., Munn, D.H., and Blazar, B.R. (2008). The indoleamine 2,3-dioxygenase pathway is essential for human plasmacytoid dendritic cell-induced adaptive T regulatory cell generation. Journal of immunology 181, 5396-5404.

Cheng, M., and Anderson, M.S. (2018). Thymic tolerance as a key brake on autoimmunity. Nature immunology.

Cheon, D.J., and Orsulic, S. (2011). Mouse models of cancer. Annual review of pathology 6, 95-119.

Chevolet, I., Speeckaert, R., Schreuer, M., Neyns, B., Krysko, O., Bachert, C., Hennart, B., Allorge, D., van Geel, N., Van Gele, M., et al. (2015). Characterization of the in vivo immune network of IDO, tryptophan metabolism, PD-L1, and CTLA-4 in circulating immune cells in melanoma. Oncoimmunology 4, e982382.

Choi, I., Lee, S., and Hong, Y.K. (2012). The new era of the lymphatic system: no longer secondary to the blood vascular system. Cold Spring Harbor perspectives in medicine 2, a006445.

Chun, J., and Hartung, H.P. (2010). Mechanism of action of oral fingolimod (FTY720) in multiple sclerosis. Clinical neuropharmacology 33, 91-101.

Cire, S., Da Rocha, S., Ferrand, M., Collins, M.K., and Galy, A. (2016). In Vivo Gene Delivery to Lymph Node Stromal Cells Leads to Transgene-specific CD8+ T Cell Anergy in Mice. Molecular therapy : the journal of the American Society of Gene Therapy 24, 1965-1973.

Cisse, B., Caton, M.L., Lehner, M., Maeda, T., Scheu, S., Locksley, R., Holmberg, D., Zweier, C., den Hollander, N.S., Kant, S.G., et al. (2008). Transcription factor E2-2 is an essential and specific regulator of plasmacytoid dendritic cell development. Cell 135, 37-48.

Clement, C.C., and Santambrogio, L. (2013). The lymph self-antigen repertoire. Frontiers in immunology 4, 424.

Clemente-Casares, X., Blanco, J., Ambalavanan, P., Yamanouchi, J., Singha, S., Fandos, C., Tsai, S., Wang, J., Garabatos, N., Izquierdo, C., et al. (2016). Expanding antigen-specific regulatory networks to treat autoimmunity. Nature 530, 434-440.

Codarri, L., Gyulveszi, G., Tosevski, V., Hesske, L., Fontana, A., Magnenat, L., Suter, T., and Becher, B. (2011). RORgammat drives production of the cytokine GM-CSF in helper T cells, which is essential for the effector phase of autoimmune neuroinflammation. Nature immunology 12, 560-567.

Cohen, J.N., Guidi, C.J., Tewalt, E.F., Qiao, H., Rouhani, S.J., Ruddell, A., Farr, A.G., Tung, K.S., and Engelhard, V.H. (2010). Lymph node-resident lymphatic endothelial cells mediate peripheral tolerance via Aire-independent direct antigen presentation. The Journal of experimental medicine 207, 681-688.

Cohen, J.N., Tewalt, E.F., Rouhani, S.J., Buonomo, E.L., Bruce, A.N., Xu, X., Bekiranov, S., Fu, Y.X., and Engelhard, V.H. (2014). Tolerogenic properties of lymphatic endothelial cells are controlled by the lymph node microenvironment. PloS one 9, e87740.

190

Collier, A.Y., Lee, J.W., and Turley, S.J. (2008). Self-encounters of the third kind: lymph node stroma promotes tolerance to peripheral tissue antigens. Mucosal immunology 1, 248-251.

Colombo, E., and Farina, C. (2016). Astrocytes: Key Regulators of Neuroinflammation. Trends in immunology 37, 608- 620.

Combes, A., Camosseto, V., N'Guessan, P., Arguello, R.J., Mussard, J., Caux, C., Bendriss-Vermare, N., Pierre, P., and Gatti, E. (2017). BAD-LAMP controls TLR9 trafficking and signalling in human plasmacytoid dendritic cells. Nature communications 8, 913.

Compston, A., and Coles, A. (2008). Multiple sclerosis. Lancet 372, 1502-1517.

Conrad, C., Gregorio, J., Wang, Y.H., Ito, T., Meller, S., Hanabuchi, S., Anderson, S., Atkinson, N., Ramirez, P.T., Liu, Y.J., et al. (2012). Plasmacytoid dendritic cells promote immunosuppression in ovarian cancer via ICOS costimulation of Foxp3(+) T-regulatory cells. Cancer research 72, 5240-5249.

Cosmi, L., Liotta, F., Lazzeri, E., Francalanci, M., Angeli, R., Mazzinghi, B., Santarlasci, V., Manetti, R., Vanini, V., Romagnani, P., et al. (2003). Human CD8+CD25+ thymocytes share phenotypic and functional features with CD4+CD25+ regulatory thymocytes. Blood 102, 4107-4114.

Costa, A., Kieffer, Y., Scholer-Dahirel, A., Pelon, F., Bourachot, B., Cardon, M., Sirven, P., Magagna, I., Fuhrmann, L., Bernard, C., et al. (2018). Fibroblast Heterogeneity and Immunosuppressive Environment in Human Breast Cancer. Cancer cell 33, 463-479 e410.

Costantino, C.M., Spooner, E., Ploegh, H.L., and Hafler, D.A. (2012). Class II MHC self-antigen presentation in human B and T lymphocytes. PloS one 7, e29805.

Crispe, I.N. (2011). Liver antigen-presenting cells. Journal of hepatology 54, 357-365.

Crotty, S. (2011). Follicular helper CD4 T cells (TFH). Annual review of immunology 29, 621-663.

Croxford, A.L., Lanzinger, M., Hartmann, F.J., Schreiner, B., Mair, F., Pelczar, P., Clausen, B.E., Jung, S., Greter, M., and Becher, B. (2015a). The Cytokine GM-CSF Drives the Inflammatory Signature of CCR2+ Monocytes and Licenses Autoimmunity. Immunity 43, 502-514.

Croxford, A.L., Spath, S., and Becher, B. (2015b). GM-CSF in Neuroinflammation: Licensing Myeloid Cells for Tissue Damage. Trends in immunology 36, 651-662.

Cua, D.J., Sherlock, J., Chen, Y., Murphy, C.A., Joyce, B., Seymour, B., Lucian, L., To, W., Kwan, S., Churakova, T., et al. (2003). Interleukin-23 rather than interleukin-12 is the critical cytokine for autoimmune inflammation of the brain. Nature 421, 744-748.

Cyster, J.G., and Schwab, S.R. (2012). Sphingosine-1-phosphate and lymphocyte egress from lymphoid organs. Annual review of immunology 30, 69-94.

D'Rozario, J., Roberts, D., Suliman, M., Knoblich, K., and Fletcher, A. (2018). Leukocyte-Stromal Interactions Within Lymph Nodes. Advances in experimental medicine and biology 1060, 1-22.

Da Mesquita, S., Louveau, A., Vaccari, A., Smirnov, I., Cornelison, R.C., Kingsmore, K.M., Contarino, C., Onengut- Gumuscu, S., Farber, E., Raper, D., et al. (2018). Functional aspects of meningeal lymphatics in ageing and Alzheimer's disease. Nature 560, 185-191.

Dalmay, F., Bhalla, D., Nicoletti, A., Cabrera-Gomez, J.A., Cabre, P., Ruiz, F., Druet-Cabanac, M., Dumas, M., and Preux, P.M. (2010). Multiple sclerosis and solar exposure before the age of 15 years: case-control study in Cuba, Martinique and Sicily. Multiple sclerosis 16, 899-908.

Dalod, M., Chelbi, R., Malissen, B., and Lawrence, T. (2014). Dendritic cell maturation: functional specialization through signaling specificity and transcriptional programming. The EMBO journal 33, 1104-1116.

Dalod, M., Hamilton, T., Salomon, R., Salazar-Mather, T.P., Henry, S.C., Hamilton, J.D., and Biron, C.A. (2003). Dendritic cell responses to early murine cytomegalovirus infection: subset functional specialization and differential regulation by interferon alpha/beta. The Journal of experimental medicine 197, 885-898.

191

Danaei, G., Vander Hoorn, S., Lopez, A.D., Murray, C.J., Ezzati, M., and Comparative Risk Assessment collaborating, g. (2005). Causes of cancer in the world: comparative risk assessment of nine behavioural and environmental risk factors. Lancet 366, 1784-1793.

Dasoveanu, D.C., Shipman, W.D., Chia, J.J., Chyou, S., and Lu, T.T. (2016). Regulation of Lymph Node Vascular-Stromal Compartment by Dendritic Cells. Trends in immunology 37, 764-777.

Davis, D.M. (2007). Intercellular transfer of cell-surface proteins is common and can affect many stages of an immune response. Nature reviews Immunology 7, 238-243.

Day, C.P., Merlino, G., and Van Dyke, T. (2015). Preclinical mouse cancer models: a maze of opportunities and challenges. Cell 163, 39-53. de Heer, H.J., Hammad, H., Soullie, T., Hijdra, D., Vos, N., Willart, M.A., Hoogsteden, H.C., and Lambrecht, B.N. (2004). Essential role of lung plasmacytoid dendritic cells in preventing asthmatic reactions to harmless inhaled antigen. The Journal of experimental medicine 200, 89-98. de Heusch, M., Blocklet, D., Egrise, D., Hauquier, B., Vermeersch, M., Goldman, S., and Moser, M. (2007). Bidirectional MHC molecule exchange between migratory and resident dendritic cells. Journal of leukocyte biology 82, 861-868.

Dean, G., and Elian, M. (1997). Age at immigration to England of Asian and Caribbean immigrants and the risk of developing multiple sclerosis. Journal of neurology, neurosurgery, and psychiatry 63, 565-568.

Defays, A., David, A., de Gassart, A., De Angelis Rigotti, F., Wenger, T., Camossetto, V., Brousset, P., Petrella, T., Dalod, M., Gatti, E., et al. (2011). BAD-LAMP is a novel biomarker of nonactivated human plasmacytoid dendritic cells. Blood 118, 609-617.

Delarasse, C., Daubas, P., Mars, L.T., Vizler, C., Litzenburger, T., Iglesias, A., Bauer, J., Della Gaspera, B., Schubart, A., Decker, L., et al. (2003). Myelin/oligodendrocyte glycoprotein-deficient (MOG-deficient) mice reveal lack of immune tolerance to MOG in wild-type mice. The Journal of clinical investigation 112, 544-553.

Demaria, O., Di Domizio, J., and Gilliet, M. (2014). Immune sensing of nucleic acids in inflammatory skin diseases. Seminars in immunopathology 36, 519-529.

Demoulin, S., Herfs, M., Delvenne, P., and Hubert, P. (2013). Tumor microenvironment converts plasmacytoid dendritic cells into immunosuppressive/tolerogenic cells: insight into the molecular mechanisms. Journal of leukocyte biology 93, 343-352. den Brok, M.H., Sutmuller, R.P., Nierkens, S., Bennink, E.J., Toonen, L.W., Figdor, C.G., Ruers, T.J., and Adema, G.J. (2006). Synergy between in situ cryoablation and TLR9 stimulation results in a highly effective in vivo dendritic cell vaccine. Cancer research 66, 7285-7292.

Dendrou, C.A., Fugger, L., and Friese, M.A. (2015). Immunopathology of multiple sclerosis. Nature reviews Immunology 15, 545-558.

Dendrou, C.A., Petersen, J., Rossjohn, J., and Fugger, L. (2018). HLA variation and disease. Nature reviews Immunology 18, 325-339.

Derbinski, J., Gabler, J., Brors, B., Tierling, S., Jonnakuty, S., Hergenhahn, M., Peltonen, L., Walter, J., and Kyewski, B. (2005). Promiscuous gene expression in thymic epithelial cells is regulated at multiple levels. The Journal of experimental medicine 202, 33-45.

Derbinski, J., Schulte, A., Kyewski, B., and Klein, L. (2001). Promiscuous gene expression in medullary thymic epithelial cells mirrors the peripheral self. Nature immunology 2, 1032-1039.

Di Pucchio, T., Chatterjee, B., Smed-Sorensen, A., Clayton, S., Palazzo, A., Montes, M., Xue, Y., Mellman, I., Banchereau, J., and Connolly, J.E. (2008). Direct proteasome-independent cross-presentation of viral antigen by plasmacytoid dendritic cells on major histocompatibility complex class I. Nature immunology 9, 551-557.

Diamond, M.S., Kinder, M., Matsushita, H., Mashayekhi, M., Dunn, G.P., Archambault, J.M., Lee, H., Arthur, C.D., White, J.M., Kalinke, U., et al. (2011). Type I interferon is selectively required by dendritic cells for immune rejection of tumors. The Journal of experimental medicine 208, 1989-2003.

192

Diana, J., Simoni, Y., Furio, L., Beaudoin, L., Agerberth, B., Barrat, F., and Lehuen, A. (2013). Crosstalk between neutrophils, B-1a cells and plasmacytoid dendritic cells initiates autoimmune diabetes. Nature medicine 19, 65-73.

Diebold, M., and Derfuss, T. (2016). Immunological treatment of multiple sclerosis. Seminars in hematology 53 Suppl 1, S54-57.

Diebold, S.S., Kaisho, T., Hemmi, H., Akira, S., and Reis e Sousa, C. (2004). Innate antiviral responses by means of TLR7- mediated recognition of single-stranded RNA. Science 303, 1529-1531.

Diehl, L., Schurich, A., Grochtmann, R., Hegenbarth, S., Chen, L., and Knolle, P.A. (2008). Tolerogenic maturation of liver sinusoidal endothelial cells promotes B7-homolog 1-dependent CD8+ T cell tolerance. Hepatology 47, 296-305.

Dieterich, L.C., and Detmar, M. (2016). Tumor lymphangiogenesis and new drug development. Advanced drug delivery reviews 99, 148-160.

Dieterich, L.C., Ikenberg, K., Cetintas, T., Kapaklikaya, K., Hutmacher, C., and Detmar, M. (2017). Tumor-Associated Lymphatic Vessels Upregulate PDL1 to Inhibit T-Cell Activation. Frontiers in immunology 8, 66.

Disanto, G., Hall, C., Lucas, R., Ponsonby, A.L., Berlanga-Taylor, A.J., Giovannoni, G., Ramagopalan, S.V., and Ausimmune Investigator, G. (2013). Assessing interactions between HLA-DRB1*15 and infectious mononucleosis on the risk of multiple sclerosis. Multiple sclerosis 19, 1355-1358.

Dominguez-Villar, M., and Hafler, D.A. (2018). Regulatory T cells in autoimmune disease. Nature immunology.

Donia, M., Andersen, R., Kjeldsen, J.W., Fagone, P., Munir, S., Nicoletti, F., Andersen, M.H., Thor Straten, P., and Svane, I.M. (2015). Aberrant Expression of MHC Class II in Melanoma Attracts Inflammatory Tumor-Specific CD4+ T- Cells, Which Dampen CD8+ T-cell Antitumor Reactivity. Cancer research 75, 3747-3759.

Doorduijn, E.M., Sluijter, M., Salvatori, D.C., Silvestri, S., Maas, S., Arens, R., Ossendorp, F., van der Burg, S.H., and van Hall, T. (2017). CD4(+) T Cell and NK Cell Interplay Key to Regression of MHC Class I(low) Tumors upon TLR7/8 Agonist Therapy. Cancer immunology research 5, 642-653.

Drobits, B., Holcmann, M., Amberg, N., Swiecki, M., Grundtner, R., Hammer, M., Colonna, M., and Sibilia, M. (2012). Imiquimod clears tumors in mice independent of adaptive immunity by converting pDCs into tumor-killing effector cells. The Journal of clinical investigation 122, 575-585.

Dubois, B., Joubert, G., Gomez de Aguero, M., Gouanvic, M., Goubier, A., and Kaiserlian, D. (2009). Sequential role of plasmacytoid dendritic cells and regulatory T cells in oral tolerance. Gastroenterology 137, 1019-1028.

Dubrot, J., Duraes, F.V., Potin, L., Capotosti, F., Brighouse, D., Suter, T., LeibundGut-Landmann, S., Garbi, N., Reith, W., Swartz, M.A., et al. (2014). Lymph node stromal cells acquire peptide-MHCII complexes from dendritic cells and induce antigen-specific CD4(+) T cell tolerance. The Journal of experimental medicine 211, 1153-1166.

Dumas, A., Amiable, N., de Rivero Vaccari, J.P., Chae, J.J., Keane, R.W., Lacroix, S., and Vallieres, L. (2014). The inflammasome pyrin contributes to pertussis toxin-induced IL-1beta synthesis, neutrophil intravascular crawling and autoimmune encephalomyelitis. PLoS pathogens 10, e1004150.

Dumont, D.J., Jussila, L., Taipale, J., Lymboussaki, A., Mustonen, T., Pajusola, K., Breitman, M., and Alitalo, K. (1998). Cardiovascular failure in mouse embryos deficient in VEGF receptor-3. Science 282, 946-949.

Dunn, G.P., Old, L.J., and Schreiber, R.D. (2004). The immunobiology of cancer immunosurveillance and immunoediting. Immunity 21, 137-148.

DuPage, M., and Bluestone, J.A. (2016). Harnessing the plasticity of CD4(+) T cells to treat immune-mediated disease. Nature reviews Immunology 16, 149-163.

DuPage, M., Dooley, A.L., and Jacks, T. (2009). Conditional mouse lung cancer models using adenoviral or lentiviral delivery of Cre recombinase. Nature protocols 4, 1064-1072.

Duraes, F.V., Lippens, C., Steinbach, K., Dubrot, J., Brighouse, D., Bendriss-Vermare, N., Issazadeh-Navikas, S., Merkler, D., and Hugues, S. (2016). pDC therapy induces recovery from EAE by recruiting endogenous pDC to sites of CNS inflammation. Journal of autoimmunity 67, 8-18.

193

Duraes, F.V., Thelemann, C., Sarter, K., Acha-Orbea, H., Hugues, S., and Reith, W. (2013). Role of major histocompatibility complex class II expression by non-hematopoietic cells in autoimmune and inflammatory disorders: facts and fiction. Tissue antigens 82, 1-15.

Durelli, L., Conti, L., Clerico, M., Boselli, D., Contessa, G., Ripellino, P., Ferrero, B., Eid, P., and Novelli, F. (2009). T- helper 17 cells expand in multiple sclerosis and are inhibited by interferon-beta. Annals of neurology 65, 499-509.

Dzionek, A., Fuchs, A., Schmidt, P., Cremer, S., Zysk, M., Miltenyi, S., Buck, D.W., and Schmitz, J. (2000). BDCA-2, BDCA-3, and BDCA-4: three markers for distinct subsets of dendritic cells in human peripheral blood. Journal of immunology 165, 6037-6046.

Dzionek, A., Sohma, Y., Nagafune, J., Cella, M., Colonna, M., Facchetti, F., Gunther, G., Johnston, I., Lanzavecchia, A., Nagasaka, T., et al. (2001). BDCA-2, a novel plasmacytoid dendritic cell-specific type II C-type lectin, mediates antigen capture and is a potent inhibitor of interferon alpha/beta induction. The Journal of experimental medicine 194, 1823-1834.

Ebers, G.C. (2008). Environmental factors and multiple sclerosis. The Lancet Neurology 7, 268-277.

El-Behi, M., Ciric, B., Dai, H., Yan, Y., Cullimore, M., Safavi, F., Zhang, G.X., Dittel, B.N., and Rostami, A. (2011). The encephalitogenicity of T(H)17 cells is dependent on IL-1- and IL-23-induced production of the cytokine GM-CSF. Nature immunology 12, 568-575.

Ellwardt, E., Walsh, J.T., Kipnis, J., and Zipp, F. (2016). Understanding the Role of T Cells in CNS Homeostasis. Trends in immunology 37, 154-165.

