Reproductive Benefits Conferred by Genetically Foreign Cells That Persist in Mothers and Offspring

Home , Catabolysis, Microchimerism

Reproductive benefits conferred by genetically foreign cells that persist in mothers and offspring

A dissertation submitted to the Graduate School of the University of

Cincinnati in partial fulfillment of the requirements for the degree of:

Doctor of Philosophy (Ph.D.)

In the Immunology Graduate Program of the University of Cincinnati College of Medicine

2016

By Jeremy M. Kinder

B.S., Ball State University, 2010 M.S., Ball State University, 2012

Committee Chair: Sing Sing Way, M.D.,Ph.D.

Abstract Our genetic makeup is equally inherited from our mother and father through classical

Mendelian genetics. However, each individual is constitutively chimeric, containing genetically foreign cells vertically transferred from their mothers during in utero development. These rare maternal microchimeric maternal cells have been identified to persist many years after birth and across a wide range of tissues demonstrating remarkable immune tolerance to our mothers persists in us long after we are born.

Reciprocally during pregnancy, mothers also receive a transfer of fetal cells that persist after parturition. We have shown that transfer of these genetically foreign cells during gestation is not accidental, but instead these cells are intentionally retained to promote reproductive fitness in subsequent pregnancies. Herein, we will explore the teleological and immunological implications of the long-term retention these genetically foreign cells.

iii Acknowledgements

I could not have made it this far without great mentorship. I would like to express my sincere gratitude to my advisor, Dr. Sing Sing Way. Your critical feedback, patience and scientific drive have shaped who I am as a scientist and as an individual. Over the last few years, I have also gained valuable insight observing how you balance a successful scientific career while also maintaining a fulfilling family life. Without your support and guidance, I would not have been able to achieve what I have done so far and what I will do in the future.

Before I started my PhD career, Dr. Heather Bruns was another mentor that helped me develop critical scientific skills during my undergraduate and master’s training. There are many intangible skills I have learned from you that I will carry with me throughout my career. These include continuous scientific innovation and organization while maintaining a strong commitment to your family life. Over the last 7 years I have valued not only your scientific mentorship but also your friendship.

I would also like to thank the members of my dissertation committee: Drs. Harinder

Singh, David Hildeman, Aimen Shaaban, and Jonathan Katz. The time you have devoted during and outside of our meetings to provide important critical feedback on my work is invaluable. Furthermore, I appreciate the investment you have made into not only my time as a graduate student but also advising me in future career decisions as well. I look forward to working with all of you as scientific peers in the future.

iv I also owe a lot to the current and past lab members who have provided valuable feedback but also contributed technically to some of the work outlined in this dissertation. I could not have done all of this work without your help and support.

I am blessed by the tremendous amount of support my family and friends have given me over the last few years. Pam, Richard and Craig, your encouragement of me and the support you have given to Christina and I is irreplaceable. I have truly felt like a part of your family over the 15 years Christina and I have been together. I owe both my scientific curiosity and tenacity to my mom, Teresa who has always told me I can accomplish whatever I set my mind to and who has forever been right behind cheering me on. To my wife, Christina, your love, dedication and constant support of my career goals has kept me going over the last several years. Importantly, you have kept me grounded during many stressful times and humble over the last 10 years. Finally, to my dad and daughter - while you cannot be here physically you have been and will continue to be my inspiration.

v Abbreviations APC antigen-presenting cell BAC bacterial-artificial chromosome CG chorionic gonadotropin CNS1 conserved non-coding sequence 1 CRE Cre recombinase DC dendritic cell DNA deoxyribonucleic acid DT diphtheria toxin DTR diphtheria toxin receptor EAE experimental autoimmune encephelomyelitis FACS fluorescence-activated cell sorting FISH fluorescence in situ hybridization FOXP3 forkhead box P3 Gal1 galectin 1 GFP green fluorescent protein GITR glucocorticoid-induced TNFR-related protein GVHD graft-versus-host disease HLA human leukocyte antigen IDO indoleamine 2,3 dioxygenase IFNg interferon-gamma IL-10 interleukin 10 IPEX immunodysregulation polyendocrinopathy enteropathy X-linked syndrome LCMV lymphocytic choriomeningitis virus Lm Listeria monocytogenes MHC major histocompatibility complex MS multiple sclerosis NFAT nuclear factor of activated-Tcells NIMA non-inherited maternal antigen NIPA non-inherited paternal antigen OVA chicken egg ovalbumin PCR polymerase chain reaction RA rheumatoid arthritis

vi Rh rhesus factor SCID severe combined immunodeficiency SLE systemic lupus erythematosus Smad3 mothers against decapentaplegic homolog 3 TCR T-cell receptor TGF-B transforming growth factor beta Th T helper cell Treg regulatory T cell

vii Table of Contents

Abstract ii Acknowledgements iv List of Abbreviations vi Table of Contents viii

CHAPTER 1: General Introduction 1 1.1 Introduction 2 1.2 Evolution of the placenta 2 1.3 Immunological conundrum of pregnancy 4 1.4 Maternal-fetal tolerance during pregnancy 6 1.5 Systemic immunological shifts during pregnancy 9 1.6 Fetal tolerance to non-inherited maternal antigen 17 1.7 Genetically foreign microchimeric cells 18 1.8 Dissertation Aims 23 1.9 References 26

CHAPTER 2: Pregnancy-induced maternal regulatory T cells, 36 bona fide memory or maintenance by antigenic reminder from fetal cell microchimerism?

CHAPTER 3: Tolerance to noninherited maternal antigens, 48 reproductive microchimerism and regulatory T cell memory: 60 years after ‘Evidence for actively acquired tolerance to Rh antigens’ 3.1 Abstract 49 3.2 Introduction, pioneering observations on immunological 50 tolerance by Dr. Ray Owen 3.3 Human immunological tolerance to noninherited 51 maternal antigens 3.4. Animal models of immune tolerance with early 57 developmental antigen exposure 3.5 Teleological benefits and immunological consequences 62 of NIMA-specific tolerance 3.6 Concluding perspectives 68

viii 3.7 References 70

CHAPTER 4: Offspring’s tolerance of mother goes viral 79

CHAPTER 5: Cross-generational reproductive fitness 87 enforced by microchimeric maternal cells 5.1 Summary 88 5.2 Introduction 89 5.3 Results 91 5.4 Discussion 100 5.5 Experimental procedures 104 5.6 References 108 5.7 Main figures 113 5.8 Supplementary figures 121

Chapter 6: Sustained protection against fetal wastage 127 conferred by prior pregnancy despite numerical loss of maternal regulatory CD4+ T cell memory 6.1 Abstract 129 6.2 Body 130 6.3 Materials and methods 138 6.4 Main figures 141 6.5 Extended data figures 145 6.6 References 149

Chapter 7: Summary and Discussion 152 7.1 Overall summary 153 7.2 Discussion 157 7.3 References 169

Appendices 172 Appendix I: Reuse Licenses 173

ix Appendix II: pdf of Chapter 2 185 Appendix III: pdf of Chapter 3 189 Appendix IV: pdf of Chapter 4 203 Appendix V: pdf of Chapter 5 206

x Chapter 1: General Introduction

1 Introduction

Reproductive success is essential for the survival of all species. In turn, this immense selective pressure drives continuous refinement and evolutionary conservation of methods for procreation that promote reproductive fitness. One such advantageous adaptation allowed the development of the embryo from within the maternal body known as viviparity. In contrast to other methods of reproduction such as laying of eggs

(oviparity), viviparity allows prolonged maturation of offspring with minimal threat of destruction from predation and permits long-term regulated transfer of parental nutrients throughout development1. Broadly speaking, despite reduced numbers of offspring, increased investment of time and resources devoted to the developing fetus with viviparity drastically increases their chances of survival and reduces the wasted allocation of resources to offspring that will likely not survive.

Evolution of the placenta

Emergence of viviparity simultaneously required additional physiological adaptations to facilitate development of offspring within the mother. Formation of a specialized organ known as the placenta allows attachment of the fetus to maternal uterine tissue and regulates exchange of nutrients, gas and waste between mother and fetus2,3. Although the placenta has evolved independently amongst multiple different animal groups

(cartilaginous fish, reptiles, amphibians, etc.) all viviparous mammalian species require a placenta for successful reproduction3,4.

2 In mammals the placenta is formed early in gestation by a specialized lineage of fetal

cells known as trophoblasts that continuously expand to form a multilayered organ3,5.