Engelhard, V.H., Rodriguez, A.B., Mauldin, I.S., Woods, A.N., Peske, J.D., and Slingluff, C.L., Jr. (2018). Immune Cell Infiltration and Tertiary Lymphoid Structures as Determinants of Antitumor Immunity. Journal of immunology 200, 432-442.

Engelhardt, B., and Ransohoff, R.M. (2012). Capture, crawl, cross: the T cell code to breach the blood-brain barriers. Trends in immunology 33, 579-589.

Engelhardt, B., Vajkoczy, P., and Weller, R.O. (2017). The movers and shapers in immune privilege of the CNS. Nature immunology 18, 123-131.

Enouz, S., Carrie, L., Merkler, D., Bevan, M.J., and Zehn, D. (2012). Autoreactive T cells bypass negative selection and respond to self-antigen stimulation during infection. The Journal of experimental medicine 209, 1769-1779.

Faget, J., Bendriss-Vermare, N., Gobert, M., Durand, I., Olive, D., Biota, C., Bachelot, T., Treilleux, I., Goddard-Leon, S., Lavergne, E., et al. (2012). ICOS-ligand expression on plasmacytoid dendritic cells supports breast cancer progression by promoting the accumulation of immunosuppressive CD4+ T cells. Cancer research 72, 6130-6141.

Fallarino, F., Asselin-Paturel, C., Vacca, C., Bianchi, R., Gizzi, S., Fioretti, M.C., Trinchieri, G., Grohmann, U., and Puccetti, P. (2004). Murine plasmacytoid dendritic cells initiate the immunosuppressive pathway of tryptophan catabolism in response to CD200 receptor engagement. Journal of immunology 173, 3748-3754.

Fankhauser, M., Broggi, M.A.S., Potin, L., Bordry, N., Jeanbart, L., Lund, A.W., Da Costa, E., Hauert, S., Rincon- Restrepo, M., Tremblay, C., et al. (2017). Tumor lymphangiogenesis promotes T cell infiltration and potentiates immunotherapy in melanoma. Science translational medicine 9.

Ferber, I.A., Brocke, S., Taylor-Edwards, C., Ridgway, W., Dinisco, C., Steinman, L., Dalton, D., and Fathman, C.G. (1996). Mice with a disrupted IFN-gamma gene are susceptible to the induction of experimental autoimmune encephalomyelitis (EAE). Journal of immunology 156, 5-7.

Fillatreau, S., and Anderton, S.M. (2007). B-cell function in CNS inflammatory demyelinating disease: a complexity of roles and a wealth of possibilities. Expert review of clinical immunology 3, 565-578.

Fletcher, A.L., Lukacs-Kornek, V., Reynoso, E.D., Pinner, S.E., Bellemare-Pelletier, A., Curry, M.S., Collier, A.R., Boyd, R.L., and Turley, S.J. (2010). Lymph node fibroblastic reticular cells directly present peripheral tissue antigen under steady-state and inflammatory conditions. The Journal of experimental medicine 207, 689-697.

Fletcher, A.L., Malhotra, D., and Turley, S.J. (2011). Lymph node stroma broaden the peripheral tolerance paradigm. Trends in immunology 32, 12-18.

194

Fox, E.J. (2004). Mechanism of action of mitoxantrone. Neurology 63, S15-18.

Frank, M.J., Reagan, P.M., Bartlett, N.L., Gordon, L.I., Friedberg, J.W., Czerwinski, D.K., Long, S.R., Hoppe, R.T., Janssen, R., Candia, A.F., et al. (2018). In Situ Vaccination with a TLR9 Agonist and Local Low-Dose Radiation Induces Systemic Responses in Untreated Indolent Lymphoma. Cancer discovery 8, 1258-1269.

Fransen, M.F., Ossendorp, F., Arens, R., and Melief, C.J. (2013). Local immunomodulation for cancer therapy: Providing treatment where needed. Oncoimmunology 2, e26493.

Freeman, G.J., Long, A.J., Iwai, Y., Bourque, K., Chernova, T., Nishimura, H., Fitz, L.J., Malenkovich, N., Okazaki, T., Byrne, M.C., et al. (2000). Engagement of the PD-1 immunoinhibitory receptor by a novel B7 family member leads to negative regulation of lymphocyte activation. The Journal of experimental medicine 192, 1027-1034.

Fridman, W.H., Pages, F., Sautes-Fridman, C., and Galon, J. (2012). The immune contexture in human tumours: impact on clinical outcome. Nature reviews Cancer 12, 298-306.

Friedlaender, M.H., Chisari, F.V., and Baer, H. (1973). The role of the inflammatory response of skin and lymph nodes in the induction of sensitization to simple chemicals. Journal of immunology 111, 164-170.

Fruhwurth, S., Pavelka, M., Bittman, R., Kovacs, W.J., Walter, K.M., Rohrl, C., and Stangl, H. (2013). High-density lipoprotein endocytosis in endothelial cells. World journal of biological chemistry 4, 131-140.

Fu, H., Ward, E.J., and Marelli-Berg, F.M. (2016). Mechanisms of T cell organotropism. Cellular and molecular life sciences : CMLS 73, 3009-3033.

Fyfe, I. (2016). Multiple sclerosis: Real-world long-term benefits of disease-modifying MS therapy. Nature reviews Neurology 12, 372.

Gale, C.R., and Martyn, C.N. (1995). Migrant studies in multiple sclerosis. Progress in neurobiology 47, 425-448.

Galea, I., Bernardes-Silva, M., Forse, P.A., van Rooijen, N., Liblau, R.S., and Perry, V.H. (2007). An antigen-specific pathway for CD8 T cells across the blood-brain barrier. The Journal of experimental medicine 204, 2023-2030.

Gallegos, A.M., and Bevan, M.J. (2004). Central tolerance to tissue-specific antigens mediated by direct and indirect antigen presentation. The Journal of experimental medicine 200, 1039-1049.

Gallotta, M., Assi, H., Degagne, E., Kannan, S.K., Coffman, R.L., and Guiducci, C. (2018). Inhaled TLR9 Agonist Renders Lung Tumors Permissive to PD-1 Blockade by Promoting Optimal CD4(+) and CD8(+) T-cell Interplay. Cancer research 78, 4943-4956.

Galluzzi, L., Vacchelli, E., Bravo-San Pedro, J.M., Buque, A., Senovilla, L., Baracco, E.E., Bloy, N., Castoldi, F., Abastado, J.P., Agostinis, P., et al. (2014). Classification of current anticancer immunotherapies. Oncotarget 5, 12472-12508.

Galon, J., Costes, A., Sanchez-Cabo, F., Kirilovsky, A., Mlecnik, B., Lagorce-Pages, C., Tosolini, M., Camus, M., Berger, A., Wind, P., et al. (2006). Type, density, and location of immune cells within human colorectal tumors predict clinical outcome. Science 313, 1960-1964.

Gardner, J.M., Devoss, J.J., Friedman, R.S., Wong, D.J., Tan, Y.X., Zhou, X., Johannes, K.P., Su, M.A., Chang, H.Y., Krummel, M.F., et al. (2008). Deletional tolerance mediated by extrathymic Aire-expressing cells. Science 321, 843- 847.

Gardner, J.M., Metzger, T.C., McMahon, E.J., Au-Yeung, B.B., Krawisz, A.K., Lu, W., Price, J.D., Johannes, K.P., Satpathy, A.T., Murphy, K.M., et al. (2013). Extrathymic Aire-expressing cells are a distinct bone marrow-derived population that induce functional inactivation of CD4(+) T cells. Immunity 39, 560-572.

Garin, M.I., Chu, C.C., Golshayan, D., Cernuda-Morollon, E., Wait, R., and Lechler, R.I. (2007). Galectin-1: a key effector of regulation mediated by CD4+CD25+ T cells. Blood 109, 2058-2065.

Gasperini, C., Haggiag, S., and Ruggieri, S. (2013). Drugs in clinical development for multiple sclerosis: focusing on anti- CD20 antibodies. Expert opinion on investigational drugs 22, 1243-1253.

Gautreau, L., Chabannes, D., Heslan, M., and Josien, R. (2011). Modulation of regulatory T cell-Th17 balance by plasmacytoid dendritic cells. Journal of leukocyte biology 90, 521-527.

195

Geary, S.M., Lemke, C.D., Lubaroff, D.M., and Salem, A.K. (2011). Tumor immunotherapy using adenovirus vaccines in combination with intratumoral doses of CpG ODN. Cancer immunology, immunotherapy : CII 60, 1309-1317.

Geginat, J., Paroni, M., Pagani, M., Galimberti, D., De Francesco, R., Scarpini, E., and Abrignani, S. (2017). The Enigmatic Role of Viruses in Multiple Sclerosis: Molecular Mimicry or Disturbed Immune Surveillance? Trends in immunology 38, 498-512.

Geng, J., Yu, S., Zhao, H., Sun, X., Li, X., Wang, P., Xiong, X., Hong, L., Xie, C., Gao, J., et al. (2017). The transcriptional coactivator TAZ regulates reciprocal differentiation of TH17 cells and Treg cells. Nature immunology 18, 800-812.

Gentek, R., and Bajenoff, M. (2017). Lymph Node Stroma Dynamics and Approaches for Their Visualization. Trends in immunology 38, 236-247.

Gerner, M.Y., Torabi-Parizi, P., and Germain, R.N. (2015). Strategically localized dendritic cells promote rapid T cell responses to lymph-borne particulate antigens. Immunity 42, 172-185.

Ghirelli, C., Reyal, F., Jeanmougin, M., Zollinger, R., Sirven, P., Michea, P., Caux, C., Bendriss-Vermare, N., Donnadieu, M.H., Caly, M., et al. (2015). Breast Cancer Cell-Derived GM-CSF Licenses Regulatory Th2 Induction by Plasmacytoid Predendritic Cells in Aggressive Disease Subtypes. Cancer research 75, 2775-2787.

Ghirelli, C., Zollinger, R., and Soumelis, V. (2010). Systematic cytokine receptor profiling reveals GM-CSF as a novel TLR-independent activator of human plasmacytoid predendritic cells. Blood 115, 5037-5040.

Ghosh, H.S., Ceribelli, M., Matos, I., Lazarovici, A., Bussemaker, H.J., Lasorella, A., Hiebert, S.W., Liu, K., Staudt, L.M., and Reizis, B. (2014). ETO family protein Mtg16 regulates the balance of dendritic cell subsets by repressing Id2. The Journal of experimental medicine 211, 1623-1635.

Ghosh, H.S., Cisse, B., Bunin, A., Lewis, K.L., and Reizis, B. (2010). Continuous expression of the transcription factor e2- 2 maintains the cell fate of mature plasmacytoid dendritic cells. Immunity 33, 905-916.

Gibson, T.J., Ramu, C., Gemund, C., and Aasland, R. (1998). The APECED polyglandular autoimmune syndrome protein, AIRE-1, contains the SAND domain and is probably a transcription factor. Trends in biochemical sciences 23, 242-244.

Giles, D.A., Duncker, P.C., Wilkinson, N.M., Washnock-Schmid, J.M., and Segal, B.M. (2018). CNS resident classical DCs play a critical role in CNS autoimmune disease. The Journal of clinical investigation.

Gilgun-Sherki, Y., Melamed, E., and Offen, D. (2004). The role of oxidative stress in the pathogenesis of multiple sclerosis: the need for effective antioxidant therapy. Journal of neurology 251, 261-268.

Gilliet, M., Cao, W., and Liu, Y.J. (2008). Plasmacytoid dendritic cells: sensing nucleic acids in viral infection and autoimmune diseases. Nature reviews Immunology 8, 594-606.

Gilliet, M., and Liu, Y.J. (2002). Generation of human CD8 T regulatory cells by CD40 ligand-activated plasmacytoid dendritic cells. The Journal of experimental medicine 195, 695-704.

Glatigny, S., Duhen, R., Oukka, M., and Bettelli, E. (2011). Cutting edge: loss of alpha4 integrin expression differentially affects the homing of Th1 and Th17 cells. Journal of immunology 187, 6176-6179.

Goc, J., Fridman, W.H., Sautes-Fridman, C., and Dieu-Nosjean, M.C. (2013). Characteristics of tertiary lymphoid structures in primary cancers. Oncoimmunology 2, e26836.

Goldacre, M.J., Seagroatt, V., Yeates, D., and Acheson, E.D. (2004). Skin cancer in people with multiple sclerosis: a record linkage study. Journal of epidemiology and community health 58, 142-144.

Goldmann, T., and Prinz, M. (2013). Role of microglia in CNS autoimmunity. Clinical & developmental immunology 2013, 208093.

Golub, R., Tan, J., Watanabe, T., and Brendolan, A. (2018). Origin and Immunological Functions of Spleen Stromal Cells. Trends in immunology 39, 503-514.

Goubier, A., Dubois, B., Gheit, H., Joubert, G., Villard-Truc, F., Asselin-Paturel, C., Trinchieri, G., and Kaiserlian, D. (2008). Plasmacytoid dendritic cells mediate oral tolerance. Immunity 29, 464-475.

196

Grajkowska, L.T., Ceribelli, M., Lau, C.M., Warren, M.E., Tiniakou, I., Nakandakari Higa, S., Bunin, A., Haecker, H., Mirny, L.A., Staudt, L.M., et al. (2017). Isoform-Specific Expression and Feedback Regulation of E Protein TCF4 Control Dendritic Cell Lineage Specification. Immunity 46, 65-77.

Grauer, O.M., Molling, J.W., Bennink, E., Toonen, L.W., Sutmuller, R.P., Nierkens, S., and Adema, G.J. (2008). TLR ligands in the local treatment of established intracerebral murine gliomas. Journal of immunology 181, 6720-6729.

Griffin, J.P., Chu, R., and Harding, C.V. (1997). Early endosomes and a late endocytic compartment generate different peptide-class II MHC complexes via distinct processing mechanisms. Journal of immunology 158, 1523-1532.

Griffin, M.D., Lutz, W., Phan, V.A., Bachman, L.A., McKean, D.J., and Kumar, R. (2001). Dendritic cell modulation by 1alpha,25 dihydroxyvitamin D3 and its analogs: a vitamin D receptor-dependent pathway that promotes a persistent state of immaturity in vitro and in vivo. Proceedings of the National Academy of Sciences of the United States of America 98, 6800-6805.

Grigorova, I.L., Schwab, S.R., Phan, T.G., Pham, T.H., Okada, T., and Cyster, J.G. (2009). Cortical sinus probing, S1P1- dependent entry and flow-based capture of egressing T cells. Nature immunology 10, 58-65.

Grouard, G., Rissoan, M.C., Filgueira, L., Durand, I., Banchereau, J., and Liu, Y.J. (1997). The enigmatic plasmacytoid T cells develop into dendritic cells with interleukin (IL)-3 and CD40-ligand. The Journal of experimental medicine 185, 1101-1111.

Guermonprez, P., Saveanu, L., Kleijmeer, M., Davoust, J., Van Endert, P., and Amigorena, S. (2003). ER-phagosome fusion defines an MHC class I cross-presentation compartment in dendritic cells. Nature 425, 397-402.

Guery, L., Dubrot, J., Lippens, C., Brighouse, D., Malinge, P., Irla, M., Pot, C., Reith, W., Waldburger, J.M., and Hugues, S. (2014). Ag-presenting CpG-activated pDCs prime Th17 cells that induce tumor regression. Cancer research 74, 6430-6440.

Guery, L., and Hugues, S. (2013). Tolerogenic and activatory plasmacytoid dendritic cells in autoimmunity. Frontiers in immunology 4, 59.

Guery, L., and Hugues, S. (2015a). New role for antigen-presenting activated pDCs in promoting Th17 cells and impacting antitumor immunity. Oncoimmunology 4, e988476.

Guery, L., and Hugues, S. (2015b). Th17 Cell Plasticity and Functions in Cancer Immunity. BioMed research international 2015, 314620.

Guiducci, C., Ghirelli, C., Marloie-Provost, M.A., Matray, T., Coffman, R.L., Liu, Y.J., Barrat, F.J., and Soumelis, V. (2008). PI3K is critical for the nuclear translocation of IRF-7 and type I IFN production by human plasmacytoid predendritic cells in response to TLR activation. The Journal of experimental medicine 205, 315-322.

Guilliams, M., Dutertre, C.A., Scott, C.L., McGovern, N., Sichien, D., Chakarov, S., Van Gassen, S., Chen, J., Poidinger, M., De Prijck, S., et al. (2016). Unsupervised High-Dimensional Analysis Aligns Dendritic Cells across Tissues and Species. Immunity 45, 669-684.

Guilliams, M., Ginhoux, F., Jakubzick, C., Naik, S.H., Onai, N., Schraml, B.U., Segura, E., Tussiwand, R., and Yona, S. (2014). Dendritic cells, monocytes and macrophages: a unified nomenclature based on ontogeny. Nature reviews Immunology 14, 571-578.

Gutierrez-Martinez, E., Planes, R., Anselmi, G., Reynolds, M., Menezes, S., Adiko, A.C., Saveanu, L., and Guermonprez, P. (2015). Cross-Presentation of Cell-Associated Antigens by MHC Class I in Dendritic Cell Subsets. Frontiers in immunology 6, 363.

Haahr, S., and Hollsberg, P. (2006). Multiple sclerosis is linked to Epstein-Barr virus infection. Reviews in medical virology 16, 297-310.

Haak, S., Croxford, A.L., Kreymborg, K., Heppner, F.L., Pouly, S., Becher, B., and Waisman, A. (2009). IL-17A and IL- 17F do not contribute vitally to autoimmune neuro-inflammation in mice. The Journal of clinical investigation 119, 61-69.

Hadeiba, H., Lahl, K., Edalati, A., Oderup, C., Habtezion, A., Pachynski, R., Nguyen, L., Ghodsi, A., Adler, S., and Butcher, E.C. (2012). Plasmacytoid dendritic cells transport peripheral antigens to the thymus to promote central tolerance. Immunity 36, 438-450.

197

Hadeiba, H., Sato, T., Habtezion, A., Oderup, C., Pan, J., and Butcher, E.C. (2008). CCR9 expression defines tolerogenic plasmacytoid dendritic cells able to suppress acute graft-versus-host disease. Nature immunology 9, 1253-1260.

Halle, S., Halle, O., and Forster, R. (2017). Mechanisms and Dynamics of T Cell-Mediated Cytotoxicity In Vivo. Trends in immunology 38, 432-443.

Hanabuchi, S., Ito, T., Park, W.R., Watanabe, N., Shaw, J.L., Roman, E., Arima, K., Wang, Y.H., Voo, K.S., Cao, W., et al. (2010). Thymic stromal lymphopoietin-activated plasmacytoid dendritic cells induce the generation of FOXP3+ regulatory T cells in human thymus. Journal of immunology 184, 2999-3007.

Hanahan, D., and Weinberg, R.A. (2011). Hallmarks of cancer: the next generation. Cell 144, 646-674.

Hara, T., Shitara, S., Imai, K., Miyachi, H., Kitano, S., Yao, H., Tani-ichi, S., and Ikuta, K. (2012). Identification of IL-7- producing cells in primary and secondary lymphoid organs using IL-7-GFP knock-in mice. Journal of immunology 189, 1577-1584.

Hartmann, E., Wollenberg, B., Rothenfusser, S., Wagner, M., Wellisch, D., Mack, B., Giese, T., Gires, O., Endres, S., and Hartmann, G. (2003). Identification and functional analysis of tumor-infiltrating plasmacytoid dendritic cells in head and neck cancer. Cancer research 63, 6478-6487.

Hartmann, F.J., Khademi, M., Aram, J., Ammann, S., Kockum, I., Constantinescu, C., Gran, B., Piehl, F., Olsson, T., Codarri, L., et al. (2014). Multiple sclerosis-associated IL2RA polymorphism controls GM-CSF production in human TH cells. Nature communications 5, 5056.

Hartung, H.P., Aktas, O., and Boyko, A.N. (2015). Alemtuzumab: a new therapy for active relapsing-remitting multiple sclerosis. Multiple sclerosis 21, 22-34.

Hauser, S.L., Waubant, E., Arnold, D.L., Vollmer, T., Antel, J., Fox, R.J., Bar-Or, A., Panzara, M., Sarkar, N., Agarwal, S., et al. (2008). B-cell depletion with rituximab in relapsing-remitting multiple sclerosis. The New England journal of medicine 358, 676-688.

Heath, W.R., Kurts, C., Miller, J.F., and Carbone, F.R. (1998). Cross-tolerance: a pathway for inducing tolerance to peripheral tissue antigens. The Journal of experimental medicine 187, 1549-1553.

Heemskerk, N., and van Egmond, M. (2018). Monoclonal antibody-mediated killing of tumour cells by neutrophils. European journal of clinical investigation 48 Suppl 2, e12962.

Hemmer, B., Kerschensteiner, M., and Korn, T. (2015). Role of the innate and adaptive immune responses in the course of multiple sclerosis. The Lancet Neurology 14, 406-419.

Hemmi, H., Kaisho, T., Takeuchi, O., Sato, S., Sanjo, H., Hoshino, K., Horiuchi, T., Tomizawa, H., Takeda, K., and Akira, S. (2002). Small anti-viral compounds activate immune cells via the TLR7 MyD88-dependent signaling pathway. Nature immunology 3, 196-200.

Hemminki, K., Li, X., Sundquist, J., Hillert, J., and Sundquist, K. (2009). Risk for multiple sclerosis in relatives and spouses of patients diagnosed with autoimmune and related conditions. Neurogenetics 10, 5-11.

Henault, J., Riggs, J.M., Karnell, J.L., Liarski, V.M., Li, J., Shirinian, L., Xu, L., Casey, K.A., Smith, M.A., Khatry, D.B., et al. (2016). Self-reactive IgE exacerbates interferon responses associated with autoimmunity. Nature immunology 17, 196-203.

Herkel, J., Jagemann, B., Wiegard, C., Lazaro, J.F., Lueth, S., Kanzler, S., Blessing, M., Schmitt, E., and Lohse, A.W. (2003). MHC class II-expressing hepatocytes function as antigen-presenting cells and activate specific CD4 T lymphocyutes. Hepatology 37, 1079-1085.

Hervas-Stubbs, S., Perez-Gracia, J.L., Rouzaut, A., Sanmamed, M.F., Le Bon, A., and Melero, I. (2011). Direct effects of type I interferons on cells of the immune system. Clinical cancer research : an official journal of the American Association for Cancer Research 17, 2619-2627.

Hirahara, K., and Nakayama, T. (2016). CD4+ T-cell subsets in inflammatory diseases: beyond the Th1/Th2 paradigm. International immunology 28, 163-171.

198

Hirakawa, S., Hong, Y.K., Harvey, N., Schacht, V., Matsuda, K., Libermann, T., and Detmar, M. (2003). Identification of vascular lineage-specific genes by transcriptional profiling of isolated blood vascular and lymphatic endothelial cells. The American journal of pathology 162, 575-586.

Hirosue, S., and Dubrot, J. (2015). Modes of Antigen Presentation by Lymph Node Stromal Cells and Their Immunological Implications. Frontiers in immunology 6, 446.

Hirosue, S., Vokali, E., Raghavan, V.R., Rincon-Restrepo, M., Lund, A.W., Corthesy-Henrioud, P., Capotosti, F., Halin Winter, C., Hugues, S., and Swartz, M.A. (2014). Steady-state antigen scavenging, cross-presentation, and CD8+ T cell priming: a new role for lymphatic endothelial cells. Journal of immunology 192, 5002-5011.

Hirsch, I., Caux, C., Hasan, U., Bendriss-Vermare, N., and Olive, D. (2010). Impaired Toll-like receptor 7 and 9 signaling: from chronic viral infections to cancer. Trends in immunology 31, 391-397.

Hoeffel, G., Ripoche, A.C., Matheoud, D., Nascimbeni, M., Escriou, N., Lebon, P., Heshmati, F., Guillet, J.G., Gannage, M., Caillat-Zucman, S., et al. (2007). Antigen crosspresentation by human plasmacytoid dendritic cells. Immunity 27, 481-492.

Hoffman, E.S., Smith, R.E., and Renaud, R.C., Jr. (2005). From the analyst's couch: TLR-targeted therapeutics. Nature reviews Drug discovery 4, 879-880.

Hohlfeld, R., Dornmair, K., Meinl, E., and Wekerle, H. (2016). The search for the target antigens of multiple sclerosis, part 1: autoreactive CD4+ T lymphocytes as pathogenic effectors and therapeutic targets. The Lancet Neurology 15, 198-209.

Hohlfeld, R., and Steinman, L. (2017). T Cell-Transfer Experimental Autoimmune Encephalomyelitis: Pillar of Multiple Sclerosis and Autoimmunity. Journal of immunology 198, 3381-3383.

Honda, K., Ohba, Y., Yanai, H., Negishi, H., Mizutani, T., Takaoka, A., Taya, C., and Taniguchi, T. (2005). Spatiotemporal regulation of MyD88-IRF-7 signalling for robust type-I interferon induction. Nature 434, 1035-1040.

Hong, S., Zhang, Z., Liu, H., Tian, M., Zhu, X., Zhang, Z., Wang, W., Zhou, X., Zhang, F., Ge, Q., et al. (2018). B Cells Are the Dominant Antigen-Presenting Cells that Activate Naive CD4(+) T Cells upon Immunization with a Virus- Derived Nanoparticle Antigen. Immunity 49, 695-708 e694.

Horst, A.K., Neumann, K., Diehl, L., and Tiegs, G. (2016). Modulation of liver tolerance by conventional and nonconventional antigen-presenting cells and regulatory immune cells. Cellular & molecular immunology 13, 277- 292.

Hou, W., Wu, Y., Sun, S., Shi, M., Sun, Y., Yang, C., Pei, G., Gu, Y., Zhong, C., and Sun, B. (2003). Pertussis toxin enhances Th1 responses by stimulation of dendritic cells. Journal of immunology 170, 1728-1736.

Hsieh, C.S., Lee, H.M., and Lio, C.W. (2012). Selection of regulatory T cells in the thymus. Nature reviews Immunology 12, 157-167.

Huang, B., Zhao, J., Unkeless, J.C., Feng, Z.H., and Xiong, H. (2008). TLR signaling by tumor and immune cells: a double-edged sword. Oncogene 27, 218-224.

Huang, G., Wang, Y., Vogel, P., Kanneganti, T.D., Otsu, K., and Chi, H. (2012). Signaling via the kinase p38alpha programs dendritic cells to drive TH17 differentiation and autoimmune inflammation. Nature immunology 13, 152- 161.

Huang, H.Y., Rivas-Caicedo, A., Renevey, F., Cannelle, H., Peranzoni, E., Scarpellino, L., Hardie, D.L., Pommier, A., Schaeuble, K., Favre, S., et al. (2018). Identification of a new subset of lymph node stromal cells involved in regulating plasma cell homeostasis. Proceedings of the National Academy of Sciences of the United States of America 115, E6826-E6835.

Hui, L., and Chen, Y. (2015). Tumor microenvironment: Sanctuary of the devil. Cancer letters 368, 7-13.

Humbert, M., Guery, L., Brighouse, D., Lemeille, S., and Hugues, S. (2018). Intratumoral CpG-B Promotes Antitumoral Neutrophil, cDC, and T-cell Cooperation without Reprograming Tolerogenic pDC. Cancer research 78, 3280-3292.

Humbert, M., Hugues, S., and Dubrot, J. (2016). Shaping of Peripheral T Cell Responses by Lymphatic Endothelial Cells. Frontiers in immunology 7, 684.

199

Hunter, M.C., Teijeira, A., and Halin, C. (2016). T Cell Trafficking through Lymphatic Vessels. Frontiers in immunology 7, 613.

Huppert, J., Closhen, D., Croxford, A., White, R., Kulig, P., Pietrowski, E., Bechmann, I., Becher, B., Luhmann, H.J., Waisman, A., et al. (2010). Cellular mechanisms of IL-17-induced blood-brain barrier disruption. FASEB journal : official publication of the Federation of American Societies for Experimental Biology 24, 1023-1034.

Iberg, C.A., Jones, A., and Hawiger, D. (2017). Dendritic Cells As Inducers of Peripheral Tolerance. Trends in immunology 38, 793-804.

Inshaw, J.R.J., Cutler, A.J., Burren, O.S., Stefana, M.I., and Todd, J.A. (2018). Approaches and advances in the genetic causes of autoimmune disease and their implications. Nature immunology.

International Multiple Sclerosis Genetics, C., Beecham, A.H., Patsopoulos, N.A., Xifara, D.K., Davis, M.F., Kemppinen, A., Cotsapas, C., Shah, T.S., Spencer, C., Booth, D., et al. (2013). Analysis of immune-related loci identifies 48 new susceptibility variants for multiple sclerosis. Nature genetics 45, 1353-1360.

International Multiple Sclerosis Genetics, C., Hafler, D.A., Compston, A., Sawcer, S., Lander, E.S., Daly, M.J., De Jager, P.L., de Bakker, P.I., Gabriel, S.B., Mirel, D.B., et al. (2007). Risk alleles for multiple sclerosis identified by a genomewide study. The New England journal of medicine 357, 851-862.

International Multiple Sclerosis Genetics, C., Wellcome Trust Case Control, C., Sawcer, S., Hellenthal, G., Pirinen, M., Spencer, C.C., Patsopoulos, N.A., Moutsianas, L., Dilthey, A., Su, Z., et al. (2011). Genetic risk and a primary role for cell-mediated immune mechanisms in multiple sclerosis. Nature 476, 214-219.

Iolyeva, M., Aebischer, D., Proulx, S.T., Willrodt, A.H., Ecoiffier, T., Haner, S., Bouchaud, G., Krieg, C., Onder, L., Ludewig, B., et al. (2013). Interleukin-7 is produced by afferent lymphatic vessels and supports lymphatic drainage. Blood 122, 2271-2281.

Irla, M., Kupfer, N., Suter, T., Lissilaa, R., Benkhoucha, M., Skupsky, J., Lalive, P.H., Fontana, A., Reith, W., and Hugues, S. (2010). MHC class II-restricted antigen presentation by plasmacytoid dendritic cells inhibits T cell-mediated autoimmunity. The Journal of experimental medicine 207, 1891-1905.

Isaksson, M., Ardesjo, B., Ronnblom, L., Kampe, O., Lassmann, H., Eloranta, M.L., and Lobell, A. (2009). Plasmacytoid DC promote priming of autoimmune Th17 cells and EAE. European journal of immunology 39, 2925-2935.

Isaksson, M., Lundgren, B.A., Ahlgren, K.M., Kampe, O., and Lobell, A. (2012). Conditional DC depletion does not affect priming of encephalitogenic Th cells in EAE. European journal of immunology 42, 2555-2563.

Ito, T., Yang, M., Wang, Y.H., Lande, R., Gregorio, J., Perng, O.A., Qin, X.F., Liu, Y.J., and Gilliet, M. (2007). Plasmacytoid dendritic cells prime IL-10-producing T regulatory cells by inducible costimulator ligand. The Journal of experimental medicine 204, 105-115.

Ivashkiv, L.B. (2018). IFNgamma: signalling, epigenetics and roles in immunity, metabolism, disease and cancer immunotherapy. Nature reviews Immunology 18, 545-558.

Jaehn, P.S., Zaenker, K.S., Schmitz, J., and Dzionek, A. (2008). Functional dichotomy of plasmacytoid dendritic cells: antigen-specific activation of T cells versus production of type I interferon. European journal of immunology 38, 1822-1832.

Jager, A., Dardalhon, V., Sobel, R.A., Bettelli, E., and Kuchroo, V.K. (2009). Th1, Th17, and Th9 effector cells induce experimental autoimmune encephalomyelitis with different pathological phenotypes. Journal of immunology 183, 7169-7177.

Jahrsdorfer, B., Vollmer, A., Blackwell, S.E., Maier, J., Sontheimer, K., Beyer, T., Mandel, B., Lunov, O., Tron, K., Nienhaus, G.U., et al. (2010). Granzyme B produced by human plasmacytoid dendritic cells suppresses T-cell expansion. Blood 115, 1156-1165.

Jakubzick, C.V., Randolph, G.J., and Henson, P.M. (2017). Monocyte differentiation and antigen-presenting functions. Nature reviews Immunology 17, 349-362.

Jarrossay, D., Napolitani, G., Colonna, M., Sallusto, F., and Lanzavecchia, A. (2001). Specialization and complementarity in microbial molecule recognition by human myeloid and plasmacytoid dendritic cells. European journal of immunology 31, 3388-3393.

200

Jelcic, I., Al Nimer, F., Wang, J., Lentsch, V., Planas, R., Jelcic, I., Madjovski, A., Ruhrmann, S., Faigle, W., Frauenknecht, K., et al. (2018). Memory B Cells Activate Brain-Homing, Autoreactive CD4(+) T Cells in Multiple Sclerosis. Cell 175, 85-100 e123.

Jeltsch, M., Kaipainen, A., Joukov, V., Meng, X., Lakso, M., Rauvala, H., Swartz, M., Fukumura, D., Jain, R.K., and Alitalo, K. (1997). Hyperplasia of lymphatic vessels in VEGF-C transgenic mice. Science 276, 1423-1425.

Jersild, C., Svejgaard, A., and Fog, T. (1972). HL-A antigens and multiple sclerosis. Lancet 1, 1240-1241.

Jia, L., and Wu, C. (2014). The biology and functions of Th22 cells. Advances in experimental medicine and biology 841, 209-230.

Joffre, O.P., Segura, E., Savina, A., and Amigorena, S. (2012). Cross-presentation by dendritic cells. Nature reviews Immunology 12, 557-569.

Johnson, L.A., Banerji, S., Lawrance, W., Gileadi, U., Prota, G., Holder, K.A., Roshorm, Y.M., Hanke, T., Cerundolo, V., Gale, N.W., et al. (2017). Dendritic cells enter lymph vessels by hyaluronan-mediated docking to the endothelial receptor LYVE-1. Nature immunology 18, 762-770.

Johnson, L.A., Clasper, S., Holt, A.P., Lalor, P.F., Baban, D., and Jackson, D.G. (2006). An inflammation-induced mechanism for leukocyte transmigration across lymphatic vessel endothelium. The Journal of experimental medicine 203, 2763-2777.

Johnson, N.C., Dillard, M.E., Baluk, P., McDonald, D.M., Harvey, N.L., Frase, S.L., and Oliver, G. (2008). Lymphatic endothelial cell identity is reversible and its maintenance requires Prox1 activity. Genes & development 22, 3282- 3291.

Joshi, N.S., Akama-Garren, E.H., Lu, Y., Lee, D.Y., Chang, G.P., Li, A., DuPage, M., Tammela, T., Kerper, N.R., Farago, A.F., et al. (2015). Regulatory T Cells in Tumor-Associated Tertiary Lymphoid Structures Suppress Anti-tumor T Cell Responses. Immunity 43, 579-590.

Kaipainen, A., Korhonen, J., Mustonen, T., van Hinsbergh, V.W., Fang, G.H., Dumont, D., Breitman, M., and Alitalo, K. (1995). Expression of the fms-like tyrosine kinase 4 gene becomes restricted to lymphatic endothelium during development. Proceedings of the National Academy of Sciences of the United States of America 92, 3566-3570.

Kambayashi, T., and Laufer, T.M. (2014). Atypical MHC class II-expressing antigen-presenting cells: can anything replace a dendritic cell? Nature reviews Immunology 14, 719-730.

Kammertoens, T., Friese, C., Arina, A., Idel, C., Briesemeister, D., Rothe, M., Ivanov, A., Szymborska, A., Patone, G., Kunz, S., et al. (2017). Tumour ischaemia by interferon-gamma resembles physiological blood vessel regression. Nature 545, 98-102.

Kanamori, M., Nakatsukasa, H., Okada, M., Lu, Q., and Yoshimura, A. (2016). Induced Regulatory T Cells: Their Development, Stability, and Applications. Trends in immunology 37, 803-811.

Kaplan, M.H., Hufford, M.M., and Olson, M.R. (2015). The development and in vivo function of T helper 9 cells. Nature reviews Immunology 15, 295-307.

Kappos, L., Antel, J., Comi, G., Montalban, X., O'Connor, P., Polman, C.H., Haas, T., Korn, A.A., Karlsson, G., Radue, E.W., et al. (2006). Oral fingolimod (FTY720) for relapsing multiple sclerosis. The New England journal of medicine 355, 1124-1140.

Karsunky, H., Merad, M., Cozzio, A., Weissman, I.L., and Manz, M.G. (2003). Flt3 ligand regulates dendritic cell development from Flt3+ lymphoid and myeloid-committed progenitors to Flt3+ dendritic cells in vivo. The Journal of experimental medicine 198, 305-313.

Kashem, S.W., Haniffa, M., and Kaplan, D.H. (2017). Antigen-Presenting Cells in the Skin. Annual review of immunology 35, 469-499.

Kasper, L.H., and Reder, A.T. (2014). Immunomodulatory activity of interferon-beta. Annals of clinical and translational neurology 1, 622-631.

Kastenmuller, K., Wille-Reece, U., Lindsay, R.W., Trager, L.R., Darrah, P.A., Flynn, B.J., Becker, M.R., Udey, M.C., Clausen, B.E., Igyarto, B.Z., et al. (2011). Protective T cell immunity in mice following protein-TLR7/8 agonist-

201

conjugate immunization requires aggregation, type I IFN, and multiple DC subsets. The Journal of clinical investigation 121, 1782-1796.

Kavousanaki, M., Makrigiannakis, A., Boumpas, D., and Verginis, P. (2010). Novel role of plasmacytoid dendritic cells in humans: induction of interleukin-10-producing Treg cells by plasmacytoid dendritic cells in patients with rheumatoid arthritis responding to therapy. Arthritis and rheumatism 62, 53-63.

Kawanokuchi, J., Shimizu, K., Nitta, A., Yamada, K., Mizuno, T., Takeuchi, H., and Suzumura, A. (2008). Production and functions of IL-17 in microglia. Journal of neuroimmunology 194, 54-61.

Kedl, R.M., Lindsay, R.S., Finlon, J.M., Lucas, E.D., Friedman, R.S., and Tamburini, B.A.J. (2017). Migratory dendritic cells acquire and present lymphatic endothelial cell-archived antigens during lymph node contraction. Nature communications 8, 2034.

Kedl, R.M., and Tamburini, B.A. (2015). Antigen archiving by lymph node stroma: A novel function for the lymphatic endothelium. European journal of immunology 45, 2721-2729.

Kellermayer, Z., Labadi, A., Czompoly, T., Arnold, H.H., and Balogh, P. (2011). Absence of Nkx2-3 homeodomain transcription factor induces the formation of LYVE-1-positive endothelial cysts without lymphatic commitment in the spleen. The journal of histochemistry and cytochemistry : official journal of the Histochemistry Society 59, 690- 700.

Kieback, E., Hilgenberg, E., Stervbo, U., Lampropoulou, V., Shen, P., Bunse, M., Jaimes, Y., Boudinot, P., Radbruch, A., Klemm, U., et al. (2016). Thymus-Derived Regulatory T Cells Are Positively Selected on Natural Self-Antigen through Cognate Interactions of High Functional Avidity. Immunity 44, 1114-1126.

Kim, B.S., Lu, H., Ichiyama, K., Chen, X., Zhang, Y.B., Mistry, N.A., Tanaka, K., Lee, Y.H., Nurieva, R., Zhang, L., et al. (2017). Generation of RORgammat(+) Antigen-Specific T Regulatory 17 Cells from Foxp3(+) Precursors in Autoimmunity. Cell reports 21, 195-207.

King, I.L., Dickendesher, T.L., and Segal, B.M. (2009). Circulating Ly-6C+ myeloid precursors migrate to the CNS and play a pathogenic role during autoimmune demyelinating disease. Blood 113, 3190-3197.