Early in pregnancy, trophoblasts initiate systemic physiological changes through release

of hormones such as chorionic gonadotropin (CG) that halts shedding or resorption of

the uterine decidua and instead triggers its expansion. In turn, fetal trophoblasts

continuously expand and invade maternal uterine tissue to form the placenta. This

process mediates anchoring of the developing fetus to the uterine wall and allows

regulation and exchange of nutrients, gas and waste products. While the structure of the

placenta and intimacy of its interaction with maternal circulation differs between

mammalian species, hemochorial placentation is shared among rodents, humans and

non-human primates (Fig 1). In contrast to other placental types, fetal trophoblasts

aggressively invade maternal tissue forming villous branches that are directly bathed in

maternal blood with hemochorial placentation, and interact directly with maternal tissue COMMENTARY allowing efficient nutrient and waste exchange (Fig 1)3,6. However, this intimate pregnancy complications such as preeclamp- abEpitheliochorial Endotheliochorial c Hemochorial sia and fetal growth restriction. Moreover, women with endometriosis have higher rates of implantation failure that are associated with defects in stromal cell decidualization and phenotypic alterations of uNK cells5,6. An increased understanding of early pregnancy biology is fundamentally important for the diagnosis, prevention and treatment of fertility and pregnancy problems as well as fertility-associated diseases in women. A key gap in our understanding of the human uterus and placenta is in early pregnancy, the time when the majority of pregnancies are Cow, pig, horse Dog, cat Human, rodent Group Publishing Vicari/Nature Katie lost. Therefore, it is essential to obtain and Figure 1 Classification of placentas based on histological assessment of the maternal-chorion infer information from animal models on interface.Figure 1. Classification of placental structures (a) Epitheliochorial placenta as seen in cow, pig and sheep: the barrier between maternal uterine and placental function during this bloodReprinted and the chorionby permission (tan cells) of Macmillanconsists of thePublishers maternal Ltd: vascular Nat Immunol, endothelium copyright and uterine 2015 epithelium critical early window, given the ethical limita- (blue cells). Red cells are vascular endothelium. (b) Endotheliochorial placenta as observed in dog and tions associated with obtaining samples from cat: the barrier between maternal blood and the chorion consists of the maternal vascular endothelium. (c) Hemochorial placenta as seen in human and small3 rodents (such as rat and mouse): maternal blood women. directly bathes the chorionic villi. All mammalian uteri contain endometrial glands that synthesize or transport and secrete transducer of Notch signaling, in mice. The placentas, differences occur in interdigitation, substances essential for survival and develop- results included impaired decidualization whereby fetal tissues branch to form villi that ment of the conceptus7. Uterine glands and and embryo implantation as a consequence of are either bathed in maternal blood or covered their secretions have biological roles in blas- decreased progesterone receptor expression. in maternal tissue. These differences result in tocyst and embryo survival and implantation, In addition, Rbpj-deficient mice had delayed variations in the degree of contact between uterine receptivity and stromal cell decidu- postpartum repair as a consequence of chronic maternal and fetal tissues at the area where alization in humans and animal models. inflammation and immunosuppression. Thus, exchange of maternal resources and nutrients Tom Spencer (Washington State University, Notch receptors play important physiological occurs. Moreover, some emerging studies Pullman) and colleagues found that infertil- roles during the establishment of pregnancy demonstrate marked differences in placental ity and recurrent pregnancy loss observed as well as postpartum repair through modula- structure between genetic strains of rat, and in ovine and mouse uterine gland–deficient tion of the immune environment. we have yet to recognize and/or exploit similar models unequivocally supports a primary From the distribution of placental types, it differences within the widely diverse strains role for uterine glands and, by inference, is clear that no other mammal has a placenta of genetic mouse models. Thus, appropriate their secretions present in uterine luminal identical to that of the human; although mor- animal models can be found, but they should fluid in the survival and development of the phologically similar, the placentas of some be chosen on the basis of the specific scien- embryo. The development and function of the nonhuman primates demonstrate less tropho- tific question being addressed. Furthermore, uterine glands within the implantation site is blast invasiveness than is observed in humans. study of a variety of animal models can, as in regulated by trophoblast, decidual and NK Placentas are classified in several ways. One the case of placental lactogens, reveal key fea- cell factors8. type of classification is based on the tissues tures conserved across taxa and their impor- During the establishment of pregnancy, that exist between the chorion and maternal tance in placental function. It is possible for decidualization in primates and rodents is a blood (Fig. 1). The major function of the researchers to obtain human term placen- multifaceted process. Notch, an evolutionarily placenta is to allow diffusion of nutrients tas after delivery and, in some instances, to conserved arbiter of cell fate, regulates diverse and oxygen from the maternal blood to the obtain placentas from elective terminations functions including cell survival, proliferation fetal blood through the chorion and of waste of pregnancy in the first trimester. However, and differentiation. In mammalian species, products from the fetus back to the mater- the period of development between embryo the role of Notch has been studied extensively nal blood. This exchange occurs through the implantation and 8 weeks after fertilization in the context of cancer biology, and Notch permeable membrane of the placenta. This is inaccessible, and embryo-maternal inter- pathway inhibitors are being developed that membrane may be composed of maternal actions during this early time frame of preg- target cancer stem cells. Notch1, one of four blood vessel endothelium, connective tissue, nancy differ substantially from those later in Notch receptors, mediates stromal cell sur- uterine luminal epithelium and chorion (epi- pregnancy8. It is during this early period that vival during decidualization, and the absence theliochorial, Fig. 1a), maternal blood vessel the major placental cell types differentiate of Notch1 results in decreased cell prolif- endothelium and chorion (endotheliochorial, and placental structures form; accordingly, eration, increased apoptosis and impaired Fig. 1b), or chorion (hemochorial, Fig. 1c). this is a critical window of time that needs to decidualization in the context of endome- Nonetheless, many features of the human be studied in order to understand the funda- triosis5. To further evaluate the role of all placenta are widely shared, and the disc-like, mental biology that contributes to placental Notch receptors in the context of endometrial hemochorial placental type of humans repre- dysfunction in humans. Laura Schulz and biology, Asgerally Fazleabas and co-workers sents ancient mammalian character states that R. Michael Roberts (University of Missouri, at Michigan State University inhibited all emerged well before the origin of primates. Columbia) have found that human embryonic canonical Notch signaling by generating a Whereas rodent placentas share hemocho- stem cells (hESCs) and induced pluripotent uterine-specific deletion of Rbpj, the nuclear rial and disc-like features with the human cells (iPSCs) provide an opportunity for

NATURE IMMUNOLOGY VOLUME 16 NUMBER 4 APRIL 2015 329 relationship between mother and offspring presents a new immunological conundrum since half of the genetic makeup expressed by the developing fetus is inherited from the father and is therefore genetically foreign to the mother. In other words, the benefits of placentation that facilitate viviparity require additional adaptations to prevent mothers from immunologically rejecting genetically foreign fetal cells and tissues.

Immunological Conundrum of Pregnancy

Classical immunological pillars discriminate antigens in individuals as either being “self” or “non-self”. This binary classification explains why adaptive immune components with specificity for genetically foreign antigens unique to microbial pathogens are readily primed after infection, whereas silenced activation of immune components with specificity to genetically encoded self-antigens protects against autoimmunity. In each individual, immunological “self” is determined by genetically encoded major histocompatibility complex (MHC) haplotype antigens7,8. All nucleated cells in an individual contain a unique, MHC signature or haplotype that is inherited equally from their mother and father. In addition to encoding self-recognition, MHC haplotypes allow each individual within a population to respond to infection in an immunologically unique fashion9. While MHC diversity favors species survival, as some individuals will inherent resistance to certain infections, it also complicates the immunology of pregnancy since the recognition of “non-self” MHC has typically been associated with immune activation that mediates rejection.

4 However, this traditional definition of “self” versus “non-self” is near exclusively based on experimental analyses using genetically-identical strains of inbred mice that artificially eliminates MHC haplotype heterogeneity that exists among individuals in outbred mammalian species. In nature however, mating between MHC haplotype identical individuals does not occur and therefore this approach likely masks important physiological shifts in immune tolerance that occur during pregnancy to compensate for compulsory exposure to foreign paternal antigens expressed by the developing fetus.