King, I.L., Kroenke, M.A., and Segal, B.M. (2010). GM-CSF-dependent, CD103+ dermal dendritic cells play a critical role in Th effector cell differentiation after subcutaneous immunization. The Journal of experimental medicine 207, 953- 961.

Klein, L., Hinterberger, M., Wirnsberger, G., and Kyewski, B. (2009). Antigen presentation in the thymus for positive selection and central tolerance induction. Nature reviews Immunology 9, 833-844.

Klein, L., Klugmann, M., Nave, K.A., Tuohy, V.K., and Kyewski, B. (2000). Shaping of the autoreactive T-cell repertoire by a splice variant of self protein expressed in thymic epithelial cells. Nature medicine 6, 56-61.

Klein, L., Kyewski, B., Allen, P.M., and Hogquist, K.A. (2014). Positive and negative selection of the T cell repertoire: what thymocytes see (and don't see). Nature reviews Immunology 14, 377-391.

Kleinewietfeld, M., and Hafler, D.A. (2013). The plasticity of human Treg and Th17 cells and its role in autoimmunity. Seminars in immunology 25, 305-312.

Klotz, L., Norman, S., Vieira, J.M., Masters, M., Rohling, M., Dube, K.N., Bollini, S., Matsuzaki, F., Carr, C.A., and Riley, P.R. (2015). Cardiac lymphatics are heterogeneous in origin and respond to injury. Nature 522, 62-67.

Knoblich, K., Cruz Migoni, S., Siew, S.M., Jinks, E., Kaul, B., Jeffery, H.C., Baker, A.T., Suliman, M., Vrzalikova, K., Mehenna, H., et al. (2018). The human lymph node microenvironment unilaterally regulates T-cell activation and differentiation. PLoS biology 16, e2005046.

Koble, C., and Kyewski, B. (2009). The thymic medulla: a unique microenvironment for intercellular self-antigen transfer. The Journal of experimental medicine 206, 1505-1513.

Kobold, S., Wiedemann, G., Rothenfusser, S., and Endres, S. (2014). Modes of action of TLR7 agonists in cancer therapy. Immunotherapy 6, 1085-1095.

Kohli, K., Janssen, A., and Forster, R. (2016). Plasmacytoid dendritic cells induce tolerance predominantly by cargoing antigen to lymph nodes. European journal of immunology 46, 2659-2668.

202

Kohm, A.P., Carpentier, P.A., Anger, H.A., and Miller, S.D. (2002). Cutting edge: CD4+CD25+ regulatory T cells suppress antigen-specific autoreactive immune responses and central nervous system inflammation during active experimental autoimmune encephalomyelitis. Journal of immunology 169, 4712-4716.

Kool, M., Geurtsvankessel, C., Muskens, F., Madeira, F.B., van Nimwegen, M., Kuipers, H., Thielemans, K., Hoogsteden, H.C., Hammad, H., and Lambrecht, B.N. (2011). Facilitated antigen uptake and timed exposure to TLR ligands dictate the antigen-presenting potential of plasmacytoid DCs. Journal of leukocyte biology 90, 1177-1190.

Korn, T., Bettelli, E., Oukka, M., and Kuchroo, V.K. (2009). IL-17 and Th17 Cells. Annual review of immunology 27, 485-517.

Korn, T., Reddy, J., Gao, W., Bettelli, E., Awasthi, A., Petersen, T.R., Backstrom, B.T., Sobel, R.A., Wucherpfennig, K.W., Strom, T.B., et al. (2007). Myelin-specific regulatory T cells accumulate in the CNS but fail to control autoimmune inflammation. Nature medicine 13, 423-431.

Koster, B.D., van den Hout, M., Sluijter, B.J.R., Molenkamp, B.G., Vuylsteke, R., Baars, A., van Leeuwen, P.A.M., Scheper, R.J., Petrousjka van den Tol, M., van den Eertwegh, A.J.M., et al. (2017). Local Adjuvant Treatment with Low-Dose CpG-B Offers Durable Protection against Disease Recurrence in Clinical Stage I-II Melanoma: Data from Two Randomized Phase II Trials. Clinical cancer research : an official journal of the American Association for Cancer Research 23, 5679-5686.

Koyama, M., Hashimoto, D., Aoyama, K., Matsuoka, K., Karube, K., Niiro, H., Harada, M., Tanimoto, M., Akashi, K., and Teshima, T. (2009). Plasmacytoid dendritic cells prime alloreactive T cells to mediate graft-versus-host disease as antigen-presenting cells. Blood 113, 2088-2095.

Krishnamoorthy, G., Lassmann, H., Wekerle, H., and Holz, A. (2006). Spontaneous opticospinal encephalomyelitis in a double-transgenic mouse model of autoimmune T cell/B cell cooperation. The Journal of clinical investigation 116, 2385-2392.

Krug, A., Veeraswamy, R., Pekosz, A., Kanagawa, O., Unanue, E.R., Colonna, M., and Cella, M. (2003). Interferon- producing cells fail to induce proliferation of naive T cells but can promote expansion and T helper 1 differentiation of antigen-experienced unpolarized T cells. The Journal of experimental medicine 197, 899-906.

Kruse, N., Neumann, K., Schrage, A., Derkow, K., Schott, E., Erben, U., Kuhl, A., Loddenkemper, C., Zeitz, M., Hamann, A., et al. (2009). Priming of CD4+ T cells by liver sinusoidal endothelial cells induces CD25low forkhead box protein 3- regulatory T cells suppressing autoimmune hepatitis. Hepatology 50, 1904-1913.

Kukk, E., Lymboussaki, A., Taira, S., Kaipainen, A., Jeltsch, M., Joukov, V., and Alitalo, K. (1996). VEGF-C receptor binding and pattern of expression with VEGFR-3 suggests a role in lymphatic vascular development. Development 122, 3829-3837.

Kunicki, M.A., Amaya Hernandez, L.C., Davis, K.L., Bacchetta, R., and Roncarolo, M.G. (2018). Identity and Diversity of Human Peripheral Th and T Regulatory Cells Defined by Single-Cell Mass Cytometry. Journal of immunology 200, 336-346.

Kurts, C. (2000). Cross-presentation: inducing CD8 T cell immunity and tolerance. Journal of molecular medicine 78, 326- 332.

Kyewski, B., Rottinger, B., and Klein, L. (2000). Making central T-cell tolerance efficient: thymic stromal cells sample distinct self-antigen pools. Current topics in microbiology and immunology 251, 139-145.

Labidi-Galy, S.I., Sisirak, V., Meeus, P., Gobert, M., Treilleux, I., Bajard, A., Combes, J.D., Faget, J., Mithieux, F., Cassignol, A., et al. (2011). Quantitative and functional alterations of plasmacytoid dendritic cells contribute to immune tolerance in ovarian cancer. Cancer research 71, 5423-5434.

Labidi-Galy, S.I., Treilleux, I., Goddard-Leon, S., Combes, J.D., Blay, J.Y., Ray-Coquard, I., Caux, C., and Bendriss- Vermare, N. (2012). Plasmacytoid dendritic cells infiltrating ovarian cancer are associated with poor prognosis. Oncoimmunology 1, 380-382.

Lahl, K., Loddenkemper, C., Drouin, C., Freyer, J., Arnason, J., Eberl, G., Hamann, A., Wagner, H., Huehn, J., and Sparwasser, T. (2007). Selective depletion of Foxp3+ regulatory T cells induces a scurfy-like disease. The Journal of experimental medicine 204, 57-63.

203

Lalor, S.J., and Segal, B.M. (2013). Th1-mediated experimental autoimmune encephalomyelitis is CXCR3 independent. European journal of immunology 43, 2866-2874.

Lammermann, T., Bader, B.L., Monkley, S.J., Worbs, T., Wedlich-Soldner, R., Hirsch, K., Keller, M., Forster, R., Critchley, D.R., Fassler, R., et al. (2008). Rapid leukocyte migration by integrin-independent flowing and squeezing. Nature 453, 51-55.

Lande, R., Ganguly, D., Facchinetti, V., Frasca, L., Conrad, C., Gregorio, J., Meller, S., Chamilos, G., Sebasigari, R., Riccieri, V., et al. (2011). Neutrophils activate plasmacytoid dendritic cells by releasing self-DNA-peptide complexes in systemic lupus erythematosus. Science translational medicine 3, 73ra19.

Lande, R., Gregorio, J., Facchinetti, V., Chatterjee, B., Wang, Y.H., Homey, B., Cao, W., Wang, Y.H., Su, B., Nestle, F.O., et al. (2007). Plasmacytoid dendritic cells sense self-DNA coupled with antimicrobial peptide. Nature 449, 564-569.

Landgraf, M., McGovern, J.A., Friedl, P., and Hutmacher, D.W. (2018). Rational Design of Mouse Models for Cancer Research. Trends in biotechnology 36, 242-251.

Lang, H.L., Jacobsen, H., Ikemizu, S., Andersson, C., Harlos, K., Madsen, L., Hjorth, P., Sondergaard, L., Svejgaard, A., Wucherpfennig, K., et al. (2002). A functional and structural basis for TCR cross-reactivity in multiple sclerosis. Nature immunology 3, 940-943.

Langrish, C.L., Chen, Y., Blumenschein, W.M., Mattson, J., Basham, B., Sedgwick, J.D., McClanahan, T., Kastelein, R.A., and Cua, D.J. (2005). IL-23 drives a pathogenic T cell population that induces autoimmune inflammation. The Journal of experimental medicine 201, 233-240.

LaSalle, J.M., Tolentino, P.J., Freeman, G.J., Nadler, L.M., and Hafler, D.A. (1992). Early signaling defects in human T cells anergized by T cell presentation of autoantigen. The Journal of experimental medicine 176, 177-186.

Lavado, A., and Oliver, G. (2007). Prox1 expression patterns in the developing and adult murine brain. Developmental dynamics : an official publication of the American Association of Anatomists 236, 518-524.

Le Mercier, I., Poujol, D., Sanlaville, A., Sisirak, V., Gobert, M., Durand, I., Dubois, B., Treilleux, I., Marvel, J., Vlach, J., et al. (2013). Tumor promotion by intratumoral plasmacytoid dendritic cells is reversed by TLR7 ligand treatment. Cancer research 73, 4629-4640.

Lee, J.W., Epardaud, M., Sun, J., Becker, J.E., Cheng, A.C., Yonekura, A.R., Heath, J.K., and Turley, S.J. (2007). Peripheral antigen display by lymph node stroma promotes T cell tolerance to intestinal self. Nature immunology 8, 181-190.

Legoux, F.P., Lim, J.B., Cauley, A.W., Dikiy, S., Ertelt, J., Mariani, T.J., Sparwasser, T., Way, S.S., and Moon, J.J. (2015). CD4+ T Cell Tolerance to Tissue-Restricted Self Antigens Is Mediated by Antigen-Specific Regulatory T Cells Rather Than Deletion. Immunity 43, 896-908.

Leibovitch, E.C., Caruso, B., Ha, S.K., Schindler, M.K., Lee, N.J., Luciano, N.J., Billioux, B.J., Guy, J.R., Yen, C., Sati, P., et al. (2018). Herpesvirus trigger accelerates neuroinflammation in a nonhuman primate model of multiple sclerosis. Proceedings of the National Academy of Sciences of the United States of America 115, 11292-11297.

LeibundGut-Landmann, S., Waldburger, J.M., Reis e Sousa, C., Acha-Orbea, H., and Reith, W. (2004). MHC class II expression is differentially regulated in plasmacytoid and conventional dendritic cells. Nature immunology 5, 899- 908.

Lennert, K., and Remmele, W. (1958). [Karyometric research on lymph node cells in man. I. Germinoblasts, lymphoblasts & lymphocytes]. Acta haematologica 19, 99-113.

Leong, S.P., Nakakura, E.K., Pollock, R., Choti, M.A., Morton, D.L., Henner, W.D., Lal, A., Pillai, R., Clark, O.H., and Cady, B. (2011). Unique patterns of metastases in common and rare types of malignancy. Journal of surgical oncology 103, 607-614.

Leung, C.S. (2015). Endogenous Antigen Presentation of MHC Class II Epitopes through Non-Autophagic Pathways. Frontiers in immunology 6, 464.

Levesque, S.A., Pare, A., Mailhot, B., Bellver-Landete, V., Kebir, H., Lecuyer, M.A., Alvarez, J.I., Prat, A., de Rivero Vaccari, J.P., Keane, R.W., et al. (2016). Myeloid cell transmigration across the CNS vasculature triggers IL-1beta- driven neuroinflammation during autoimmune encephalomyelitis in mice. The Journal of experimental medicine 213, 929-949.

204

Li, M.O., and Rudensky, A.Y. (2016). T cell receptor signalling in the control of regulatory T cell differentiation and function. Nature reviews Immunology 16, 220-233.

Li, R., Patterson, K.R., and Bar-Or, A. (2018). Reassessing B cell contributions in multiple sclerosis. Nature immunology.

Lim, K., Hyun, Y.M., Lambert-Emo, K., Capece, T., Bae, S., Miller, R., Topham, D.J., and Kim, M. (2015). Neutrophil trails guide influenza-specific CD8(+) T cells in the airways. Science 349, aaa4352.

Limmer, A., Ohl, J., Kurts, C., Ljunggren, H.G., Reiss, Y., Groettrup, M., Momburg, F., Arnold, B., and Knolle, P.A. (2000). Efficient presentation of exogenous antigen by liver endothelial cells to CD8+ T cells results in antigen- specific T-cell tolerance. Nature medicine 6, 1348-1354.

Lindau, D., Mussard, J., Wagner, B.J., Ribon, M., Ronnefarth, V.M., Quettier, M., Jelcic, I., Boissier, M.C., Rammensee, H.G., and Decker, P. (2013). Primary blood neutrophils express a functional cell surface Toll-like receptor 9. European journal of immunology 43, 2101-2113.

Link, A., Vogt, T.K., Favre, S., Britschgi, M.R., Acha-Orbea, H., Hinz, B., Cyster, J.G., and Luther, S.A. (2007). Fibroblastic reticular cells in lymph nodes regulate the homeostasis of naive T cells. Nature immunology 8, 1255- 1265.

Linthicum, D.S., Munoz, J.J., and Blaskett, A. (1982). Acute experimental autoimmune encephalomyelitis in mice. I. Adjuvant action of Bordetella pertussis is due to vasoactive amine sensitization and increased vascular permeability of the central nervous system. Cellular immunology 73, 299-310.

Lippens, C., Duraes, F.V., Dubrot, J., Brighouse, D., Lacroix, M., Irla, M., Aubry-Lachainaye, J.P., Reith, W., Mandl, J.N., and Hugues, S. (2016). IDO-orchestrated crosstalk between pDCs and Tregs inhibits autoimmunity. Journal of autoimmunity 75, 39-49.

Liston, A., Gray, D.H., Lesage, S., Fletcher, A.L., Wilson, J., Webster, K.E., Scott, H.S., Boyd, R.L., Peltonen, L., and Goodnow, C.C. (2004). Gene dosage--limiting role of Aire in thymic expression, clonal deletion, and organ-specific autoimmunity. The Journal of experimental medicine 200, 1015-1026.

Liu, C., Lou, Y., Lizee, G., Qin, H., Liu, S., Rabinovich, B., Kim, G.J., Wang, Y.H., Ye, Y., Sikora, A.G., et al. (2008a). Plasmacytoid dendritic cells induce NK cell-dependent, tumor antigen-specific T cell cross-priming and tumor regression in mice. The Journal of clinical investigation 118, 1165-1175.

Liu, K., Victora, G.D., Schwickert, T.A., Guermonprez, P., Meredith, M.M., Yao, K., Chu, F.F., Randolph, G.J., Rudensky, A.Y., and Nussenzweig, M. (2009). In vivo analysis of dendritic cell development and homeostasis. Science 324, 392-397.

Liu, X., Mishra, P., Yu, S., Beckmann, J., Wendland, M., Kocks, J., Seth, S., Hoffmann, K., Hoffmann, M., Kremmer, E., et al. (2011). Tolerance induction towards cardiac allografts under costimulation blockade is impaired in CCR7- deficient animals but can be restored by adoptive transfer of syngeneic plasmacytoid dendritic cells. European journal of immunology 41, 611-623.

Liu, Y., Teige, I., Birnir, B., and Issazadeh-Navikas, S. (2006). Neuron-mediated generation of regulatory T cells from encephalitogenic T cells suppresses EAE. Nature medicine 12, 518-525.

Liu, Y., Zhang, P., Li, J., Kulkarni, A.B., Perruche, S., and Chen, W. (2008b). A critical function for TGF-beta signaling in the development of natural CD4+CD25+Foxp3+ regulatory T cells. Nature immunology 9, 632-640.

Liu, Y.J. (2005). IPC: professional type 1 interferon-producing cells and plasmacytoid dendritic cell precursors. Annual review of immunology 23, 275-306.

Lohse, A.W., Dinkelmann, M., Kimmig, M., Herkel, J., and Meyer zum Buschenfelde, K.H. (1996). Estimation of the frequency of self-reactive T cells in health and inflammatory diseases by limiting dilution analysis and single cell cloning. Journal of autoimmunity 9, 667-675.

Lombardi, V.C., Khaiboullina, S.F., and Rizvanov, A.A. (2015). Plasmacytoid dendritic cells, a role in neoplastic prevention and progression. European journal of clinical investigation 45 Suppl 1, 1-8.

Lopes Pinheiro, M.A., Kamermans, A., Garcia-Vallejo, J.J., van Het Hof, B., Wierts, L., O'Toole, T., Boeve, D., Verstege, M., van der Pol, S.M., van Kooyk, Y., et al. (2016). Internalization and presentation of myelin antigens by the brain endothelium guides antigen-specific T cell migration. eLife 5.

205

Lopez-Diego, R.S., and Weiner, H.L. (2008). Novel therapeutic strategies for multiple sclerosis--a multifaceted adversary. Nature reviews Drug discovery 7, 909-925.

Loschko, J., Heink, S., Hackl, D., Dudziak, D., Reindl, W., Korn, T., and Krug, A.B. (2011a). Antigen targeting to plasmacytoid dendritic cells via Siglec-H inhibits Th cell-dependent autoimmunity. Journal of immunology 187, 6346-6356.

Loschko, J., and Krug, A. (2012). Antigen delivery to plasmacytoid dendritic cells -induction of tolerance and immunity. Critical reviews in immunology 32, 489-501.

Loschko, J., Schlitzer, A., Dudziak, D., Drexler, I., Sandholzer, N., Bourquin, C., Reindl, W., and Krug, A.B. (2011b). Antigen delivery to plasmacytoid dendritic cells via BST2 induces protective T cell-mediated immunity. Journal of immunology 186, 6718-6725.

Lou, Y., Liu, C., Lizee, G., Peng, W., Xu, C., Ye, Y., Rabinovich, B.A., Hailemichael, Y., Gelbard, A., Zhou, D., et al. (2011). Antitumor activity mediated by CpG: the route of administration is critical. Journal of immunotherapy 34, 279-288.

Louveau, A., Harris, T.H., and Kipnis, J. (2015a). Revisiting the Mechanisms of CNS Immune Privilege. Trends in immunology 36, 569-577.

Louveau, A., Herz, J., Alme, M.N., Salvador, A.F., Dong, M.Q., Viar, K.E., Herod, S.G., Knopp, J., Setliff, J.C., Lupi, A.L., et al. (2018). CNS lymphatic drainage and neuroinflammation are regulated by meningeal lymphatic vasculature. Nature neuroscience 21, 1380-1391.

Louveau, A., Plog, B.A., Antila, S., Alitalo, K., Nedergaard, M., and Kipnis, J. (2017). Understanding the functions and relationships of the glymphatic system and meningeal lymphatics. The Journal of clinical investigation 127, 3210- 3219.

Louveau, A., Smirnov, I., Keyes, T.J., Eccles, J.D., Rouhani, S.J., Peske, J.D., Derecki, N.C., Castle, D., Mandell, J.W., Lee, K.S., et al. (2015b). Structural and functional features of central nervous system lymphatic vessels. Nature 523, 337- 341.

Lovett-Racke, A.E., Rocchini, A.E., Choy, J., Northrop, S.C., Hussain, R.Z., Ratts, R.B., Sikder, D., and Racke, M.K. (2004). Silencing T-bet defines a critical role in the differentiation of autoreactive T lymphocytes. Immunity 21, 719- 731.

Lu, L., Barbi, J., and Pan, F. (2017). The regulation of immune tolerance by FOXP3. Nature reviews Immunology 17, 703-717.

Lublin, F., Miller, D.H., Freedman, M.S., Cree, B.A.C., Wolinsky, J.S., Weiner, H., Lubetzki, C., Hartung, H.P., Montalban, X., Uitdehaag, B.M.J., et al. (2016). Oral fingolimod in primary progressive multiple sclerosis (INFORMS): a phase 3, randomised, double-blind, placebo-controlled trial. Lancet 387, 1075-1084.

Lucas, E.D., Finlon, J.M., Burchill, M.A., McCarthy, M.K., Morrison, T.E., Colpitts, T.M., and Tamburini, B.A.J. (2018). Type 1 IFN and PD-L1 Coordinate Lymphatic Endothelial Cell Expansion and Contraction during an Inflammatory Immune Response. Journal of immunology 201, 1735-1747.

Lui, G., Manches, O., Angel, J., Molens, J.P., Chaperot, L., and Plumas, J. (2009). Plasmacytoid dendritic cells capture and cross-present viral antigens from influenza-virus exposed cells. PloS one 4, e7111.

Lukacs-Kornek, V. (2016). The Role of Lymphatic Endothelial Cells in Liver Injury and Tumor Development. Frontiers in immunology 7, 548.

Lukacs-Kornek, V., Burgdorf, S., Diehl, L., Specht, S., Kornek, M., and Kurts, C. (2008). The kidney-renal lymph node- system contributes to cross-tolerance against innocuous circulating antigen. Journal of immunology 180, 706-715.

Lund, A.W., Duraes, F.V., Hirosue, S., Raghavan, V.R., Nembrini, C., Thomas, S.N., Issa, A., Hugues, S., and Swartz, M.A. (2012). VEGF-C promotes immune tolerance in B16 melanomas and cross-presentation of tumor antigen by lymph node lymphatics. Cell reports 1, 191-199.

Lund, A.W., Wagner, M., Fankhauser, M., Steinskog, E.S., Broggi, M.A., Spranger, S., Gajewski, T.F., Alitalo, K., Eikesdal, H.P., Wiig, H., et al. (2016). Lymphatic vessels regulate immune microenvironments in human and murine melanoma. The Journal of clinical investigation 126, 3389-3402.

206

Lund, J.M., Alexopoulou, L., Sato, A., Karow, M., Adams, N.C., Gale, N.W., Iwasaki, A., and Flavell, R.A. (2004). Recognition of single-stranded RNA viruses by Toll-like receptor 7. Proceedings of the National Academy of Sciences of the United States of America 101, 5598-5603.

Luth, S., Huber, S., Schramm, C., Buch, T., Zander, S., Stadelmann, C., Bruck, W., Wraith, D.C., Herkel, J., and Lohse, A.W. (2008). Ectopic expression of neural autoantigen in mouse liver suppresses experimental autoimmune neuroinflammation by inducing antigen-specific Tregs. The Journal of clinical investigation 118, 3403-3410.

Luther, S.A., Bidgol, A., Hargreaves, D.C., Schmidt, A., Xu, Y., Paniyadi, J., Matloubian, M., and Cyster, J.G. (2002). Differing activities of homeostatic chemokines CCL19, CCL21, and CXCL12 in lymphocyte and dendritic cell recruitment and lymphoid neogenesis. Journal of immunology 169, 424-433.

Lyssiotis, C.A., and Kimmelman, A.C. (2017). Metabolic Interactions in the Tumor Microenvironment. Trends in cell biology 27, 863-875.

Macal, M., Jo, Y., Dallari, S., Chang, A.Y., Dai, J., Swaminathan, S., Wehrens, E.J., Fitzgerald-Bocarsly, P., and Zuniga, E.I. (2018). Self-Renewal and Toll-like Receptor Signaling Sustain Exhausted Plasmacytoid Dendritic Cells during Chronic Viral Infection. Immunity 48, 730-744 e735.

Magnusson, F.C., Liblau, R.S., von Boehmer, H., Pittet, M.J., Lee, J.W., Turley, S.J., and Khazaie, K. (2008). Direct presentation of antigen by lymph node stromal cells protects against CD8 T-cell-mediated intestinal autoimmunity. Gastroenterology 134, 1028-1037.

Maisel, K., Sasso, M.S., Potin, L., and Swartz, M.A. (2017). Exploiting lymphatic vessels for immunomodulation: Rationale, opportunities, and challenges. Advanced drug delivery reviews 114, 43-59.

Malhotra, D., Fletcher, A.L., Astarita, J., Lukacs-Kornek, V., Tayalia, P., Gonzalez, S.F., Elpek, K.G., Chang, S.K., Knoblich, K., Hemler, M.E., et al. (2012). Transcriptional profiling of stroma from inflamed and resting lymph nodes defines immunological hallmarks. Nature immunology 13, 499-510.

Malhotra, D., Fletcher, A.L., and Turley, S.J. (2013). Stromal and hematopoietic cells in secondary lymphoid organs: partners in immunity. Immunological reviews 251, 160-176.

Manches, O., Munn, D., Fallahi, A., Lifson, J., Chaperot, L., Plumas, J., and Bhardwaj, N. (2008). HIV-activated human plasmacytoid DCs induce Tregs through an indoleamine 2,3-dioxygenase-dependent mechanism. The Journal of clinical investigation 118, 3431-3439.

Mandl, J.N., Monteiro, J.P., Vrisekoop, N., and Germain, R.N. (2013). T cell-positive selection uses self-ligand binding strength to optimize repertoire recognition of foreign antigens. Immunity 38, 263-274.

Mann, E.R., and Li, X. (2014). Intestinal antigen-presenting cells in mucosal immune homeostasis: crosstalk between dendritic cells, macrophages and B-cells. World journal of gastroenterology 20, 9653-9664.

Marabelle, A., Kohrt, H., Caux, C., and Levy, R. (2014). Intratumoral immunization: a new paradigm for cancer therapy. Clinical cancer research : an official journal of the American Association for Cancer Research 20, 1747-1756.

Marta, M., Andersson, A., Isaksson, M., Kampe, O., and Lobell, A. (2008). Unexpected regulatory roles of TLR4 and TLR9 in experimental autoimmune encephalomyelitis. European journal of immunology 38, 565-575.

Martin-Gayo, E., Sierra-Filardi, E., Corbi, A.L., and Toribio, M.L. (2010). Plasmacytoid dendritic cells resident in human thymus drive natural Treg cell development. Blood 115, 5366-5375.

Martin, B.N., Wang, C., Zhang, C.J., Kang, Z., Gulen, M.F., Zepp, J.A., Zhao, J., Bian, G., Do, J.S., Min, B., et al. (2016). T cell-intrinsic ASC critically promotes T(H)17-mediated experimental autoimmune encephalomyelitis. Nature immunology 17, 583-592.

Martinez-Corral, I., Stanczuk, L., Frye, M., Ulvmar, M.H., Dieguez-Hurtado, R., Olmeda, D., Makinen, T., and Ortega, S. (2016). Vegfr3-CreER (T2) mouse, a new genetic tool for targeting the lymphatic system. Angiogenesis 19, 433-445.

Massara, M., Persico, P., Bonavita, O., Mollica Poeta, V., Locati, M., Simonelli, M., and Bonecchi, R. (2017). Neutrophils in Gliomas. Frontiers in immunology 8, 1349.

Matta, B.M., Castellaneta, A., and Thomson, A.W. (2010). Tolerogenic plasmacytoid DC. European journal of immunology 40, 2667-2676.

207

Matusevicius, D., Kivisakk, P., He, B., Kostulas, N., Ozenci, V., Fredrikson, S., and Link, H. (1999). Interleukin-17 mRNA expression in blood and CSF mononuclear cells is augmented in multiple sclerosis. Multiple sclerosis 5, 101- 104.

Mbeunkui, F., and Johann, D.J., Jr. (2009). Cancer and the tumor microenvironment: a review of an essential relationship. Cancer chemotherapy and pharmacology 63, 571-582.

McGeachy, M.J., Stephens, L.A., and Anderton, S.M. (2005). Natural recovery and protection from autoimmune encephalomyelitis: contribution of CD4+CD25+ regulatory cells within the central nervous system. Journal of immunology 175, 3025-3032.

McQualter, J.L., Darwiche, R., Ewing, C., Onuki, M., Kay, T.W., Hamilton, J.A., Reid, H.H., and Bernard, C.C. (2001). Granulocyte macrophage colony-stimulating factor: a new putative therapeutic target in multiple sclerosis. The Journal of experimental medicine 194, 873-882.

Mebius, R.E., and Kraal, G. (2005). Structure and function of the spleen. Nature reviews Immunology 5, 606-616.

Mehrfeld, C., Zenner, S., Kornek, M., and Lukacs-Kornek, V. (2018). The Contribution of Non-Professional Antigen- Presenting Cells to Immunity and Tolerance in the Liver. Frontiers in immunology 9, 635.

Mellman, I., Coukos, G., and Dranoff, G. (2011). Cancer immunotherapy comes of age. Nature 480, 480-489.

Menager-Marcq, I., Pomie, C., Romagnoli, P., and van Meerwijk, J.P. (2006). CD8+CD28- regulatory T lymphocytes prevent experimental inflammatory bowel disease in mice. Gastroenterology 131, 1775-1785.

Mendel, I., Kerlero de Rosbo, N., and Ben-Nun, A. (1995). A myelin oligodendrocyte glycoprotein peptide induces typical chronic experimental autoimmune encephalomyelitis in H-2b mice: fine specificity and T cell receptor V beta expression of encephalitogenic T cells. European journal of immunology 25, 1951-1959.

Menon, M., Blair, P.A., Isenberg, D.A., and Mauri, C. (2016). A Regulatory Feedback between Plasmacytoid Dendritic Cells and Regulatory B Cells Is Aberrant in Systemic Lupus Erythematosus. Immunity 44, 683-697.

Merad, M., Sathe, P., Helft, J., Miller, J., and Mortha, A. (2013). The dendritic cell lineage: ontogeny and function of dendritic cells and their subsets in the steady state and the inflamed setting. Annual review of immunology 31, 563- 604.

Metzger, T.C., and Anderson, M.S. (2011). Control of central and peripheral tolerance by Aire. Immunological reviews 241, 89-103.

Meyer-Wentrup, F., Benitez-Ribas, D., Tacken, P.J., Punt, C.J., Figdor, C.G., de Vries, I.J., and Adema, G.J. (2008). Targeting DCIR on human plasmacytoid dendritic cells results in antigen presentation and inhibits IFN-alpha production. Blood 111, 4245-4253.

Meyer, C., Martin-Blondel, G., and Liblau, R.S. (2017). Endothelial cells and lymphatics at the interface between the immune and central nervous systems: implications for multiple sclerosis. Current opinion in neurology 30, 222-230.

Mezey, E., and Palkovits, M. (2015). Neuroanatomy: Forgotten findings of brain lymphatics. Nature 524, 415.

Michea, P., Noel, F., Zakine, E., Czerwinska, U., Sirven, P., Abouzid, O., Goudot, C., Scholer-Dahirel, A., Vincent- Salomon, A., Reyal, F., et al. (2018). Adjustment of dendritic cells to the breast-cancer microenvironment is subset specific. Nature immunology 19, 885-897.

Miller, C.N., Hartigan-O'Connor, D.J., Lee, M.S., Laidlaw, G., Cornelissen, I.P., Matloubian, M., Coughlin, S.R., McDonald, D.M., and McCune, J.M. (2013). IL-7 production in murine lymphatic endothelial cells and induction in the setting of peripheral lymphopenia. International immunology 25, 471-483.

Milo, R., and Kahana, E. (2010). Multiple sclerosis: geoepidemiology, genetics and the environment. Autoimmunity reviews 9, A387-394.

Minn, A.J. (2015). Interferons and the Immunogenic Effects of Cancer Therapy. Trends in immunology 36, 725-737.

Mirchandani, A.S., Besnard, A.G., Yip, E., Scott, C., Bain, C.C., Cerovic, V., Salmond, R.J., and Liew, F.Y. (2014). Type 2 innate lymphoid cells drive CD4+ Th2 cell responses. Journal of immunology 192, 2442-2448.

208

Moffat, J.M., Segura, E., Khoury, G., Caminschi, I., Cameron, P.U., Lewin, S.R., Villadangos, J.A., and Mintern, J.D. (2013). Targeting antigen to bone marrow stromal cell-2 expressed by conventional and plasmacytoid dendritic cells elicits efficient antigen presentation. European journal of immunology 43, 595-605.

Mohammad, M.G., Tsai, V.W., Ruitenberg, M.J., Hassanpour, M., Li, H., Hart, P.H., Breit, S.N., Sawchenko, P.E., and Brown, D.A. (2014). Immune cell trafficking from the brain maintains CNS immune tolerance. The Journal of clinical investigation 124, 1228-1241.

Montealegre, S., and van Endert, P.M. (2018). Endocytic Recycling of MHC Class I Molecules in Non-professional Antigen Presenting and Dendritic Cells. Frontiers in immunology 9, 3098.

Moon, J.J., Chu, H.H., Pepper, M., McSorley, S.J., Jameson, S.C., Kedl, R.M., and Jenkins, M.K. (2007). Naive CD4(+) T cell frequency varies for different epitopes and predicts repertoire diversity and response magnitude. Immunity 27, 203-213.

Moseman, E.A., Liang, X., Dawson, A.J., Panoskaltsis-Mortari, A., Krieg, A.M., Liu, Y.J., Blazar, B.R., and Chen, W. (2004). Human plasmacytoid dendritic cells activated by CpG oligodeoxynucleotides induce the generation of CD4+CD25+ regulatory T cells. Journal of immunology 173, 4433-4442.

Moser, B., and Brandes, M. (2006). Gammadelta T cells: an alternative type of professional APC. Trends in immunology 27, 112-118.

Mosmann, T.R., and Coffman, R.L. (1989). TH1 and TH2 cells: different patterns of lymphokine secretion lead to different functional properties. Annual review of immunology 7, 145-173.

Mouries, J., Moron, G., Schlecht, G., Escriou, N., Dadaglio, G., and Leclerc, C. (2008). Plasmacytoid dendritic cells efficiently cross-prime naive T cells in vivo after TLR activation. Blood 112, 3713-3722.

Mrdjen, D., Pavlovic, A., Hartmann, F.J., Schreiner, B., Utz, S.G., Leung, B.P., Lelios, I., Heppner, F.L., Kipnis, J., Merkler, D., et al. (2018). High-Dimensional Single-Cell Mapping of Central Nervous System Immune Cells Reveals Distinct Myeloid Subsets in Health, Aging, and Disease. Immunity 48, 380-395 e386.

Mueller, S.N., and Ahmed, R. (2008). Lymphoid stroma in the initiation and control of immune responses. Immunological reviews 224, 284-294.

Munger, K.L., Zhang, S.M., O'Reilly, E., Hernan, M.A., Olek, M.J., Willett, W.C., and Ascherio, A. (2004). Vitamin D intake and incidence of multiple sclerosis. Neurology 62, 60-65.

Munn, D.H., Sharma, M.D., Hou, D., Baban, B., Lee, J.R., Antonia, S.J., Messina, J.L., Chandler, P., Koni, P.A., and Mellor, A.L. (2004). Expression of indoleamine 2,3-dioxygenase by plasmacytoid dendritic cells in tumor-draining lymph nodes. The Journal of clinical investigation 114, 280-290.

Munz, C. (2015). Of LAP, CUPS, and DRibbles - Unconventional Use of Autophagy Proteins for MHC Restricted Antigen Presentation. Frontiers in immunology 6, 200.

Murphy, K., Travers, P., Walport, M., and Janeway, C. (2008). Janeway's immunobiology, 7th edn (New York: Garland Science).

Naito, S., Namerow, N., Mickey, M.R., and Terasaki, P.I. (1972). Multiple sclerosis: association with HL-A3. Tissue antigens 2, 1-4.

Nakayama, M. (2014). Antigen Presentation by MHC-Dressed Cells. Frontiers in immunology 5, 672.

Neefjes, J., Jongsma, M.L., Paul, P., and Bakke, O. (2011). Towards a systems understanding of MHC class I and MHC class II antigen presentation. Nature reviews Immunology 11, 823-836.

Neuhaus, O., Farina, C., Yassouridis, A., Wiendl, H., Then Bergh, F., Dose, T., Wekerle, H., and Hohlfeld, R. (2000). Multiple sclerosis: comparison of copolymer-1- reactive T cell lines from treated and untreated subjects reveals cytokine shift from T helper 1 to T helper 2 cells. Proceedings of the National Academy of Sciences of the United States of America 97, 7452-7457.

Neyt, K., Perros, F., GeurtsvanKessel, C.H., Hammad, H., and Lambrecht, B.N. (2012). Tertiary lymphoid organs in infection and autoimmunity. Trends in immunology 33, 297-305.

209

Nicenboim, J., Malkinson, G., Lupo, T., Asaf, L., Sela, Y., Mayseless, O., Gibbs-Bar, L., Senderovich, N., Hashimshony, T., Shin, M., et al. (2015). Lymphatic vessels arise from specialized angioblasts within a venous niche. Nature 522, 56- 61.

Nichols, L.A., Chen, Y., Colella, T.A., Bennett, C.L., Clausen, B.E., and Engelhard, V.H. (2007). Deletional self-tolerance to a melanocyte/melanoma antigen derived from tyrosinase is mediated by a radio-resistant cell in peripheral and mesenteric lymph nodes. Journal of immunology 179, 993-1003.

Nicolas-Avila, J.A., Adrover, J.M., and Hidalgo, A. (2017). Neutrophils in Homeostasis, Immunity, and Cancer. Immunity 46, 15-28.

Nierkens, S., den Brok, M.H., Garcia, Z., Togher, S., Wagenaars, J., Wassink, M., Boon, L., Ruers, T.J., Figdor, C.G., Schoenberger, S.P., et al. (2011). Immune adjuvant efficacy of CpG oligonucleotide in cancer treatment is founded specifically upon TLR9 function in plasmacytoid dendritic cells. Cancer research 71, 6428-6437.

Nierkens, S., Tel, J., Janssen, E., and Adema, G.J. (2013). Antigen cross-presentation by dendritic cell subsets: one general or all sergeants? Trends in immunology 34, 361-370.

Norder, M., Gutierrez, M.G., Zicari, S., Cervi, E., Caruso, A., and Guzman, C.A. (2012). Lymph node-derived lymphatic endothelial cells express functional costimulatory molecules and impair dendritic cell-induced allogenic T-cell proliferation. FASEB journal : official publication of the Federation of American Societies for Experimental Biology 26, 2835-2846.

O'Connor, P., Wolinsky, J.S., Confavreux, C., Comi, G., Kappos, L., Olsson, T.P., Benzerdjeb, H., Truffinet, P., Wang, L., Miller, A., et al. (2011). Randomized trial of oral teriflunomide for relapsing multiple sclerosis. The New England journal of medicine 365, 1293-1303.

Obar, J.J., Khanna, K.M., and Lefrancois, L. (2008). Endogenous naive CD8+ T cell precursor frequency regulates primary and memory responses to infection. Immunity 28, 859-869.

Oberkampf, M., Guillerey, C., Mouries, J., Rosenbaum, P., Fayolle, C., Bobard, A., Savina, A., Ogier-Denis, E., Enninga, J., Amigorena, S., et al. (2018). Mitochondrial reactive oxygen species regulate the induction of CD8(+) T cells by plasmacytoid dendritic cells. Nature communications 9, 2241.