Evidence that exposure to genetically foreign antigens during gestation actively primes immune tolerance was first described by Ray Owen in a seminal characterization of fraternal twin cattle10. Here, the fused blood circulation [vascular anastomoses] between these dizygotic twins permitted compulsory exposure to genetically foreign blood cells from the other twin that remarkably persisted into adulthood resulting in natural blood chimerism10. Expanding on Owen’s findings, Sir Peter Medawar remarked that the phenomenon of ‘rejection’ observed following transplant of allogeneic skin grafts was remarkably absent during pregnancy despite an equivalent presence of genetically foreign tissues11,12. Based on this observation, Medawar proposed several theories to explain how tolerance to these foreign antigens may occur. One hypothesis was that the placenta forms an impenetrable barrier that prevents intermingling of foreign fetal placental antigen and maternal immune cells thereby averting rejection11,13,14. However, the requirement for the placenta to mediate exchange of essential nutrients and waste between mother and fetus implies some amount of porosity must exist within in placental tissue.

5 Medawar additionally hypothesized that broad maternal immune suppression during

pregnancy may also promote survival of the fetal allograft. Given that pregnant women

are still able to mount sufficient immune responses, and therefore survive minor

infections, this is also not likely the case. Given the importance of reproductive success

for all mammalian species, it is perhaps not surprising that multiple non-overlapping

layers of immunological tolerance mechanisms have evolved that protects the

developing fetus from immunological attack or rejection by maternal immune cells.

Maternal-fetal tolerance during pregnancy

Throughout, gestation maternal immune cells are present within the decidua and

therefore actively encounter fetal antigens at the placenta interface15. Viewing the fetus REVIEW REVIEW as a semi-allogeneic graft, averting “rejection” requires preventing aberrant activation of

Figure 2 Molecular basis of T cellFigure 2 Molecular basisa of T cellDirect a DirectSemidirect SemidirectIndirect adaptive Indirect immune components with allorecognition and activation. allorecognition and activation. Allogeneic Allogeneic (a) Allorecognition occurs in one(a )of Allorecognition three occurs in one of three MHC or minor MHC or minor ways when T cells interact withways allogeneic when T cells interact with allogeneic antigen–derived antigen–derivedspecificity to genetically foreign fetal target cells or peptides. In directtarget cells or peptides. In direct proteins proteins allorecognition, T cells are activatedallorecognition, directly T cells are activatedAllogeneic directly Allogeneic target cell target cell Autologous antigens.Autologous Graft rejection following by allogeneic cells (antigen-presentingby allogeneic cells cells (antigen-presentingor APC cells or APC APC APC (APCs) or any cell expressing allogeneic(APCs) or any cell expressing allogeneic MHC). T cell receptor activationMHC). is triggered T cell receptor activation is triggered transplantation, is typically mediated by recognition of the complex comprisingby recognition of the complex comprising Costimulatory Costimulatory Costimulatory Costimulatory Costimulatory MHC AllogeneicMHC MHC Costimulatory AllogeneicMHC MHC class II MHC class II pathway pathway pathway determinants of the allogeneic determinantsMHC of the allogeneic MHC pathway pathway pathway class I or MHC class II moleculeclass loaded I or withMHC class II molecule loaded with Peptide PeptidePeptide ProcessedPeptide Processedthrough two non-overlapping pathways peptide (the peptide origin is irrelevant).peptide (the peptide origin is irrelevant). allogeneic allogeneic In semidirect allorecognition, allogeneicIn semidirect allorecognition,TCR allogeneic TCR TCR peptideTCR TCR peptide TCR MHC class I or MHC class II moleculesMHC class I or MHC class II molecules known as direct or indirect allo- are acquired and MHC-peptide arecomplexes acquired and MHC-peptide complexes displayed by autologous APCs. displayedThe peptide by autologous APCs. The peptide 16 origin is again irrelevant. For indirectorigin is again irrelevant. For indirectT cell T cell recognition (Fig 2) . Direct recognition allorecognition to occur, allogeneicallorecognition to occur, allogeneic transplantation-relevant proteinstransplantation-relevant must proteins must involves interaction of the intact first be processed by autologousfirst APCs. be processed by autologous APCs. Subsequently, peptides derivedSubsequently, from these peptidesb derived from these Figureb 2. Pathways of T2 cell allo-recognition.2 allogeneic antigens are cross-presentedallogeneic by antigens are cross-presented by foreign MHC present on a donor autologous MHCII on autologousautologous APCs. T cellMHCII on autologous APCs. T cell ReprintedCD80 or CD86 by permissionCD28 CD80 of Macmillan or CD86 CD28 APC APC receptor activation is triggered receptorby recognition activation is triggered by recognition Publishers Ltd: Nat Biotechnol, copyright 2014 of the complex comprising determinantsof the complex of comprising determinants of antigen-presenting cell (APC) with the the autologous MHCII moleculethe loaded autologous with MHCII molecule loaded with MHC + MHCTCR + TCR peptide peptide an allogeneic peptide. (b) T cellan activation allogeneic by peptide. (b) T cell activation by 1 161 antigen-presenting cells requiresantigen-presenting two types cells requires two types T cell receptor (TCR) of a host T cell (Fig 2) . In this case, direct recognition of the non- of signals. Signal 1 is received ofby signals. the TCR Signal 1 is received by the TCR CD80 or CD86 CTLA-4 CD80 or CD86 T cell CTLA-4 T cell when it interacts with a cognatewhen peptide-MHC it interacts with a cognate peptide-MHC complex. Signal 2 is received bycomplex. costimulatory Signal 2 is received by costimulatory PD-L1 PD-1 PD-L1 PD-1 Kim Caesar/Nature Publishing Group molecules (e.g., CD28) when theymolecules interact (e.g., CD28) when they interact Kim Caesar/Nature Publishing Group with cognate ligands (e.g., CD80with or cognateCD86). Moleculesligands (e.g., that CD80 inhibit or TCD86). cell activation Molecules and/or that inhibitinduce T cell activationanergy include and/or cytotoxic induce T cell anergylymphocyte include cytotoxic T cell lymphocyte antigen 4 (CTLA-4) and programmedantigen cell 4 (CTLA-4) death protein and programmed1 (PD-1). Programmed cell death proteindeath ligand 1 (PD-1). 1 (PD-L1) Programmed is the main death ligand ligand for 1 the(PD-L1) PD-1 is receptor. the main6 ligand for the PD-1 receptor. evidence, show that adoptiveevidence, transfer showof antigen-presenting that adoptive transfer cells, of orantigen-presenting host NKT cells26 reducedcells, orexpansion host NKT of cellsalloreactive26 reduced donor expansion T cells of alloreactive donor T cells regulatory T cells, NKT cells orregulatory mesenchymal T cells, stromal NKT cells cells or may mesenchymal have and stromal ameliorated cells may graft-versus-host have and ameliorated disease without graft-versus-host impairing graft- disease without impairing graft- tolerogenic effects, suggestingtolerogenic that they could effects, provide suggesting an alternative that they could versus-tumor provide an effects alternative in mouse versus-tumor HSC transplantation effects in mousemodels. HSC NKT transplantation models. NKT to pharmacological immunosuppressionto pharmacological in the context immunosuppression of regenerative in thecells context have the of regenerativeimportant advantage cells have that the tolerization important to advantage alloantigens that tolerization to alloantigens medicine. However, it is importantmedicine. to recognize However, thatit is anyimportant allogeneic to recognize can be thatachieved any allogeneic using autologous can be cells, achieved whereas using the autologous evidence to cells, date whereas the evidence to date cell is inherently immunogeniccell and, is inherently if injected immunogenic into an immunocom and, if injected- suggests into anthat immunocom dendritic cells- andsuggests NK cells that must dendritic be of cellsdonor and origin. NK cells must be of donor origin. petent recipient, risks eliminationpetent byrecipient, alloreactive risks hostelimination T cells andby alloreactive Regulatory host T cells andhave beenRegulatory at the center T cells of attention have been of attrans the- center of attention of trans- NK cells. It may therefore beNK necessary cells. It tomay reduce therefore this riskbe necessary through toplantation reduce this immunologists risk through forplantation more than immunologists a decade. Extensive for more data than a decade. Extensive data pharmacological or antibody-basedpharmacological immunosuppression, or antibody-based at least durimmunosuppression,- from animal studiesat least durshow- thatfrom a animalpolyclonal studies population show that of donora polyclonal population of donor ing the initial phase after transplantation.ing the initial phase after transplantation. regulatory T cells can induceregulatory tolerance T tocells alloantigens, can induce provided tolerance to alloantigens, provided Dendritic cells are powerfulDendritic antigen-presenting cells are powerful cells and antigen-presenting have that it contains cells and sufficient have numbersthat it contains of alloantigen-specific sufficient numbers regu of- alloantigen-specific regu- traditionally been used to activatetraditionally the immune been responseused to activate against the anti immune- latory response T cells against to suppress anti- donorlatory alloreactive T cells to suppresseffector Tdonor cells 27–alloreactive30. effector T cells27–30. gens such as a tumor-associatedgens antigenssuch as 1a9 .tumor-associated Nevertheless, infusion antigens 1Although9. Nevertheless, purified infusion antigen-specific Although regulatory purified Tantigen-specific cells are generally regulatory T cells are generally of immature or ‘semi-mature’of dendriticimmature cells or ‘semi-mature’ (the latter character dendritic- cellsmore (the effective latter thancharacter polyclonal- more regulatory effective T than cells, polyclonal as yet there regulatory are no T cells, as yet there are no ized by increased expressionized of MHC by increased and costimulatory expression moleculesof MHC and standardized costimulatory protocols molecules for efficient, standardized good manufacturing protocols for practice efficient, (GMP)- good manufacturing practice (GMP)- in the absence of proinflammatoryin the cytokine absence ofproduction) proinflammatory20 was shown cytokine compliant production) generation20 was shown of large compliant numbers of generation antigen-specific of large regulatory numbers of antigen-specific regulatory to promote transplantation toleranceto promote in transplantationseveral preclinical tolerance studies, in severalT cells. preclinical But GMP-compliant studies, T isolation cells. But and GMP-compliant expansion of polyclonal isolation and expansion of polyclonal including prevention of kidneyincluding allograft prevention rejection in of a kidneyrhesus macaqueallograft rejectionregulatory in a rhesus T cells macaque is already feasibleregulatory31, and T cells this isapproach already couldfeasible make31, and this approach could make model21 and treatment of lethalmodel graft-versus-host21 and treatment disease of lethal in mousegraft-versus-host banking diseaseof MHC-typed in mouse polyclonal banking regulatory of MHC-typed T cells forpolyclonal off-the-shelf regulatory T cells for off-the-shelf HSC transplantation models2HSC2. The transplantation tolerogenic properties models2 2of. The imma tolerogenic- use in properties the setting of of imma autoimmunity- use in theor alloimmunitysetting of autoimmunity a reality in orthe alloimmunity a reality in the ture dendritic cells have beenture attributed dendritic to cells a number have been of possible attributed nearto a future.number of possible near future. mechanisms, including lack mechanisms,of efficient co-stimulation including lack that of efficientresults co-stimulationEx vivo–expanded that results polyclonal Ex regulatory vivo–expanded T cells polyclonal have been regulatorytested T cells have been tested in T cell anergy or inhibitionin of T clonalcell anergy T cell or expansion, inhibition as of well clonal as T incell clinical expansion, trials. as The well results as in of clinical four trials trials. in Thetype results 1 diabetes of four and trials in type 1 diabetes and promotion of regulatory T cellspromotion23. of regulatory T cells23. graft-versus-host disease havegraft-versus-host been reported 3disease2, and severalhave been trials reported 32, and several trials Induction of antigen-specificInduction tolerance of by antigen-specific antigen-presenting tolerance comparing by antigen-presenting regulatory T cell immunotherapycomparing regulatory to conventional T cell immunotherapy immuno- to conventional immuno- cells is not limited to dendriticcells cells. is not B cellslimited genetically to dendritic modified cells. toB cellssuppressive genetically regimens modified are to undersuppressive way32,33 .regimens The available are under clinical way evi32-,33. The available clinical evi- express a tolerogenic IgG-antigenexpress fusion a tolerogenic protein have IgG-antigen shown efficacy fusion proteindence have to date shown suggests efficacy that regulatorydence to date T cell suggests therapy that is feasible, regulatory safe T cell therapy is feasible, safe in experimental models of autoimmunein experimental diseases, models such of as autoimmune encephalitis diseases,and does such not as encephalitiscompromise protectiveand does notimmunity. compromise A few protectivetrials have immunity. A few trials have or uveitis24. And adoptive immunotherapyor uveitis24. And using adoptive donor immunotherapy NK cells25 demonstrated using donor efficacyNK cells in25 acute demonstrated graft-versus-host efficacy disease in acute32,33 graft-versus-host. disease32,33.