Ohnmacht, C., Pullner, A., King, S.B., Drexler, I., Meier, S., Brocker, T., and Voehringer, D. (2009). Constitutive ablation of dendritic cells breaks self-tolerance of CD4 T cells and results in spontaneous fatal autoimmunity. The Journal of experimental medicine 206, 549-559.

Oliphant, C.J., Hwang, Y.Y., Walker, J.A., Salimi, M., Wong, S.H., Brewer, J.M., Englezakis, A., Barlow, J.L., Hams, E., Scanlon, S.T., et al. (2014). MHCII-mediated dialog between group 2 innate lymphoid cells and CD4(+) T cells potentiates type 2 immunity and promotes parasitic helminth expulsion. Immunity 41, 283-295.

Oliveira, C.C., and van Hall, T. (2015). Alternative Antigen Processing for MHC Class I: Multiple Roads Lead to Rome. Frontiers in immunology 6, 298.

Oliver, G. (2004). Lymphatic vasculature development. Nature reviews Immunology 4, 35-45.

Olsson, T., Barcellos, L.F., and Alfredsson, L. (2017). Interactions between genetic, lifestyle and environmental risk factors for multiple sclerosis. Nature reviews Neurology 13, 25-36.

Onder, L., Narang, P., Scandella, E., Chai, Q., Iolyeva, M., Hoorweg, K., Halin, C., Richie, E., Kaye, P., Westermann, J., et al. (2012). IL-7-producing stromal cells are critical for lymph node remodeling. Blood 120, 4675-4683.

Ooi, J.D., Petersen, J., Tan, Y.H., Huynh, M., Willett, Z.J., Ramarathinam, S.H., Eggenhuizen, P.J., Loh, K.L., Watson, K.A., Gan, P.Y., et al. (2017). Dominant protection from HLA-linked autoimmunity by antigen-specific regulatory T cells. Nature 545, 243-247.

Osorio, F., Fuentes, C., Lopez, M.N., Salazar-Onfray, F., and Gonzalez, F.E. (2015). Role of Dendritic Cells in the Induction of Lymphocyte Tolerance. Frontiers in immunology 6, 535.

Panda, S.K., Kolbeck, R., and Sanjuan, M.A. (2017). Plasmacytoid dendritic cells in autoimmunity. Current opinion in immunology 44, 20-25.

210

Pardoll, D.M. (2012). The blockade of immune checkpoints in cancer immunotherapy. Nature reviews Cancer 12, 252- 264.

Patel, S., Fu, S., Mastio, J., Dominguez, G.A., Purohit, A., Kossenkov, A., Lin, C., Alicea-Torres, K., Sehgal, M., Nefedova, Y., et al. (2018). Unique pattern of neutrophil migration and function during tumor progression. Nature immunology 19, 1236-1247.

Pedroza-Gonzalez, A., Zhou, G., Vargas-Mendez, E., Boor, P.P., Mancham, S., Verhoef, C., Polak, W.G., Grunhagen, D., Pan, Q., Janssen, H., et al. (2015). Tumor-infiltrating plasmacytoid dendritic cells promote immunosuppression by Tr1 cells in human liver tumors. Oncoimmunology 4, e1008355.

Pender, M.P. (2003). Infection of autoreactive B lymphocytes with EBV, causing chronic autoimmune diseases. Trends in immunology 24, 584-588.

Pender, M.P., and Burrows, S.R. (2014). Epstein-Barr virus and multiple sclerosis: potential opportunities for immunotherapy. Clinical & translational immunology 3, e27.

Perrot, I., Blanchard, D., Freymond, N., Isaac, S., Guibert, B., Pacheco, Y., and Lebecque, S. (2007). Dendritic cells infiltrating human non-small cell lung cancer are blocked at immature stage. Journal of immunology 178, 2763-2769.

Peske, J.D., Thompson, E.D., Gemta, L., Baylis, R.A., Fu, Y.X., and Engelhard, V.H. (2015). Effector lymphocyte- induced lymph node-like vasculature enables naive T-cell entry into tumours and enhanced anti-tumour immunity. Nature communications 6, 7114.

Peters, A., Fowler, K.D., Chalmin, F., Merkler, D., Kuchroo, V.K., and Pot, C. (2015). IL-27 Induces Th17 Differentiation in the Absence of STAT1 Signaling. Journal of immunology 195, 4144-4153.

Petrova, T.V., and Koh, G.Y. (2018). Organ-specific lymphatic vasculature: From development to pathophysiology. The Journal of experimental medicine 215, 35-49.

Petrova, T.V., Makinen, T., Makela, T.P., Saarela, J., Virtanen, I., Ferrell, R.E., Finegold, D.N., Kerjaschki, D., Yla- Herttuala, S., and Alitalo, K. (2002). Lymphatic endothelial reprogramming of vascular endothelial cells by the Prox- 1 homeobox transcription factor. The EMBO journal 21, 4593-4599.

Pflicke, H., and Sixt, M. (2009). Preformed portals facilitate dendritic cell entry into afferent lymphatic vessels. The Journal of experimental medicine 206, 2925-2935.

Pham, T.H., Baluk, P., Xu, Y., Grigorova, I., Bankovich, A.J., Pappu, R., Coughlin, S.R., McDonald, D.M., Schwab, S.R., and Cyster, J.G. (2010). Lymphatic endothelial cell sphingosine kinase activity is required for lymphocyte egress and lymphatic patterning. The Journal of experimental medicine 207, 17-27.

Pham, T.H., Okada, T., Matloubian, M., Lo, C.G., and Cyster, J.G. (2008). S1P1 receptor signaling overrides retention mediated by G alpha i-coupled receptors to promote T cell egress. Immunity 28, 122-133.

Pikor, N.B., Astarita, J.L., Summers-Deluca, L., Galicia, G., Qu, J., Ward, L.A., Armstrong, S., Dominguez, C.X., Malhotra, D., Heiden, B., et al. (2015). Integration of Th17- and Lymphotoxin-Derived Signals Initiates Meningeal- Resident Stromal Cell Remodeling to Propagate Neuroinflammation. Immunity 43, 1160-1173.

Pipi, E., Nayar, S., Gardner, D.H., Colafrancesco, S., Smith, C., and Barone, F. (2018). Tertiary Lymphoid Structures: Autoimmunity Goes Local. Frontiers in immunology 9, 1952.

Pisapia, L., Barba, P., Cortese, A., Cicatiello, V., Morelli, F., and Del Pozzo, G. (2015). EBP1 protein modulates the expression of human MHC class II molecules in non-hematopoietic cancer cells. International journal of oncology 47, 481-489.

Pober, J.S., Merola, J., Liu, R., and Manes, T.D. (2017). Antigen Presentation by Vascular Cells. Frontiers in immunology 8, 1907.

Podgrabinska, S., Braun, P., Velasco, P., Kloos, B., Pepper, M.S., and Skobe, M. (2002). Molecular characterization of lymphatic endothelial cells. Proceedings of the National Academy of Sciences of the United States of America 99, 16069-16074.

211

Podgrabinska, S., Kamalu, O., Mayer, L., Shimaoka, M., Snoeck, H., Randolph, G.J., and Skobe, M. (2009). Inflamed lymphatic endothelium suppresses dendritic cell maturation and function via Mac-1/ICAM-1-dependent mechanism. Journal of immunology 183, 1767-1779.

Polman, C.H., O'Connor, P.W., Havrdova, E., Hutchinson, M., Kappos, L., Miller, D.H., Phillips, J.T., Lublin, F.D., Giovannoni, G., Wajgt, A., et al. (2006). A randomized, placebo-controlled trial of natalizumab for relapsing multiple sclerosis. The New England journal of medicine 354, 899-910.

Pomie, C., Menager-Marcq, I., and van Meerwijk, J.P. (2008). Murine CD8+ regulatory T lymphocytes: the new era. Human immunology 69, 708-714.

Ponath, G., Park, C., and Pitt, D. (2018). The Role of Astrocytes in Multiple Sclerosis. Frontiers in immunology 9, 217.

Prinz, M., Schmidt, H., Mildner, A., Knobeloch, K.P., Hanisch, U.K., Raasch, J., Merkler, D., Detje, C., Gutcher, I., Mages, J., et al. (2008). Distinct and nonredundant in vivo functions of IFNAR on myeloid cells limit autoimmunity in the central nervous system. Immunity 28, 675-686.

Pure, E., and Lo, A. (2016). Can Targeting Stroma Pave the Way to Enhanced Antitumor Immunity and Immunotherapy of Solid Tumors? Cancer immunology research 4, 269-278.

Puttur, F., Francozo, M., Solmaz, G., Bueno, C., Lindenberg, M., Gohmert, M., Swallow, M., Tufa, D., Jacobs, R., Lienenklaus, S., et al. (2016). Conventional Dendritic Cells Confer Protection against Mouse Cytomegalovirus Infection via TLR9 and MyD88 Signaling. Cell reports 17, 1113-1127.

Randolph, G.J., Ivanov, S., Zinselmeyer, B.H., and Scallan, J.P. (2017). The Lymphatic System: Integral Roles in Immunity. Annual review of immunology 35, 31-52.

Rangachari, M., Kerfoot, S.M., Arbour, N., and Alvarez, J.I. (2017). Editorial: Lymphocytes in MS and EAE: More Than Just a CD4(+) World. Frontiers in immunology 8, 133.

Rangachari, M., and Kuchroo, V.K. (2013). Using EAE to better understand principles of immune function and autoimmune pathology. Journal of autoimmunity 45, 31-39.

Ransohoff, R.M., Hafler, D.A., and Lucchinetti, C.F. (2015). Multiple sclerosis-a quiet revolution. Nature reviews Neurology 11, 134-142.

Rasouli, J., Ciric, B., Imitola, J., Gonnella, P., Hwang, D., Mahajan, K., Mari, E.R., Safavi, F., Leist, T.P., Zhang, G.X., et al. (2015). Expression of GM-CSF in T Cells Is Increased in Multiple Sclerosis and Suppressed by IFN-beta Therapy. Journal of immunology 194, 5085-5093.

Rautela, J., Baschuk, N., Slaney, C.Y., Jayatilleke, K.M., Xiao, K., Bidwell, B.N., Lucas, E.C., Hawkins, E.D., Lock, P., Wong, C.S., et al. (2015). Loss of Host Type-I IFN Signaling Accelerates Metastasis and Impairs NK-cell Antitumor Function in Multiple Models of Breast Cancer. Cancer immunology research 3, 1207-1217.

Ray, A., Das, D.S., Song, Y., Macri, V., Richardson, P., Brooks, C.L., Chauhan, D., and Anderson, K.C. (2017). A novel agent SL-401 induces anti-myeloma activity by targeting plasmacytoid dendritic cells, osteoclastogenesis and cancer stem-like cells. Leukemia 31, 2652-2660.

Ray, A., Das, D.S., Song, Y., Richardson, P., Munshi, N.C., Chauhan, D., and Anderson, K.C. (2015). Targeting PD1- PDL1 immune checkpoint in plasmacytoid dendritic cell interactions with T cells, natural killer cells and multiple myeloma cells. Leukemia 29, 1441-1444.

Ray, A., Tian, Z., Das, D.S., Coffman, R.L., Richardson, P., Chauhan, D., and Anderson, K.C. (2014). A novel TLR-9 agonist C792 inhibits plasmacytoid dendritic cell-induced myeloma cell growth and enhance cytotoxicity of bortezomib. Leukemia 28, 1716-1724.

Razakandrainibe, R., Pelleau, S., Grau, G.E., and Jambou, R. (2012). Antigen presentation by endothelial cells: what role in the pathophysiology of malaria? Trends in parasitology 28, 151-160.

Reboldi, A., Coisne, C., Baumjohann, D., Benvenuto, F., Bottinelli, D., Lira, S., Uccelli, A., Lanzavecchia, A., Engelhardt, B., and Sallusto, F. (2009). C-C chemokine receptor 6-regulated entry of TH-17 cells into the CNS through the choroid plexus is required for the initiation of EAE. Nature immunology 10, 514-523.

212

Reith, W., LeibundGut-Landmann, S., and Waldburger, J.M. (2005). Regulation of MHC class II gene expression by the class II transactivator. Nature reviews Immunology 5, 793-806.

Reizis, B. (2019). Plasmacytoid Dendritic Cells: Development, Regulation, and Function. Immunity 50, 37-50.

Reizis, B., Bunin, A., Ghosh, H.S., Lewis, K.L., and Sisirak, V. (2011a). Plasmacytoid dendritic cells: recent progress and open questions. Annual review of immunology 29, 163-183.

Reizis, B., Colonna, M., Trinchieri, G., Barrat, F., and Gilliet, M. (2011b). Plasmacytoid dendritic cells: one-trick ponies or workhorses of the immune system? Nature reviews Immunology 11, 558-565.

Reynoso, E.D., Elpek, K.G., Francisco, L., Bronson, R., Bellemare-Pelletier, A., Sharpe, A.H., Freeman, G.J., and Turley, S.J. (2009). Intestinal tolerance is converted to autoimmune enteritis upon PD-1 ligand blockade. Journal of immunology 182, 2102-2112.

Reynoso, E.D., and Turley, S.J. (2009). Unconventional antigen-presenting cells in the induction of peripheral CD8(+) T cell tolerance. Journal of leukocyte biology 86, 795-801.

Ribas, A., Medina, T., Kummar, S., Amin, A., Kalbasi, A., Drabick, J.J., Barve, M., Daniels, G.A., Wong, D.J., Schmidt, E.V., et al. (2018). SD-101 in Combination with Pembrolizumab in Advanced Melanoma: Results of a Phase Ib, Multicenter Study. Cancer discovery 8, 1250-1257.

Richards, D.M., Kyewski, B., and Feuerer, M. (2016). Re-examining the Nature and Function of Self-Reactive T cells. Trends in immunology 37, 114-125.

Riedel, A., Shorthouse, D., Haas, L., Hall, B.A., and Shields, J. (2016). Tumor-induced stromal reprogramming drives lymph node transformation. Nature immunology 17, 1118-1127.

Roberts, E.W., Broz, M.L., Binnewies, M., Headley, M.B., Nelson, A.E., Wolf, D.M., Kaisho, T., Bogunovic, D., Bhardwaj, N., and Krummel, M.F. (2016). Critical Role for CD103(+)/CD141(+) Dendritic Cells Bearing CCR7 for Tumor Antigen Trafficking and Priming of T Cell Immunity in Melanoma. Cancer cell 30, 324-336.

Roche, P.A., and Furuta, K. (2015). The ins and outs of MHC class II-mediated antigen processing and presentation. Nature reviews Immunology 15, 203-216.

Rodda, L.B., Lu, E., Bennett, M.L., Sokol, C.L., Wang, X., Luther, S.A., Barres, B.A., Luster, A.D., Ye, C.J., and Cyster, J.G. (2018). Single-Cell RNA Sequencing of Lymph Node Stromal Cells Reveals Niche-Associated Heterogeneity. Immunity 48, 1014-1028 e1016.

Rodrigues, P.F., Alberti-Servera, L., Eremin, A., Grajales-Reyes, G.E., Ivanek, R., and Tussiwand, R. (2018). Distinct progenitor lineages contribute to the heterogeneity of plasmacytoid dendritic cells. Nature immunology 19, 711-722.

Rogers, G.L., Shirley, J.L., Zolotukhin, I., Kumar, S.R.P., Sherman, A., Perrin, G.Q., Hoffman, B.E., Srivastava, A., Basner-Tschakarjan, E., Wallet, M.A., et al. (2017). Plasmacytoid and conventional dendritic cells cooperate in crosspriming AAV capsid-specific CD8(+) T cells. Blood 129, 3184-3195.

Roman, C.A., and Arnett, P.A. (2016). Structural brain indices and executive functioning in multiple sclerosis: A review. Journal of clinical and experimental neuropsychology 38, 261-274.

Ronchi, F., Basso, C., Preite, S., Reboldi, A., Baumjohann, D., Perlini, L., Lanzavecchia, A., and Sallusto, F. (2016). Experimental priming of encephalitogenic Th1/Th17 cells requires pertussis toxin-driven IL-1beta production by myeloid cells. Nature communications 7, 11541.

Roozendaal, R., Mempel, T.R., Pitcher, L.A., Gonzalez, S.F., Verschoor, A., Mebius, R.E., von Andrian, U.H., and Carroll, M.C. (2009). Conduits mediate transport of low-molecular-weight antigen to lymph node follicles. Immunity 30, 264- 276.

Rothenfusser, S., Tuma, E., Endres, S., and Hartmann, G. (2002). Plasmacytoid dendritic cells: the key to CpG. Human immunology 63, 1111-1119.

Rothhammer, V., Heink, S., Petermann, F., Srivastava, R., Claussen, M.C., Hemmer, B., and Korn, T. (2011). Th17 lymphocytes traffic to the central nervous system independently of alpha4 integrin expression during EAE. The Journal of experimental medicine 208, 2465-2476.

213

Rouhani, S.J., Eccles, J.D., Riccardi, P., Peske, J.D., Tewalt, E.F., Cohen, J.N., Liblau, R., Makinen, T., and Engelhard, V.H. (2015). Roles of lymphatic endothelial cells expressing peripheral tissue antigens in CD4 T-cell tolerance induction. Nature communications 6, 6771.

Royer, D.J., Elliott, M.H., Le, Y.Z., and Carr, D.J.J. (2018). Corneal Epithelial Cells Exhibit Myeloid Characteristics and Present Antigen via MHC Class II. Investigative ophthalmology & visual science 59, 1512-1522.

Ruben, J.M., Garcia-Romo, G.S., Breman, E., van der Kooij, S., Redeker, A., Arens, R., and van Kooten, C. (2018). Human plasmacytoid dendritic cells acquire phagocytic capacity by TLR9 ligation in the presence of soluble factors produced by renal epithelial cells. Kidney international 93, 355-364.

Ruddle, N.H. (2016). High Endothelial Venules and Lymphatic Vessels in Tertiary Lymphoid Organs: Characteristics, Functions, and Regulation. Frontiers in immunology 7, 491.

Rudolph, A., Chang-Claude, J., and Schmidt, M.K. (2016). Gene-environment interaction and risk of breast cancer. British journal of cancer 114, 125-133.

Russo, E., Nitschke, M., and Halin, C. (2013). Dendritic cell interactions with lymphatic endothelium. Lymphatic research and biology 11, 172-182.

Russo, E., Teijeira, A., Vaahtomeri, K., Willrodt, A.H., Bloch, J.S., Nitschke, M., Santambrogio, L., Kerjaschki, D., Sixt, M., and Halin, C. (2016). Intralymphatic CCL21 Promotes Tissue Egress of Dendritic Cells through Afferent Lymphatic Vessels. Cell reports 14, 1723-1734.

Saaristo, A., Veikkola, T., Tammela, T., Enholm, B., Karkkainen, M.J., Pajusola, K., Bueler, H., Yla-Herttuala, S., and Alitalo, K. (2002). Lymphangiogenic gene therapy with minimal blood vascular side effects. The Journal of experimental medicine 196, 719-730.

Sadaka, C., Marloie-Provost, M.A., Soumelis, V., and Benaroch, P. (2009). Developmental regulation of MHC II expression and transport in human plasmacytoid-derived dendritic cells. Blood 113, 2127-2135.

Sage, A.P., Murphy, D., Maffia, P., Masters, L.M., Sabir, S.R., Baker, L.L., Cambrook, H., Finigan, A.J., Ait-Oufella, H., Grassia, G., et al. (2014). MHC Class II-restricted antigen presentation by plasmacytoid dendritic cells drives proatherogenic T cell immunity. Circulation 130, 1363-1373.

Sagiv-Barfi, I., Czerwinski, D.K., Levy, S., Alam, I.S., Mayer, A.T., Gambhir, S.S., and Levy, R. (2018). Eradication of spontaneous malignancy by local immunotherapy. Science translational medicine 10.

Sakaguchi, S., Sakaguchi, N., Asano, M., Itoh, M., and Toda, M. (1995). Immunologic self-tolerance maintained by activated T cells expressing IL-2 receptor alpha-chains (CD25). Breakdown of a single mechanism of self-tolerance causes various autoimmune diseases. Journal of immunology 155, 1151-1164.

Sakata, K., Nakayamada, S., Miyazaki, Y., Kubo, S., Ishii, A., Nakano, K., and Tanaka, Y. (2018). Up-Regulation of TLR7- Mediated IFN-alpha Production by Plasmacytoid Dendritic Cells in Patients With Systemic Lupus Erythematosus. Frontiers in immunology 9, 1957.

Salio, M., Palmowski, M.J., Atzberger, A., Hermans, I.F., and Cerundolo, V. (2004). CpG-matured murine plasmacytoid dendritic cells are capable of in vivo priming of functional CD8 T cell responses to endogenous but not exogenous antigens. The Journal of experimental medicine 199, 567-579.

Salmon, H., Idoyaga, J., Rahman, A., Leboeuf, M., Remark, R., Jordan, S., Casanova-Acebes, M., Khudoynazarova, M., Agudo, J., Tung, N., et al. (2016). Expansion and Activation of CD103(+) Dendritic Cell Progenitors at the Tumor Site Enhances Tumor Responses to Therapeutic PD-L1 and BRAF Inhibition. Immunity 44, 924-938.

Salou, M., Nicol, B., Garcia, A., and Laplaud, D.A. (2015). Involvement of CD8(+) T Cells in Multiple Sclerosis. Frontiers in immunology 6, 604.

Sapoznikov, A., Fischer, J.A., Zaft, T., Krauthgamer, R., Dzionek, A., and Jung, S. (2007). Organ-dependent in vivo priming of naive CD4+, but not CD8+, T cells by plasmacytoid dendritic cells. The Journal of experimental medicine 204, 1923-1933.

Sato-Kaneko, F., Yao, S., Ahmadi, A., Zhang, S.S., Hosoya, T., Kaneda, M.M., Varner, J.A., Pu, M., Messer, K.S., Guiducci, C., et al. (2017). Combination immunotherapy with TLR agonists and checkpoint inhibitors suppresses head and neck cancer. JCI insight 2.

214

Schaedler, E., Remy-Ziller, C., Hortelano, J., Kehrer, N., Claudepierre, M.C., Gatard, T., Jakobs, C., Preville, X., Carpentier, A.F., and Rittner, K. (2017). Sequential administration of a MVA-based MUC1 cancer vaccine and the TLR9 ligand Litenimod (Li28) improves local immune defense against tumors. Vaccine 35, 577-585.

Schettini, J., Kidiyoor, A., Besmer, D.M., Tinder, T.L., Roy, L.D., Lustgarten, J., Gendler, S.J., and Mukherjee, P. (2012). Intratumoral delivery of CpG-conjugated anti-MUC1 antibody enhances NK cell anti-tumor activity. Cancer immunology, immunotherapy : CII 61, 2055-2065.

Schiavoni, G., Mattei, F., and Gabriele, L. (2013). Type I Interferons as Stimulators of DC-Mediated Cross-Priming: Impact on Anti-Tumor Response. Frontiers in immunology 4, 483.

Schineis, P., Runge, P., and Halin, C. (2018). Cellular traffic through afferent lymphatic vessels. Vascular pharmacology.

Schlager, C., Korner, H., Krueger, M., Vidoli, S., Haberl, M., Mielke, D., Brylla, E., Issekutz, T., Cabanas, C., Nelson, P.J., et al. (2016). Effector T-cell trafficking between the leptomeninges and the cerebrospinal fluid. Nature 530, 349-353.

Schluns, K.S., Kieper, W.C., Jameson, S.C., and Lefrancois, L. (2000). Interleukin-7 mediates the homeostasis of naive and memory CD8 T cells in vivo. Nature immunology 1, 426-432.

Schluns, K.S., and Lefrancois, L. (2003). Cytokine control of memory T-cell development and survival. Nature reviews Immunology 3, 269-279.

Schneberger, D., Caldwell, S., Kanthan, R., and Singh, B. (2013). Expression of Toll-like receptor 9 in mouse and human lungs. Journal of anatomy 222, 495-503.

Schuijs, M.J., Hammad, H., and Lambrecht, B.N. (2019). Professional and 'Amateur' Antigen-Presenting Cells In Type 2 Immunity. Trends in immunology 40, 22-34.

Schuster, S., Hurrell, B., and Tacchini-Cottier, F. (2013). Crosstalk between neutrophils and dendritic cells: a context- dependent process. Journal of leukocyte biology 94, 671-675.

Schutz, C., Inselmann, S., Sausslele, S., Dietz, C.T., Mu Ller, M.C., Eigendorff, E., Brendel, C.A., Metzelder, S.K., Bru Mmendorf, T.H., Waller, C., et al. (2017). Expression of the CTLA-4 ligand CD86 on plasmacytoid dendritic cells (pDC) predicts risk of disease recurrence after treatment discontinuation in CML. Leukemia 31, 829-836.

Schwab, N., Schneider-Hohendorf, T., and Wiendl, H. (2015). Therapeutic uses of anti-alpha4-integrin (anti-VLA-4) antibodies in multiple sclerosis. International immunology 27, 47-53.

Scott, C.L., Soen, B., Martens, L., Skrypek, N., Saelens, W., Taminau, J., Blancke, G., Van Isterdael, G., Huylebroeck, D., Haigh, J., et al. (2016). The transcription factor Zeb2 regulates development of conventional and plasmacytoid DCs by repressing Id2. The Journal of experimental medicine 213, 897-911.

See, P., Dutertre, C.A., Chen, J., Gunther, P., McGovern, N., Irac, S.E., Gunawan, M., Beyer, M., Handler, K., Duan, K., et al. (2017). Mapping the human DC lineage through the integration of high-dimensional techniques. Science 356.

Segura, E., Durand, M., and Amigorena, S. (2013a). Similar antigen cross-presentation capacity and phagocytic functions in all freshly isolated human lymphoid organ-resident dendritic cells. The Journal of experimental medicine 210, 1035-1047.

Segura, E., Touzot, M., Bohineust, A., Cappuccio, A., Chiocchia, G., Hosmalin, A., Dalod, M., Soumelis, V., and Amigorena, S. (2013b). Human inflammatory dendritic cells induce Th17 cell differentiation. Immunity 38, 336-348.

Sharma, M.D., Baban, B., Chandler, P., Hou, D.Y., Singh, N., Yagita, H., Azuma, M., Blazar, B.R., Mellor, A.L., and Munn, D.H. (2007). Plasmacytoid dendritic cells from mouse tumor-draining lymph nodes directly activate mature Tregs via indoleamine 2,3-dioxygenase. The Journal of clinical investigation 117, 2570-2582.

Sharma, S., Dominguez, A.L., Manrique, S.Z., Cavallo, F., Sakaguchi, S., and Lustgarten, J. (2008). Systemic targeting of CpG-ODN to the tumor microenvironment with anti-neu-CpG hybrid molecule and T regulatory cell depletion induces memory responses in BALB-neuT tolerant mice. Cancer research 68, 7530-7540.

Sharma, S., Karakousis, C.P., Takita, H., Shin, K., and Brooks, S.P. (2004). Cytokines and chemokines are expressed at different levels in small and large murine colon-26 tumors following intratumoral injections of CpG ODN. Neoplasia 6, 523-528.

215

Sharpe, A.H., and Pauken, K.E. (2018). The diverse functions of the PD1 inhibitory pathway. Nature reviews Immunology 18, 153-167.

Shaul, M.E., and Fridlender, Z.G. (2018). Cancer related circulating and tumor-associated neutrophils - subtypes, sources and function. The FEBS journal.

Shaygannejad, V., Golabchi, K., Haghighi, S., Dehghan, H., and Moshayedi, A. (2010). A Comparative Study of 25 (OH) Vitamin D Serum Levels in Patients with Multiple Sclerosis and Control Group in Isfahan, Iran. International journal of preventive medicine 1, 195-201.

Shetty, S., Lalor, P.F., and Adams, D.H. (2018). Liver sinusoidal endothelial cells - gatekeepers of hepatic immunity. Nature reviews Gastroenterology & hepatology 15, 555-567.

Shi, W., Li, X., Porter, J.L., Ostrodi, D.H., Yang, B., Li, J., Wang, Y., Zhang, J., Bai, L., and Jiao, S. (2014). Level of plasmacytoid dendritic cells is increased in non-small cell lung carcinoma. Tumour biology : the journal of the International Society for Oncodevelopmental Biology and Medicine 35, 2247-2252.

Shields, J.D., Kourtis, I.C., Tomei, A.A., Roberts, J.M., and Swartz, M.A. (2010). Induction of lymphoidlike stroma and immune escape by tumors that express the chemokine CCL21. Science 328, 749-752.

Shirani, A., and Stuve, O. (2017). Natalizumab for Multiple Sclerosis: A Case in Point for the Impact of Translational Neuroimmunology. Journal of immunology 198, 1381-1386.

Sichien, D., Scott, C.L., Martens, L., Vanderkerken, M., Van Gassen, S., Plantinga, M., Joeris, T., De Prijck, S., Vanhoutte, L., Vanheerswynghels, M., et al. (2016). IRF8 Transcription Factor Controls Survival and Function of Terminally Differentiated Conventional and Plasmacytoid Dendritic Cells, Respectively. Immunity 45, 626-640.

Siegal, F.P., Kadowaki, N., Shodell, M., Fitzgerald-Bocarsly, P.A., Shah, K., Ho, S., Antonenko, S., and Liu, Y.J. (1999). The nature of the principal type 1 interferon-producing cells in human blood. Science 284, 1835-1837.

Siegert, S., and Luther, S.A. (2012). Positive and negative regulation of T cell responses by fibroblastic reticular cells within paracortical regions of lymph nodes. Frontiers in immunology 3, 285.

Sinha, S., Boyden, A.W., Itani, F.R., Crawford, M.P., and Karandikar, N.J. (2015). CD8(+) T-Cells as Immune Regulators of Multiple Sclerosis. Frontiers in immunology 6, 619.

Sisirak, V., Faget, J., Gobert, M., Goutagny, N., Vey, N., Treilleux, I., Renaudineau, S., Poyet, G., Labidi-Galy, S.I., Goddard-Leon, S., et al. (2012). Impaired IFN-alpha production by plasmacytoid dendritic cells favors regulatory T- cell expansion that may contribute to breast cancer progression. Cancer research 72, 5188-5197.

Sisirak, V., Vey, N., Goutagny, N., Renaudineau, S., Malfroy, M., Thys, S., Treilleux, I., Labidi-Galy, S.I., Bachelot, T., Dezutter-Dambuyant, C., et al. (2013). Breast cancer-derived transforming growth factor-beta and tumor necrosis factor-alpha compromise interferon-alpha production by tumor-associated plasmacytoid dendritic cells. International journal of cancer 133, 771-778.

Sixt, M., Kanazawa, N., Selg, M., Samson, T., Roos, G., Reinhardt, D.P., Pabst, R., Lutz, M.B., and Sorokin, L. (2005). The conduit system transports soluble antigens from the afferent lymph to resident dendritic cells in the T cell area of the lymph node. Immunity 22, 19-29.

Smith, T.T., Moffett, H.F., Stephan, S.B., Opel, C.F., Dumigan, A.G., Jiang, X., Pillarisetty, V.G., Pillai, S.P.S., Wittrup, K.D., and Stephan, M.T. (2017). Biopolymers codelivering engineered T cells and STING agonists can eliminate heterogeneous tumors. The Journal of clinical investigation 127, 2176-2191.

Smolders, J., Peelen, E., Thewissen, M., Menheere, P., Tervaert, J.W., Hupperts, R., and Damoiseaux, J. (2009). The relevance of vitamin D receptor gene polymorphisms for vitamin D research in multiple sclerosis. Autoimmunity reviews 8, 621-626.

Snell, L.M., McGaha, T.L., and Brooks, D.G. (2017). Type I Interferon in Chronic Virus Infection and Cancer. Trends in immunology 38, 542-557.

Sorensen, P.S., and Blinkenberg, M. (2016). The potential role for ocrelizumab in the treatment of multiple sclerosis: current evidence and future prospects. Therapeutic advances in neurological disorders 9, 44-52.

216

Sosa-Pineda, B., Wigle, J.T., and Oliver, G. (2000). Hepatocyte migration during liver development requires Prox1. Nature genetics 25, 254-255.

Soumelis, V., and Liu, Y.J. (2006). From plasmacytoid to dendritic cell: morphological and functional switches during plasmacytoid pre-dendritic cell differentiation. European journal of immunology 36, 2286-2292.

Spath, S., Komuczki, J., Hermann, M., Pelczar, P., Mair, F., Schreiner, B., and Becher, B. (2017). Dysregulation of the Cytokine GM-CSF Induces Spontaneous Phagocyte Invasion and Immunopathology in the Central Nervous System. Immunity 46, 245-260.

Sprent, J. (1995). Antigen-presenting cells. Professionals and amateurs. Current biology : CB 5, 1095-1097.

Stacker, S.A., Williams, S.P., Karnezis, T., Shayan, R., Fox, S.B., and Achen, M.G. (2014). Lymphangiogenesis and lymphatic vessel remodelling in cancer. Nature reviews Cancer 14, 159-172.

Stary, G., Bangert, C., Tauber, M., Strohal, R., Kopp, T., and Stingl, G. (2007). Tumoricidal activity of TLR7/8-activated inflammatory dendritic cells. The Journal of experimental medicine 204, 1441-1451.

Stefanova, I., Dorfman, J.R., and Germain, R.N. (2002). Self-recognition promotes the foreign antigen sensitivity of naive T lymphocytes. Nature 420, 429-434.

Steinman, R.M., and Cohn, Z.A. (1973). Identification of a novel cell type in peripheral lymphoid organs of mice. I. Morphology, quantitation, tissue distribution. The Journal of experimental medicine 137, 1142-1162.

Steinman, R.M., Hawiger, D., and Nussenzweig, M.C. (2003). Tolerogenic dendritic cells. Annual review of immunology 21, 685-711.

Steinman, R.M., Turley, S., Mellman, I., and Inaba, K. (2000). The induction of tolerance by dendritic cells that have captured apoptotic cells. The Journal of experimental medicine 191, 411-416.

Stergiopoulos, A., Elkouris, M., and Politis, P.K. (2014). Prospero-related homeobox 1 (Prox1) at the crossroads of diverse pathways during adult neural fate specification. Frontiers in cellular neuroscience 8, 454.

Stockinger, B., and Omenetti, S. (2017). The dichotomous nature of T helper 17 cells. Nature reviews Immunology 17, 535-544.

Stromnes, I.M., Cerretti, L.M., Liggitt, D., Harris, R.A., and Goverman, J.M. (2008). Differential regulation of central nervous system autoimmunity by T(H)1 and T(H)17 cells. Nature medicine 14, 337-342.

Swartz, M.A. (2014). Immunomodulatory roles of lymphatic vessels in cancer progression. Cancer immunology research 2, 701-707.

Swartz, M.A., and Lund, A.W. (2012). Lymphatic and interstitial flow in the tumour microenvironment: linking mechanobiology with immunity. Nature reviews Cancer 12, 210-219.

Swiecki, M., and Colonna, M. (2015). The multifaceted biology of plasmacytoid dendritic cells. Nature reviews Immunology 15, 471-485.

Swiecki, M., Gilfillan, S., Vermi, W., Wang, Y., and Colonna, M. (2010). Plasmacytoid dendritic cell ablation impacts early interferon responses and antiviral NK and CD8(+) T cell accrual. Immunity 33, 955-966.

Szot, C., Saha, S., Zhang, X.M., Zhu, Z., Hilton, M.B., Morris, K., Seaman, S., Dunleavey, J.M., Hsu, K.S., Yu, G.J., et al. (2018). Tumor stroma-targeted antibody-drug conjugate triggers localized anticancer drug release. The Journal of clinical investigation 128, 2927-2943.

Taflin, C., Favier, B., Baudhuin, J., Savenay, A., Hemon, P., Bensussan, A., Charron, D., Glotz, D., and Mooney, N. (2011). Human endothelial cells generate Th17 and regulatory T cells under inflammatory conditions. Proceedings of the National Academy of Sciences of the United States of America 108, 2891-2896.

Takaba, H., Morishita, Y., Tomofuji, Y., Danks, L., Nitta, T., Komatsu, N., Kodama, T., and Takayanagi, H. (2015). Fezf2 Orchestrates a Thymic Program of Self-Antigen Expression for Immune Tolerance. Cell 163, 975-987.

Takaba, H., and Takayanagi, H. (2017). The Mechanisms of T Cell Selection in the Thymus. Trends in immunology 38, 805-816.

217

Takagi, H., Fukaya, T., Eizumi, K., Sato, Y., Sato, K., Shibazaki, A., Otsuka, H., Hijikata, A., Watanabe, T., Ohara, O., et al. (2011). Plasmacytoid dendritic cells are crucial for the initiation of inflammation and T cell immunity in vivo. Immunity 35, 958-971.

Takeuchi, O., and Akira, S. (2010). Pattern recognition receptors and inflammation. Cell 140, 805-820.

Tamburini, B.A., Burchill, M.A., and Kedl, R.M. (2014). Antigen capture and archiving by lymphatic endothelial cells following vaccination or viral infection. Nature communications 5, 3989.

Tammela, T., Saaristo, A., Holopainen, T., Lyytikka, J., Kotronen, A., Pitkonen, M., Abo-Ramadan, U., Yla-Herttuala, S., Petrova, T.V., and Alitalo, K. (2007). Therapeutic differentiation and maturation of lymphatic vessels after lymph node dissection and transplantation. Nature medicine 13, 1458-1466.

Targoni, O.S., and Lehmann, P.V. (1998). Endogenous myelin basic protein inactivates the high avidity T cell repertoire. The Journal of experimental medicine 187, 2055-2063.

Teijeira, A., Russo, E., and Halin, C. (2014). Taking the lymphatic route: dendritic cell migration to draining lymph nodes. Seminars in immunopathology 36, 261-274.

Tel, J., Aarntzen, E.H., Baba, T., Schreibelt, G., Schulte, B.M., Benitez-Ribas, D., Boerman, O.C., Croockewit, S., Oyen, W.J., van Rossum, M., et al. (2013a). Natural human plasmacytoid dendritic cells induce antigen-specific T-cell responses in melanoma patients. Cancer research 73, 1063-1075.

Tel, J., Anguille, S., Waterborg, C.E., Smits, E.L., Figdor, C.G., and de Vries, I.J. (2014). Tumoricidal activity of human dendritic cells. Trends in immunology 35, 38-46.

Tel, J., Benitez-Ribas, D., Hoosemans, S., Cambi, A., Adema, G.J., Figdor, C.G., Tacken, P.J., and de Vries, I.J. (2011). DEC-205 mediates antigen uptake and presentation by both resting and activated human plasmacytoid dendritic cells. European journal of immunology 41, 1014-1023.

Tel, J., Lambeck, A.J., Cruz, L.J., Tacken, P.J., de Vries, I.J., and Figdor, C.G. (2010). Human plasmacytoid dendritic cells phagocytose, process, and present exogenous particulate antigen. Journal of immunology 184, 4276-4283.

Tel, J., Sittig, S.P., Blom, R.A., Cruz, L.J., Schreibelt, G., Figdor, C.G., and de Vries, I.J. (2013b). Targeting uptake receptors on human plasmacytoid dendritic cells triggers antigen cross-presentation and robust type I IFN secretion. Journal of immunology 191, 5005-5012.

Tel, J., Smits, E.L., Anguille, S., Joshi, R.N., Figdor, C.G., and de Vries, I.J. (2012a). Human plasmacytoid dendritic cells are equipped with antigen-presenting and tumoricidal capacities. Blood 120, 3936-3944.

Tel, J., van der Leun, A.M., Figdor, C.G., Torensma, R., and de Vries, I.J. (2012b). Harnessing human plasmacytoid dendritic cells as professional APCs. Cancer immunology, immunotherapy : CII 61, 1279-1288.