788 788 VOLUME 32 NUMBER 8 AUGUSTVOLUME 2014 32 NATURENUMBER BIOTECHNOLOGY 8 AUGUST 2014 NATURE BIOTECHNOLOGY self MHC stimulates an extremely robust inflammatory response. During pregnancy, there is drastically reduced expression of MHC molecules on placental trophoblasts whereby the risk of direct allo-recognition of foreign fetal-paternal MHC complexes by maternal immune cells is minimized17,18.

Indirect recognition, results from recognition of foreign MHC or other minor alloantigens that are processed and presented by host APCs to host T cells (Fig 2)16. During pregnancy, maternal APCs such as dendritic cells (DCs) that reside within the decidua consistently encounter shed placental antigens and therefore require mechanisms to prevent unnecessary indirect allo-activation that could mediate fetal demise. One mechanism for how this is achieved is through restriction of the recruitment of maternal

APCs to the outer myometrium and therefore limiting potential contact with fetal antigens present within the decidua19,20 (Fig 3). Furthermore, the paucity of lymphatic vessels within endometrial and decidual tissues leads to ‘DC entrapment’ preventing any activated DCs within the decidua from emigrating to the nearby draining lymph nodes21,22 (Fig 3).

7 Many other non-overlapping immune suppressive pathways are in place to help re- enforce fetal tolerance within the local maternal-fetal interface. These include secretion of the enzyme indoleamine 2,3-dioxygenase (IDO) by multiple maternal decidual and fetal trophoblast cells23. The presence of IDO facilitates the catabolysis of the essential amino acid tryptophan into its metabolites 3-OH-kynurenine and 3-OH-anthranilic acid24,25. Here, both the depletion of tryptophan that is essential for proper T cell activation and increased presences of its metabolites have all been shown to mediate suppression of T-cell responsiveness24,25. The importance of IDO during fetal tolerance is demonstrated during mouse pregnancy where pharmacological inhibition of IDO results in fetal loss26.

In addition to IDO, expression of other anti-inflammatory molecules including TGF-β and

Galectin-1 has been demonstrated in maternal endometrial cells27-29. TGF-β expression mediates fetal tolerance through dampening pro-inflammatory cytokine production and promoting anti-Th1 responses30,31. In turn, forced expression of the TGF-β agonist known as “lefty” within the mouse endometrium resulted in significantly increased infertility due to implantation failure32,33. Galectin-1 is expressed not only on maternal but also healthy fetal placental tissue34, and mediates suppression by a variety of mechanisms including induction of tolerogenic IL-10-producing DCs in addition to directly antagonizing TCR-induced T-cell activation35,36. Its importance in reproduction is suggested by studies showing decreased galectin-1 expression in placental trophoblasts associated with human pregnancy complications37. This requirement is confirmed in mice where maternal galectin-1 deficiency (Gal1-/-) triggers increased fetal

8 loss 38. Together, these checks and balances along with many others are the first line of defense that has evolved to avert fetal antigen recognition and subsequent rejection.

Systemic immunological shifts during pregnancy

In addition to development of immune suppressive mechanisms at the maternal fetal interface, complementary shifts in systemic immune responsiveness also occur during pregnancy. The best evidence of systemic immunological shifts in mothers in non- reproductive tissues is consistent alterations in the severity of autoimmune disorders that occur during pregnancy. During pregnancy, women suffering from multiple sclerosis

(MS), regularly experience reductions in relapse and even complete remission of symptoms39-41. Additionally, nearly 75% of women suffering from rheumatoid arthritis

(RA) also experience marked improvement and relief from symptoms during pregnancy42,43. These clinical observations in humans are also recapitulated in animal models of autoimmunity during pregnancy. In mice, progression and relapse of experimental autoimmune encephalomyelitis (EAE) that mimics demyelination and paralysis associated with MS in humans is also diminished during pregnancy44-46.

Similarly, immune shifts during pregnancy reduce symptoms associated with experimental autoimmune arthritis47. Interestingly, in nearly all women, relief during pregnancy is followed by relapse following delivery suggesting these shifts that dampen immune responsiveness to self-antigens is restricted to the gestational period39,41,43,48.

Interestingly, these alterations to maternal immune responsiveness during pregnancy do not always yield beneficial relief of autoimmune symptoms. Systemic lupus

9 erythmatosus (SLE) is associated with antibody-mediated destruction of multiple tissues

(joints, skin, lung, etc.) and disproportionally affects reproductive age women. For SLE, pregnancy is associated with exacerbated symptoms. Together these findings suggest systemic maternal immune cell shifts during pregnancy that favor tolerance to the genetically foreign fetus also extend to immune components with specificity to certain self-antigens. Given that many of these conserved self-antigens are likely to be shared between mother and offspring, these diminished responses would also help avert fetal rejection.

FOXP3+ regulatory T cells

While uncomplicated healthy pregnancy was initially associated with Th2 polarization of maternal CD4+ T cells, a more comprehensive understanding of previously undiscovered CD4+ subsets has shifted this perspective to instead favor skewing away from IFNγ-producing Th1 cells. In turn, following the discovery of the immune- suppressive subset of CD4+ known as regulatory T cells (Tregs), how this population contributes to peripheral immune tolerance during pregnancy has become a main focus of reproductive immunology.

Tregs are an immune-suppressive subset of CD4+ T cells defined by expression of the transcription factor FOXP3 49-51. Initial studies of these cells revealed their dominant role in preventing autoimmunity through silencing activation of adaptive immune components with specificity to self-antigens that bypass central tolerance 52-54. In humans, this is demonstrated by development of a systemic autoimmune disorder known as immune dysregulation, polyendocrinopathy, enteropathy X-linked (IPEX)

10 syndrome that parallels spontaneous mutations in FOXP3 and results in fatality within the first 6 months of life 55,56. In mice, both FOXP3 mutations and targeted depletion of

FOXP3+ cells precipitates mortality associated with immune-mediated multi-organ destruction 49,57-59.

Additional studies revealed expression of FOXP3 is not just restricted to CD4+ T cells during thymic development but can also be induced in CD4+ FOXP3- cells in the periphery following TCR stimulation in the presence of TGF-β60. Induction of Tregs from

FOXP3- cells relies on interaction of transcription factors including Smad3 and NFAT with the intronic FOXP3 enhancer element known as CNS1 61,62. Interestingly, targeted deletion of CNS1 abrogates induction of Tregs in the periphery but not thymic-derived

Tregs and is associated with inflammation at mucosal sites commonly associated with commensal microbe colonization63. Evidence for peripheral FOXP3 induction in non-

Treg CD4+ cells that contain an infinite number of antigen-specificities suggest in addition to necessity of Tregs to protect against autoimmune diseases, their suppressive properties can also extend to beneficial non-self antigens.

Maternal regulatory T cells during pregnancy

In humans, the temporal expansion of maternal Tregs in both the peripheral blood and decidua is associated with normal healthy pregnancy64-66. While these immune cell shifts in humans provide evidence that maternal Tregs may enforce systemic tolerance to fetal antigen, the inability to selectively manipulate these cells in humans results in a limiting understanding of their cause and effect relationship. Here animal models of

11 pregnancy have allowed a more definitive analysis of the importance of maintained expansion of Tregs in mediating fetal tolerance.

The most useful studies involve experimental allogeneic matings, where MHC- discordant strains of mice are mated to more closely recapitulate the natural heterogeneity observed in outbred populations. In these models, accumulation of maternal Tregs during pregnancy parallels that observed in humans with expansion beginning early in pregnancy and peaking around mid-gestation67-70. Interestingly, blunted maternal Treg expansion is observed in an abortion-prone mating [DBA/2J male x CBA/J female] and adoptive transfer of Tregs from normal pregnancy mice rescue this phenotype71-73.

Employing animal models also allows a more definitive assessment of the importance of maternal Tregs to enforce fetal tolerance through observing pregnancy outcomes following experimental Treg manipulation. Initial studies showed both adoptive transfer of Treg-depleted (α-CD25) lymphocytes into T cell-deficient females prior to pregnancy or α-CD25 depletion in intact pregnant females drove increased fetal loss early in gestation (embryonic day [E] 2.5 -7.5)69,74. Later studies utilized more selective depletion of Tregs based on FOXP3 expression that circumvents limitations of α-CD25 depletion that targets Tregs but also activated FOXP3- T cells as well. Specifically, transgenic mice that express the high-affinity human diphtheria toxin receptor (DTR) with FOXP3 allows targeted ablation of Tregs with low-dose diphtheria toxin (DT)58. In line with previous observations, depletion of maternal Tregs during pregnancy in this manner resulted in near complete fetal loss68,75. Furthermore, the X-linked inheritance of

FOXP3 allows exploitation of random X-chromosome inactivation in females to generate

12 a Treg population in FOXP3DTR/WT females where only 50% of maternal Tregs express

DTR and are susceptible to DT-mediated ablation. Importantly, in these females administration of DT during pregnancy results in only transient depletion of Tregs to pre- pregnancy levels that circumvents pregnancy complications that result from systemic inflammation associated with total Treg depletion58. Here even partial transient ablation of maternal Tregs at mid-gestation [E11.5] is sufficient to drive fetal resorption68.

Together these results show targeted manipulation of maternal Tregs during pregnancy drives fetal loss and support the requirement for Tregs in successful pregnancy.

Maternal regulatory T cells and pregnancy complications

In line with the requirement for maternal Treg expansion to enforce fetal tolerance during normal pregnancy, many idiopathic human pregnancy complications are associated with blunted expansion or dampened functional capacity of maternal

Tregs65. For example, development of pre-eclampsia occurs in roughly 10% of all first time pregnancies and is associated on average with a ~30% decrease in maternal Treg expansion76-78. Interestingly, the underlying symptoms including systemic inflammation associated with pre-eclampsia is likely driven in response to fetal antigen, as delivery of the fetus and other products of conception still remains the only cure65,77. Similar decreases in maternal Treg expansion have been observed in cases of spontaneous abortion65. Here an ~33% decrease of Tregs identified as CD4+ CD25bright cells was observed in the peripheral blood of women with spontaneous abortion compared to healthy pregnancies79. Additional studies have demonstrated, a 3-fold decrease in CD4+

CD25bright Tregs within decidual tissue of women experiencing spontaneous abortion

13 compared to women following elective abortion79,80. In addition to blunted accumulation

of maternal Tregs, shifts in functional Treg potency have also been associated with

pregnancy complications including spontaneous abortion and pre-term birth81,82. Taken

together these studies suggest expansion of maternal Tregs throughout gestation is

critical to sustain fetal tolerance. In turn, their blunted accumulation or diminished

suppressive potency may be responsible for many of the leading pregnancy

complications. RESEARCH SUPPLEMENTARY INFORMATION

RESEARCH SUPPLEMENTARY INFORMATION Specificity of maternal regulatory T cells during pregnancy

Similar to other T cell subsets,

Tregs have defined antigen

specificity, controlled by the T cell

receptor (TCR) expressed by each

cell. Increasing evidence

shows the suppressive

function of Tregs require

cell-intrinsic TCR

stimulation, highlighting

an important role for

Treg antigen Figure 4. Models comparing maternal regulatory CD4 cell accumulation Reprinted by permission of Macmillan Publishers Ltd: Nature, copyright 2012 recognition Supplementary Figure 10. Models comparing maternal regulatory CD4 cell accumulation. In model 1 (top), pregnancy stimulates non-specific expansion of maternal FOXP3+ regulatory T in immune tolerance83. To address whether expansion of maternal Tregs is the result of cellsSupplementary before investigation Figure 10.using Models antigen comparing-specific tools. maternal In model regulatory 2 (bottom), CD4 materna cell accumulation.l regulatory T cells with fetal-antigen specificity selectively expand and accumulate during pregnancy.+ This nonIn- specificmodel 1 (top), expansion pregnancy during stimulates pregnancy non- spe[Figcific 4 expansion(Model 1)] of ormaternal is specifically FOXP3 regulatory driven by T modelcells before is supported investigation by data using presented antigen in-specific this paper tools. where In model maternal 2 (bottom), regulatory materna CD4 l cells regulatory with speciT cellsficity with for fetal a single-antigen peptide specificity antigen selectively expressed expand by the anddeveloping accumulate fetus during are found pregnancy. to expand This greatermodel isthan supported 100-fold, by while data FOXP presented3 expression in this paperamong where bulk maternal maternal CD4 regulatory cells accumulate CD4 cells less with thanspeci twoficityfold. for Furthermore,a single peptide we showantigen pregnancy expressed-induced by the maternal developing regulatory fetus are T foundcells with to expand fetal specificitygreater than are 100 pheno-fold,-typically while FOXP distinct,3 expression persist14 after among delivery bulk, maternaland rapidly CD4 re- expandcells accumulate and provide less protectionthan twofold. from Furthermore, fetal resorption we during show secondarypregnancy -pregnancy.induced maternal regulatory T cells with fetal specificity are pheno-typically distinct, persist after delivery, and rapidly re-expand and provide protection from fetal resorption during secondary pregnancy.