Temizoz, B., Kuroda, E., and Ishii, K.J. (2016). Vaccine adjuvants as potential cancer immunotherapeutics. International immunology 28, 329-338.

Terra, M., Oberkampf, M., Fayolle, C., Rosenbaum, P., Guillerey, C., Dadaglio, G., and Leclerc, C. (2018). Tumor- Derived TGFbeta Alters the Ability of Plasmacytoid Dendritic Cells to Respond to Innate Immune Signaling. Cancer research 78, 3014-3026.

Tewalt, E.F., Cohen, J.N., Rouhani, S.J., Guidi, C.J., Qiao, H., Fahl, S.P., Conaway, M.R., Bender, T.P., Tung, K.S., Vella, A.T., et al. (2012). Lymphatic endothelial cells induce tolerance via PD-L1 and lack of costimulation leading to high- level PD-1 expression on CD8 T cells. Blood 120, 4772-4782.

Thaiss, C.A., Levy, M., Itav, S., and Elinav, E. (2016). Integration of Innate Immune Signaling. Trends in immunology 37, 84-101.

The, L. (2018). GLOBOCAN 2018: counting the toll of cancer. Lancet 392, 985.

Theofilopoulos, A.N., Kono, D.H., and Baccala, R. (2017). The multiple pathways to autoimmunity. Nature immunology 18, 716-724.

218

Thomas, S.N., Rutkowski, J.M., Pasquier, M., Kuan, E.L., Alitalo, K., Randolph, G.J., and Swartz, M.A. (2012). Impaired humoral immunity and tolerance in K14-VEGFR-3-Ig mice that lack dermal lymphatic drainage. Journal of immunology 189, 2181-2190.

Thompson, E.D., Enriquez, H.L., Fu, Y.X., and Engelhard, V.H. (2010). Tumor masses support naive T cell infiltration, activation, and differentiation into effectors. The Journal of experimental medicine 207, 1791-1804.

Treffers, L.W., Hiemstra, I.H., Kuijpers, T.W., van den Berg, T.K., and Matlung, H.L. (2016). Neutrophils in cancer. Immunological reviews 273, 312-328.

Treilleux, I., Blay, J.Y., Bendriss-Vermare, N., Ray-Coquard, I., Bachelot, T., Guastalla, J.P., Bremond, A., Goddard, S., Pin, J.J., Barthelemy-Dubois, C., et al. (2004). Dendritic cell infiltration and prognosis of early stage breast cancer. Clinical cancer research : an official journal of the American Association for Cancer Research 10, 7466-7474.

Trifilo, M.J., and Lane, T.E. (2004). The CC chemokine ligand 3 regulates CD11c+CD11b+CD8alpha- dendritic cell maturation and activation following viral infection of the central nervous system: implications for a role in T cell activation. Virology 327, 8-15.

Turley, S.J., Cremasco, V., and Astarita, J.L. (2015). Immunological hallmarks of stromal cells in the tumour microenvironment. Nature reviews Immunology 15, 669-682.

Turley, S.J., Fletcher, A.L., and Elpek, K.G. (2010). The stromal and haematopoietic antigen-presenting cells that reside in secondary lymphoid organs. Nature reviews Immunology 10, 813-825.

Ulvmar, M.H., and Makinen, T. (2016). Heterogeneity in the lymphatic vascular system and its origin. Cardiovascular research 111, 310-321.

Urosevic, M., Dummer, R., Conrad, C., Beyeler, M., Laine, E., Burg, G., and Gilliet, M. (2005). Disease-independent skin recruitment and activation of plasmacytoid predendritic cells following imiquimod treatment. Journal of the National Cancer Institute 97, 1143-1153.

Uto, T., Takagi, H., Fukaya, T., Nasu, J., Fukui, T., Miyanaga, N., Arimura, K., Nakamura, T., Choijookhuu, N., Hishikawa, Y., et al. (2018). Critical role of plasmacytoid dendritic cells in induction of oral tolerance. The Journal of allergy and clinical immunology 141, 2156-2167 e2159.

Valitutti, S., and Espinosa, E. (2010). Cognate interactions between mast cells and helper T lymphocytes. Self/nonself 1, 114-122. van Gisbergen, K.P., Sanchez-Hernandez, M., Geijtenbeek, T.B., and van Kooyk, Y. (2005). Neutrophils mediate immune modulation of dendritic cells through glycosylation-dependent interactions between Mac-1 and DC-SIGN. The Journal of experimental medicine 201, 1281-1292.

Veerappan Ganesan, A.P., and Eisenlohr, L.C. (2017). The elucidation of non-classical MHC class II antigen processing through the study of viral antigens. Current opinion in virology 22, 71-76.

Vieira, J.M., Norman, S., Villa Del Campo, C., Cahill, T.J., Barnette, D.N., Gunadasa-Rohling, M., Johnson, L.A., Greaves, D.R., Carr, C.A., Jackson, D.G., et al. (2018). The cardiac lymphatic system stimulates resolution of inflammation following myocardial infarction. The Journal of clinical investigation 128, 3402-3412.

Vigl, B., Aebischer, D., Nitschke, M., Iolyeva, M., Rothlin, T., Antsiferova, O., and Halin, C. (2011). Tissue inflammation modulates gene expression of lymphatic endothelial cells and dendritic cell migration in a stimulus-dependent manner. Blood 118, 205-215.

Viglietta, V., Baecher-Allan, C., Weiner, H.L., and Hafler, D.A. (2004). Loss of functional suppression by CD4+CD25+ regulatory T cells in patients with multiple sclerosis. The Journal of experimental medicine 199, 971-979.

Vignali, D. (2008). How many mechanisms do regulatory T cells need? European journal of immunology 38, 908-911.

Villadangos, J.A., and Young, L. (2008). Antigen-presentation properties of plasmacytoid dendritic cells. Immunity 29, 352-361.

Villani, A.C., Satija, R., Reynolds, G., Sarkizova, S., Shekhar, K., Fletcher, J., Griesbeck, M., Butler, A., Zheng, S., Lazo, S., et al. (2017). Single-cell RNA-seq reveals new types of human blood dendritic cells, monocytes, and progenitors. Science 356.

219

Vollmer, J., and Krieg, A.M. (2009). Immunotherapeutic applications of CpG oligodeoxynucleotide TLR9 agonists. Advanced drug delivery reviews 61, 195-204. von Oppen, N., Schurich, A., Hegenbarth, S., Stabenow, D., Tolba, R., Weiskirchen, R., Geerts, A., Kolanus, W., Knolle, P., and Diehl, L. (2009). Systemic antigen cross-presented by liver sinusoidal endothelial cells induces liver-specific CD8 T-cell retention and tolerization. Hepatology 49, 1664-1672.

Vono, M., Lin, A., Norrby-Teglund, A., Koup, R.A., Liang, F., and Lore, K. (2017). Neutrophils acquire the capacity for antigen presentation to memory CD4(+) T cells in vitro and ex vivo. Blood 129, 1991-2001.

Vuddamalay, Y., Attia, M., Vicente, R., Pomie, C., Enault, G., Leobon, B., Joffre, O., Romagnoli, P., and van Meerwijk, J.P. (2016). Mouse and human CD8(+) CD28(low) regulatory T lymphocytes differentiate in the thymus. Immunology 148, 187-196.

Wack, A., Terczynska-Dyla, E., and Hartmann, R. (2015). Guarding the frontiers: the biology of type III interferons. Nature immunology 16, 802-809.

Wakim, L.M., and Bevan, M.J. (2011). Cross-dressed dendritic cells drive memory CD8+ T-cell activation after viral infection. Nature 471, 629-632.

Walker, L.S., and Abbas, A.K. (2002). The enemy within: keeping self-reactive T cells at bay in the periphery. Nature reviews Immunology 2, 11-19.

Wang, B.X., Rahbar, R., and Fish, E.N. (2011). Interferon: current status and future prospects in cancer therapy. Journal of interferon & cytokine research : the official journal of the International Society for Interferon and Cytokine Research 31, 545-552.

Wang, C., Singer, M., and Anderson, A.C. (2017). Molecular Dissection of CD8(+) T-Cell Dysfunction. Trends in immunology 38, 567-576.

Wang, D., Jiang, W., Zhu, F., Mao, X., and Agrawal, S. (2018a). Modulation of the tumor microenvironment by intratumoral administration of IMO-2125, a novel TLR9 agonist, for cancer immunotherapy. International journal of oncology 53, 1193-1203.

Wang, S., Campos, J., Gallotta, M., Gong, M., Crain, C., Naik, E., Coffman, R.L., and Guiducci, C. (2016). Intratumoral injection of a CpG oligonucleotide reverts resistance to PD-1 blockade by expanding multifunctional CD8+ T cells. Proceedings of the National Academy of Sciences of the United States of America 113, E7240-E7249.

Wang, Y., Su, L., Morin, M.D., Jones, B.T., Mifune, Y., Shi, H., Wang, K.W., Zhan, X., Liu, A., Wang, J., et al. (2018b). Adjuvant effect of the novel TLR1/TLR2 agonist Diprovocim synergizes with anti-PD-L1 to eliminate melanoma in mice. Proceedings of the National Academy of Sciences of the United States of America 115, E8698-E8706.

Wei, S., Kryczek, I., Zou, L., Daniel, B., Cheng, P., Mottram, P., Curiel, T., Lange, A., and Zou, W. (2005). Plasmacytoid dendritic cells induce CD8+ regulatory T cells in human ovarian carcinoma. Cancer research 65, 5020-5026.

Weiden, J., Tel, J., and Figdor, C.G. (2018). Synthetic immune niches for cancer immunotherapy. Nature reviews Immunology 18, 212-219.

Wekerle, H. (2015). Nature plus nurture: the triggering of multiple sclerosis. Swiss medical weekly 145, w14189.

Wekerle, H., Flugel, A., Fugger, L., Schett, G., and Serreze, D. (2012). Autoimmunity's next top models. Nature medicine 18, 66-70.

Wen, Y.R., Yang, J.H., Wang, X., and Yao, Z.B. (2018). Induced dural lymphangiogenesis facilities soluble amyloid-beta clearance from brain in a transgenic mouse model of Alzheimer's disease. Neural regeneration research 13, 709-716.

Whitacre, C.C. (2001). Sex differences in autoimmune disease. Nature immunology 2, 777-780.

Wigle, J.T., Harvey, N., Detmar, M., Lagutina, I., Grosveld, G., Gunn, M.D., Jackson, D.G., and Oliver, G. (2002). An essential role for Prox1 in the induction of the lymphatic endothelial cell phenotype. The EMBO journal 21, 1505- 1513.

Wigle, J.T., and Oliver, G. (1999). Prox1 function is required for the development of the murine lymphatic system. Cell 98, 769-778.

220

Wilson, N.S., El-Sukkari, D., Belz, G.T., Smith, C.M., Steptoe, R.J., Heath, W.R., Shortman, K., and Villadangos, J.A. (2003). Most lymphoid organ dendritic cell types are phenotypically and functionally immature. Blood 102, 2187- 2194.

Wisse, E. (1970). An electron microscopic study of the fenestrated endothelial lining of rat liver sinusoids. Journal of ultrastructure research 31, 125-150.

Wittlich, M., Dudek, M., Bottcher, J.P., Schanz, O., Hegenbarth, S., Bopp, T., Schmitt, E., Kurts, C., Garbers, C., Rose John, S., et al. (2017). Liver sinusoidal endothelial cell cross-priming is supported by CD4 T cell-derived IL-2. Journal of hepatology 66, 978-986.

Wohlleber, D., and Knolle, P.A. (2016). The role of liver sinusoidal cells in local hepatic immune surveillance. Clinical & translational immunology 5, e117.

Wosen, J.E., Mukhopadhyay, D., Macaubas, C., and Mellins, E.D. (2018). Epithelial MHC Class II Expression and Its Role in Antigen Presentation in the Gastrointestinal and Respiratory Tracts. Frontiers in immunology 9, 2144.

Wucherpfennig, K.W., and Strominger, J.L. (1995). Molecular mimicry in T cell-mediated autoimmunity: viral peptides activate human T cell clones specific for myelin basic protein. Cell 80, 695-705.

Wynn, D., Kaufman, M., Montalban, X., Vollmer, T., Simon, J., Elkins, J., O'Neill, G., Neyer, L., Sheridan, J., Wang, C., et al. (2010). Daclizumab in active relapsing multiple sclerosis (CHOICE study): a phase 2, randomised, double-blind, placebo-controlled, add-on trial with interferon beta. The Lancet Neurology 9, 381-390.

Xing, Y., and Hogquist, K.A. (2012). T-cell tolerance: central and peripheral. Cold Spring Harbor perspectives in biology 4.

Xu, R.H., Wong, E.B., Rubio, D., Roscoe, F., Ma, X., Nair, S., Remakus, S., Schwendener, R., John, S., Shlomchik, M., et al. (2015). Sequential Activation of Two Pathogen-Sensing Pathways Required for Type I Interferon Expression and Resistance to an Acute DNA Virus Infection. Immunity 43, 1148-1159.

Yang, B.H., Hagemann, S., Mamareli, P., Lauer, U., Hoffmann, U., Beckstette, M., Fohse, L., Prinz, I., Pezoldt, J., Suerbaum, S., et al. (2016). Foxp3(+) T cells expressing RORgammat represent a stable regulatory T-cell effector lineage with enhanced suppressive capacity during intestinal inflammation. Mucosal immunology 9, 444-457.

Yang, F., Feng, C., Zhang, X., Lu, J., and Zhao, Y. (2017). The Diverse Biological Functions of Neutrophils, Beyond the Defense Against Infections. Inflammation 40, 311-323.

Yang, Y. (2015). Cancer immunotherapy: harnessing the immune system to battle cancer. The Journal of clinical investigation 125, 3335-3337.

Yaniv, K., Isogai, S., Castranova, D., Dye, L., Hitomi, J., and Weinstein, B.M. (2006). Live imaging of lymphatic development in the zebrafish. Nature medicine 12, 711-716.

Yanofsky, V.R., Mitsui, H., Felsen, D., and Carucci, J.A. (2013). Understanding dendritic cells and their role in cutaneous carcinoma and cancer immunotherapy. Clinical & developmental immunology 2013, 624123.

Yeo, K.P., and Angeli, V. (2017). Bidirectional Crosstalk between Lymphatic Endothelial Cell and T Cell and Its Implications in Tumor Immunity. Frontiers in immunology 8, 83.

Yip, L., Creusot, R.J., Pager, C.T., Sarnow, P., and Fathman, C.G. (2013). Reduced DEAF1 function during type 1 diabetes inhibits translation in lymph node stromal cells by suppressing Eif4g3. Journal of molecular cell biology 5, 99-110.

Yip, L., and Fathman, C.G. (2014). Type 1 diabetes in mice and men: gene expression profiling to investigate disease pathogenesis. Immunologic research 58, 340-350.

Yip, L., Su, L., Sheng, D., Chang, P., Atkinson, M., Czesak, M., Albert, P.R., Collier, A.R., Turley, S.J., Fathman, C.G., et al. (2009). Deaf1 isoforms control the expression of genes encoding peripheral tissue antigens in the pancreatic lymph nodes during type 1 diabetes. Nature immunology 10, 1026-1033.

Yogev, N., Frommer, F., Lukas, D., Kautz-Neu, K., Karram, K., Ielo, D., von Stebut, E., Probst, H.C., van den Broek, M., Riethmacher, D., et al. (2012). Dendritic cells ameliorate autoimmunity in the CNS by controlling the homeostasis of PD-1 receptor(+) regulatory T cells. Immunity 37, 264-275.

221

Young, L.J., Wilson, N.S., Schnorrer, P., Proietto, A., ten Broeke, T., Matsuki, Y., Mount, A.M., Belz, G.T., O'Keeffe, M., Ohmura-Hoshino, M., et al. (2008). Differential MHC class II synthesis and ubiquitination confers distinct antigen- presenting properties on conventional and plasmacytoid dendritic cells. Nature immunology 9, 1244-1252.

Yu, C.F., Peng, W.M., Oldenburg, J., Hoch, J., Bieber, T., Limmer, A., Hartmann, G., Barchet, W., Eis-Hubinger, A.M., and Novak, N. (2010). Human plasmacytoid dendritic cells support Th17 cell effector function in response to TLR7 ligation. Journal of immunology 184, 1159-1167.

Yu, X., Cai, B., Wang, M., Tan, P., Ding, X., Wu, J., Li, J., Li, Q., Liu, P., Xing, C., et al. (2016). Cross-Regulation of Two Type I Interferon Signaling Pathways in Plasmacytoid Dendritic Cells Controls Anti-malaria Immunity and Host Mortality. Immunity 45, 1093-1107.

Yuseff, M.I., and Lennon-Dumenil, A.M. (2015). B Cells use Conserved Polarity Cues to Regulate Their Antigen Processing and Presentation Functions. Frontiers in immunology 6, 251.

Zhang, H., Gregorio, J.D., Iwahori, T., Zhang, X., Choi, O., Tolentino, L.L., Prestwood, T., Carmi, Y., and Engleman, E.G. (2017). A distinct subset of plasmacytoid dendritic cells induces activation and differentiation of B and T lymphocytes. Proceedings of the National Academy of Sciences of the United States of America 114, 1988-1993.

Zhang, J., Raper, A., Sugita, N., Hingorani, R., Salio, M., Palmowski, M.J., Cerundolo, V., and Crocker, P.R. (2006). Characterization of Siglec-H as a novel endocytic receptor expressed on murine plasmacytoid dendritic cell precursors. Blood 107, 3600-3608.

Zhang, Q.J., Li, X.L., Wang, D., Huang, X.C., Mathis, J.M., Duan, W.M., Knight, D., Shi, R., Glass, J., Zhang, D.Q., et al. (2008). Trogocytosis of MHC-I/peptide complexes derived from tumors and infected cells enhances dendritic cell cross-priming and promotes adaptive T cell responses. PloS one 3, e3097.

Zhang, X., Koldzic, D.N., Izikson, L., Reddy, J., Nazareno, R.F., Sakaguchi, S., Kuchroo, V.K., and Weiner, H.L. (2004). IL-10 is involved in the suppression of experimental autoimmune encephalomyelitis by CD25+CD4+ regulatory T cells. International immunology 16, 249-256.

Ziani, L., Chouaib, S., and Thiery, J. (2018). Alteration of the Antitumor Immune Response by Cancer-Associated Fibroblasts. Frontiers in immunology 9, 414.

Ziemssen, T., and Schrempf, W. (2007). Glatiramer acetate: mechanisms of action in multiple sclerosis. International review of neurobiology 79, 537-570.

Zilio, S., and Serafini, P. (2016). Neutrophils and Granulocytic MDSC: The Janus God of Cancer Immunotherapy. Vaccines 4.

Zitvogel, L., Galluzzi, L., Kepp, O., Smyth, M.J., and Kroemer, G. (2015). Type I interferons in anticancer immunity. Nature reviews Immunology 15, 405-414.

Zitvogel, L., Pitt, J.M., Daillere, R., Smyth, M.J., and Kroemer, G. (2016). Mouse models in oncoimmunology. Nature reviews Cancer 16, 759-773.

Zitvogel, L., Tesniere, A., and Kroemer, G. (2006). Cancer despite immunosurveillance: immunoselection and immunosubversion. Nature reviews Immunology 6, 715-727.

Zou, W., Machelon, V., Coulomb-L'Hermin, A., Borvak, J., Nome, F., Isaeva, T., Wei, S., Krzysiek, R., Durand-Gasselin, I., Gordon, A., et al. (2001). Stromal-derived factor-1 in human tumors recruits and alters the function of plasmacytoid precursor dendritic cells. Nature medicine 7, 1339-1346.

Zufferey, A., Speck, E.R., Machlus, K.R., Aslam, R., Guo, L., McVey, M.J., Kim, M., Kapur, R., Boilard, E., Italiano, J.E., Jr., et al. (2017). Mature murine megakaryocytes present antigen-MHC class I molecules to T cells and transfer them to platelets. Blood advances 1, 1773-1785.

222