10 | WWW.NATURE.COM/NATURE

10 | WWW.NATURE.COM/NATURE maternal recognition of fetal antigen [(Fig 4 (Model 2)], recent studies have employed antigen-specific tools that allow tracking of CD4+ T cells with defined specificity to model antigens in vivo84,85.

Using male mice with ubiquitous cell-surface expression of MHC class II restricted (I-Ab)

86 2W1S55-68 (2W1S) to sire pregnancies in non-transgenic females transforms 2W1S into a surrogate fetal antigen and allows tracking of maternal CD4+ T cells with fetal

2W1S-specificity throughout gestation. In these studies, fetal-antigen specific Tregs expand by ~100-fold compared to only ~2-fold expansion in bulk maternal Tregs68,75.

Similarly, adoptive transfer of monoclonal TCR transgenic CD4+ T cells that recognize fetal-expressed antigen or Treg-depleted bulk CD4+ cells in intact females results in accumulation of CD4+ FOXP3+ maternal Tregs with fetal specificity68,87. Together these studies demonstrate expansion of maternal Tregs during pregnancy is primarily driven by induction of Tregs with specificity to fetal antigen. Furthermore, increased fetal resorption in mice with targeted deletion in CNS1 that lack induced but not thymic derived Tregs highlights the necessity for induction of FOXP3 in CD4+ cells in response to fetal antigen stimulation in order to establish and sustain fetal tolerance88.

Interestingly, this CNS1 genetic element required for induction of FOXP3 expression in non-Treg CD4+ cells is conserved in humans and other placental mammals but absent in marsupials and egg-laying mammals suggesting this adaptation co-evolved along with development of in utero placental gestation88.

15 Extension of these analyses reveals maternal Tregs with fetal-specificity are retained at

enriched levels up to 100 days postpartum. Furthermore, following secondary fetal-

antigen re-stimulation during secondary pregnancy retained maternal Tregs with fetal-

specificity re-expanded with accelerated tempo that suggests functional memory (Fig

5)68. In line with this observation, this expanded pool of maternal Tregs during RESEARCH SUPPLEMENTARY INFORMATION secondary pregnancy parallels highly efficient protection against fetal wastage following

partial transient Treg ablation in FOXP3DTR/WT females. In turn, these findings

demonstrate fetal-antigen stimulation during primary pregnancy establishes protective

maternal Treg memory that enforces fetal tolerance during subsequent pregnancies.

Here, protective maternal Treg memory observed in mouse allogeneic pregnancy,

parallels partner-specific benefits associated with reductions in the rate of pre-

eclampsia observed in successive pregnancies89,90.

Figure 5. Model showing postpartum retention and secondary re-expansion of Supplementarymaternal Tregs with fetal Figure 10. Models-specificity comparing maternal regulatory CD4 cell accumulation. In model 1 (top), pregnancy stimulates non-specific expansion of maternal FOXP3+ regulatory T Reprinted by permission of Macmillan Publishers Ltd: Nature, copyright 2012 cells before investigation using antigen-specific tools. In model 2 (bottom), maternal regulatory T cells with fetal-antigen specificity selectively expand and accumulate during pregnancy. This model is supported by data presented in this paper where maternal regulatory CD4 cells with Whilespeci ficity these for findingsa single peptide collectively antigen suggest expressed successful by the developing pregnancy fetus establishesare found to immuneexpand greater than 100-fold, while FOXP3 expression among bulk maternal CD4 cells accumulate less suppressivethan twofold. memoryFurthermore, that we is specificshow pregnancy to paternal-induced-antigens maternal expressed regulatory byT cells the with fetus fetal, the specificity are pheno-typically distinct, persist after delivery, and rapidly re-expand and provide stabilityprotection and from requirements fetal resorption for during maintenance secondary pregnancy.of maternal Treg memory to fetal antigen

10 | WWW.NATURE.COM/NATURE remain undefined. For effector T cells whose activation is associated with clearance of pathogenic microbes, requirements between CD8+ and CD4+ cell is better defined. In contrast to CD8+ T cell memory that persists despite complete antigen removal, CD4+ effector T cell memory relies on low-level antigen stimulation to persist 91-94. However, whether the need for antigenic reminders is also necessary to sustain CD4+ Treg memory remains unknown.

Fetal tolerance to non-inherited maternal antigen

Although reproductive immunology has primarily focused on maternal tolerance to discordant paternal antigens expressed by the developing fetus, the fetus is also reciprocally exposed to an equally vast array of immunologically foreign non-inherited maternal antigens (NIMA) during in utero development. This is especially important in humans and other species with a full complement of adaptive immune components at birth, where immune tolerance to NIMA is essential to prevent anti-maternal immunity during development95. In humans, TCR stimulation of CD4+ T cells during fetal development favors induction of FOXP3 expression and expansion of fetal Tregs functionally suppress expansion of NIMA-responsive T cells in utero95.

Exposure to NIMA beginning during in utero development has also been shown to confer remarkable long-lasting antigen-specific immune tolerance properties, analogous to postpartum maternal regulatory T cell memory to fetal antigen. For example in kidney transplantation among MHC mis-matched adult sibling donors, graft survival is markedly improved when matched for NIMA compared with non-inherited paternal antigen (NIPA)

MHC haploytes96. Additionally, the incidence of graft-versus-host disease is also

17 diminished after stem cell transplantation when recipients share MHC haplotypes with donor NIMAs 97,98. Similarly, increased tolerance to Rh erythrocyte antigens occurs among women born to Rh+ mothers99, and diminished antibodies to NIMA specific HLA haplotypes are found among transfusion development individuals broadly exposed to foreign HLA haplotype antigens100.

Interestingly, the sharply improved long-term survival of NIMA-matched solid organ donor allografts is recapitulated in NIMA-matched tissue allografts in mice. The best example of this is the long-term survival of heterotopic allogeneic heart transplants (H-

2d) in H-2d NIMA-exposed mice that are rapidly rejected in genetically-identical mice that were not exposed to maternal H-2d during development101. Likewise, protection from graft-versus-host disease (GVHD) from NIMA-matched donors in human stem cell transplantation is recapitulated in mouse models of donor bone marrow cell transfer and engraftment into irradiated recipients. For example, while wild-type mice quickly succumb to GVHD, an overwhelming proportion of genetically identical mice exposed to donor MHC as a NIMA survived following bone marrow transfer associated with a reduced GVHD score102. The conservation of tolerance to NIMA in mice and across other mammalian species with delayed immune cell ontogeny suggests in addition to averting anti-maternal immunity in utero, tolerance to NIMA may provide other postnatal teleological benefit that favors its evolutionary conservation101-105.

Genetically foreign microchimeric cells

Given the requirement for transfer of nutrients and waste products through a porous placental interface, it is perhaps not surprising that genetically foreign maternal and fetal

18 cells are bi-directionally transferred in parallel. However, what is remarkable is the long- term persistence of genetically foreign fetal cells in mothers after pregnancy and the postnatal persistence of maternal cells in offspring. While the survival of these cells that establish low-level microchimerism is perhaps the most definitive evidence that tolerance to reproductive-associated foreign antigens persists following gestation the question remains whether the retention of these cells is accidental or deliberate. In turn, if genetically foreign cells are deliberately retained, uncovering the teleological benefits that preclude their survival may unlock key clues into their preservation. Additionally, understanding how microchimeric cells are able to avoid rejection has the potential to redefine immunological identity beyond classical self versus non-self distinction by extending this definition to encompass developmentally relevant antigen sources.

Fetal microchimeric cells

Early studies that identified fetal cells based on discordant expression of Y-chromosome

DNA in mothers bearing male offspring uncovered that fetal cells existed in maternal circulation during gestation and following delivery106-109. Further technological advances in amplification of DNA through polymerase chain reaction (PCR) allowed increased specificity and quantification of these extremely rare cells 110. Detectible levels of fetal cells can be observed in maternal circulation early in gestation (4-5 weeks) rising throughout gestation and declining post-parturition111,112. Despite their decline postpartum, later studies that employed both fluorescent in situ hybridization (FISH) and polymerase chain reaction (PCR) to amplify Y-chromosome or HLA-discordant DNA revealed fetal cells are retained across multiple tissues even in healthy mothers113-117.

19 More in-depth analysis of fetal cells has uncovered a wide array of tissue-specific (e.g. cardiomyocytes, hepatocytes) and hematopoietic cell subsets suggesting fetal microchimeric cells have multi-lineage potential114,116,118,119. Other studies focusing on progenitor-like properties of microchimeric cells have also detected fetal cells in hematopoietic (CD34+) and mesenchymal stem cell populations during and following gestation120-124.

In line with the conservation of maternal immune tolerance mechanisms that develop in response to fetal-antigen stimulation, the trafficking and postpartum retention of fetal cells is also preserved across mammalian species including rodents and non-human primates125-131. Given the parallel existence of maternal Treg memory to fetal antigen together with postpartum persistence of fetal microchimeric cells, whether these cells serve as constant antigenic reminders that maintain maternal Treg memory is important to establish potential clinical application of protective properties of maternal Tregs.

Additional studies examining the consequences of persistent microchimeric cells suggest a role for fetal cells in autoimmune disease as increased levels of fetal cells have been recorded in target tissues of women with, SLE132,133, systemic sclerosis134-136 and many other autoimmune disorders115,137,138. However, the technical limitations preventing manipulation of fetal cells in vivo has led to controversy regarding the role of fetal microchimerism in driving or resolving disease138-140. Further development of animal models of autoimmunity where fetal microchimeric cells are detectable and can be manipulated will be imperative to understanding whether these cells contribute to

20 tissue repair or whether in some cases autoimmunity is driven by allo-immunity against discordant MHC haplotypes.

Maternal microchimeric cells

The transfer of maternal cells to offspring during gestation was first identified following observation of offspring with severe combined immunodeficiency (SCID) where maternal cells expand and can result in a graft-versus-host (GVHD) like syndrome 141-

143. However, development of tools including FISH and PCR allowed detection of low- level XX or HLA-discordant maternal cells in cord blood and later multiple tissues of healthy offspring postpartum144-148 Similar to fetal microchimerism, the direct biological effects of maternal microchimeric cells have been investigated with increased levels of maternal cells in patients with neonatal lupus syndrome, dermatomyositis and biliary atresia149-153. The presence of maternal cells across multiple hematopoietic cell lineages has led to hypotheses these cells may contribute to autoimmunity115,138,145,146,154.

However studies have also shown maternal cells directly contribute to insulin production within the pancreatic beta islets of patients with Type I diabetes suggesting a more beneficial role for these cells in disease148.

Use of animal models where mothers are heterozygous for fluorescent transgenes

(GFP) or MHC haplotypes have allowed detection and further characterization of maternal microchimeric cells in mice and non-human primates101,155-157. Interestingly, in mouse models of NIMA-matched heart allograft transplantation, the detection and tissue distribution of microchimeric maternal cells was associated with graft survival 101,157.

Furthermore, optimal levels of microchimerism and tolerance to NIMA-matched

21 allografts require exposure to NIMA both in utero and postnatally through breastfeeding as NIMA-matched grafts were rejected when either exposure was eliminated101,158. This may reflect the necessity for in utero and also oral exposure to soluble-HLA and intact maternal cells present in breast milk159,160. Although these studies are correlative, this provides evidence that the retention of maternal microchimeric cells may drive persistence of NIMA tolerance. However, models allowing in vivo depletion of maternal microchimeric cells are necessary to establish a definitive cause and effect relationship between these two phenomena. Furthermore, while tolerance to NIMA and persistence of maternal microchimeric cells is associated with reduced anti-maternal immunity for human concepti95, the biological benefits of NIMA tolerance for other mammalian species remains undefined.

22 Hypothesis and Specific Aims Each of us is constitutively chimeric bearing genetically foreign cells from our mothers.

In addition women contain multiple sets of chimeric cells from their mother (Fig 6A) and 532 H.S. Gammill and J.L. Nelson later their own offspring (Fig 6B,C). I have fetus acquires maternal cells. Early evidence of maternal cell transfer to the fetus came from the demonstration of transplacental metastases of maternal melanoma (Reynolds 1955; Freed- been fascinatedman et al. 1960). by the idea that these Exchange from mother to fetus may have different immunologic consequences than fetus to mother transfer, as the acquisi- cells cantion of maternal persist cells by contrary the fetus occurs to in a nascent classical immune system. In 1953, Billingham and Medawar conducted experiments in animal models demonstrating the fundamental concept tenantsof ofactive immunology acquisition of tolerance that in the dictatefetus. At this early these date, A they began to speculate as to the relevance of their results to physiologic processes: “Actively acquired tolerance may not be a wholly artificial phenomenon. We are inquiring into the possibility cells shouldthat it may occur be naturally rejected by the accidental by the incorporation host of maternal cells into a foetus during normal development (Billingham B et al. 1953). immune Human system. correlation To of this thisphenomenon end, derives my from mainobserva- tions in transplantation. In 1988, Claas et al., investigated a group of renal transplant patients who had previously received large C numbers of transfusions. In an attempt to find permissible matched focus grafts throu for theseghout individuals, my the investigators dissertation found that there has was preferential nonresponsiveness to their non-inherited maternal Fig. 1. Microchimerism (Mc) in three generations. (A) Proband as antigen (NIMA), as compared with the non-inherited paternal infant (red) exchanges Mc with her mother (blue), resulting in maternal antigen (NIPA). They suggested this tolerance may have been Mc in the infant and fetal Mc in the mother. (B) As an adult, proband (red), been to understand the biological still harboring maternal Mc, experiences pregnancy herself (green) and induced by fetal exposure to maternal cells, resulting in “true Figure 6. Layers of microchimerism acquires new source of fetal Mc. (C) Later, proband (red), child (green), tolerance maintained by chimerism” (Claas et al. 1988). acrossand proband’s multiple mother (blue), allgenerations. with persistent Mc from maternal and/ benefits thatMore recently promote long-term the persistence preservation of fetal microchimerism of or fetal sources. (FMc) in the mother has been recognized. A 1996 study of women Reprinted by permission of the author H. who had given birth to sons found male DNA in 6 of 8 women chromosome detection in lymphocytes in maternal circulation including women who had given birth to their sons decades Gammill: was used in Int J Dev Biol this regard by several, copyright 2010 groups decades ago microchimericpreviously (Bianchi cells et al . in1996). addition Testing for male to DNA how (or male (Walknowska et al. 1969; Schroder et al. 1972; Schroder et al. cells) in a woman with prior pregnancies is the most common 1974). approach used to identify presumptive FMc. This approach is Various techniques for fetal cell enrichment have been devel- these takenimmunological not because of any biological foreign difference incells FMc from a activelyson oped (Herzenberg avert et al rejection. 1979; Bianchi etand al. 1990; promote Cox et al. the than from a daughter, but rather for practical reasons because a 2003). As molecular techniques have advanced, realization of the single test can be employed and many women tested. Persis- goal of achieving noninvasive prenatal diagnosis has become tence of maternal microchimerism (MMc) had been recognized by more likely. In addition, consideration of plasma fetal DNA con- immunologicalthe 1980s, but tolerance studies were limited associated to children with with immune theircentration persistence. as an adjunct to The current screeningtrafficking tests for ofaneuploidy cells during deficiencies (Pollack et al. 1982). Long-term persistence of MMc has also been suggested, as the amount of fetal DNA in maternal in individuals without immune deficiency including in healthy circulation is higher in pregnancies with aneuploid fetuses than gestationadults is was bidescribed-direction in a 1999 alreport such (Maloney that et al. 1999). both Thus, maternalwith euploid fetuses and (Bianchi fetal et al cells. 1997; Lo cross et al. 1999; the Lo et placenta.al. bi-directional trafficking during pregnancy routinely results in the 2007). acquisition of both FMc in the mother and MMc in her offspring. Given the(See Fig.distinct 1.) immunological states of motherFMc kinetics and in normalchild pregnancy I divided my dissertation into This review will summarize current understanding of naturally Prenatal diagnostic investigations have resulted in a body of acquired Mc, focusing on fetal and maternal sources. The fre- literature describing the kinetics of fetal DNA detection in mater- quency with which Mc occurs, the form in which it originates, and nal plasma. The primary markers utilized to evaluate fetal material two logicalrelevant questions. disease associations will be discussed. We will explore in maternal blood and tissues are based on Y-chromosome unanswered questions about Mc that, when better understood, detection. Polymerase chain reaction (PCR) and in situ hybridiza- will enable us to appreciate its functionality and, ultimately, to tion analyses (example in Fig. 2) have been used most commonly, harness its therapeutic potential. with increasing utilization of quantitative PCR approaches (Q- Aim 1: Investigate the biological consequencesPCR). Employing of maternalthese techniques, microchimericit has been shown that FMc, cells on Fetal microchimerism both in cellular form and in free DNA, is quite common, with 70- 80% of maternal plasma or serum specimens positive for male offspringFeasibility. Given of noninvasive previous prenatal reports diagnosis that maternalDNA, andmicrochimeric at least 17% positivity cells in cellular are material associated (Lo et al. with Many early investigations of fetomaternal cell trafficking prima- 1997). Quantitative studies indicate that male DNA accounts for rily sought to develop methods whereby noninvasive prenatal 3-6% of total DNA in maternal plasma (Lo et al. 1998). FMc is diagnosis of aneuploidy and other genetic disorders could be detectable during pregnancy in maternal blood as early as 4-5 toleranceaccomplished. to NIMA, Karyotypic I hypothesized analysis and quinacrine thesestaining for cells Y weeks may gestational drive age expansion(Thomas et al. 1994). of The regulatory concentration of T cells with their specificity in offspring. To test this I employed female mice with heterozygous expression of transgene encoding ubiquitous, cell-surface expression of I-Ab:2W1S and ovalbumin (OVA) (Act-2W mice) that allowed simultaneous quantification of OVA+

23 maternal microchimeric cells and detection of CD4+ cells with defined NIMA specificity.

Combined with commercially available anti-OVA antibodies we were to deplete maternal microchimeric cells in offspring tissues and defitive analysis of the cause and effect relationship between maternal microchimeric cells and NIMA responsive T cells.

To address the teleological benefits that favor the persistence of maternal microchimeric cells we considered the previously published importance of Tregs in maintaining fetal tolerance during pregnancy. Here we hypothesized that if female offspring with enriched

Tregs with specificity to NIMA encountered male mates that expressed antigens with overlap to NIMA (e.g. MHC) these pregnancies would be more protected from complications. To accomplish this we utilized female mice exposed to MHC haplotypes as NIMA to mate to male mice expressing overlapping or irrelevant MHC haplotypes. To probed fetal tolerance during pregnancy I employed both prenatal pathogen (Listeria monocytogenes) and non-infectious (partial maternal Treg ablation) models of pregnancy complication (Manuscript published in Cell 162(3):505-15.)

Aim 2: Examine the requirement for fetal microchimeric cells to maintain expanded memory maternal Tregs with pre-existing fetal specificity. A former student in the lab

(Jared Rowe) had characterized the profound expansion of maternal Tregs with fetal specificity in females mated to transgenic Act-2W male mice. He showed maternal

Tregs expand throughout gestation and persist postpartum to mediate protective memory to fetal antigen upon secondary pregnancy. Given the parallel existence of fetal microchimeric cells I have re-directed tools generated in Aim 1 to determine the requirement for fetal microchimeric cells in maintenance of maternal Treg memory.

Following primary pregnancy with Act-2W males fetal microchimeric cells were depleted

24 with anti-OVA antibody. Both quantitative retention of maternal Tregs with fetal-2W1S specificity and protection during secondary pregnancy with and without fetal microchimeric cell depletion were tested. Similar to Aim 1, fetal tolerance was challenged with infectious and non-infectious probes. (Manuscript in preparation).

25 References

1 Bainbridge, D. R. The evolution of pregnancy. Early Hum Dev 90, 741-745, doi:10.1016/j.earlhumdev.2014.08.013 (2014). 2 Mossman, H. W. Comparative morphogenesis of the fetal membranes and accessory uterine structures. Contrib Embryol 26, 133-247 (1937). 3 Roberts, R. M., Green, J. A. & Schulz, L. C. The evolution of the placenta. Reproduction 152, R179-189, doi:10.1530/REP-16-0325 (2016). 4 Wildman, D. E. et al. Evolution of the mammalian placenta revealed by phylogenetic analysis. Proceedings of the National Academy of Sciences of the United States of America 103, 3203-3208, doi:10.1073/pnas.0511344103 (2006). 5 Hubrecht, A. A. The Trophoblast: A Rejoinder. Science 20, 367-370, doi:10.1126/science.20.507.367 (1904). 6 PrabhuDas, M. et al. Immune mechanisms at the maternal-fetal interface: perspectives and challenges. Nat Immunol 16, 328-334, doi:10.1038/ni.3131 (2015). 7 Zinkernagel, R. M. & Doherty, P. C. MHC-restricted cytotoxic T cells: studies on the biological role of polymorphic major transplantation antigens determining T-cell restriction-specificity, function, and responsiveness. Advances in immunology 27, 51-177 (1979). 8 Klein, J. The major histocompatibility complex of the mouse. Science 203, 516-521 (1979). 9 Spurgin, L. G. & Richardson, D. S. How pathogens drive genetic diversity: MHC, mechanisms and misunderstandings. Proc. Biol. Sci. 277, 979-988 (2010). 10 Owen, R. D. Immunogenetic Consequences of Vascular Anastomoses between Bovine Twins. Science 102, 400-401 (1945). 11 Medawar, P. B. Some immunological and endocrinological problems raised by the evolution of viviparity in vertebrates. Symp. Soc. Exp. Biol. 7, 320-338 (1953). 12 Medawar, P. B. Biological problems of skin surgery. J Int Chir 13, 385-391; includes translations (1953). 13 Billington, W. et al. Placental barrier to maternal cells. Nature 224, 704-706 (1969). 14 Seller, M. J. Lack of porosity of the mouse placenta to maternal cells. Nature 225, 1254-1255 (1970). 15 Erlebacher, A. Immunology of the maternal-fetal interface. Annual review of immunology 31, 387-411, doi:10.1146/annurev-immunol-032712-100003 (2013). 16 Game, D. S. & Lechler, R. I. Pathways of allorecognition: implications for transplantation tolerance. Transpl Immunol 10, 101-108 (2002). 17 Madeja, Z. et al. Paternal MHC expression on mouse trophoblast affects uterine vascularization and fetal growth. Proceedings of the National Academy of Sciences of the United States of America 108, 4012-4017, doi:10.1073/pnas.1005342108 (2011). 18 Zuckermann, F. & Head, J. Expression of MHC antigens on murine trophoblast and their modulation by interferon. Journal of immunology 137, 846-853 (1986). 19 Tagliani, E. & Erlebacher, A. Dendritic cell function at the maternal-fetal interface. Expert Rev Clin Immunol 7, 593-602, doi:10.1586/eci.11.52 (2011).

26 20 Tagliani, E. et al. Coordinate regulation of tissue macrophage and dendritic cell population dynamics by CSF-1. The Journal of experimental medicine 208, 1901- 1916, doi:10.1084/jem.20110866 (2011). 21 Collins, M. K., Tay, C. S. & Erlebacher, A. Dendritic cell entrapment within the pregnant uterus inhibits immune surveillance of the maternal/fetal interface in mice. The Journal of clinical investigation 119, 2062-2073, doi:10.1172/JCI38714 (2009). 22 Volchek, M. et al. Lymphatics in the human endometrium disappear during decidualization. Hum Reprod 25, 2455-2464, doi:10.1093/humrep/deq224 (2010). 23 Kamimura, S., Eguchi, K., Yonezawa, M. & Sekiba, K. Localization and developmental change of indoleamine 2,3-dioxygenase activity in the human placenta. Acta Med Okayama 45, 135-139 (1991). 24 Frumento, G. et al. Tryptophan-derived catabolites are responsible for inhibition of T and natural killer cell proliferation induced by indoleamine 2,3-dioxygenase. The Journal of experimental medicine 196, 459-468 (2002). 25 Terness, P. et al. Inhibition of allogeneic T cell proliferation by indoleamine 2,3- dioxygenase-expressing dendritic cells: mediation of suppression by tryptophan metabolites. The Journal of experimental medicine 196, 447-457 (2002). 26 Munn, D. et al. Prevention of allogeneic fetal rejection by tryptophan catabolism. Science 281, 1191-1193 (1998). 27 Simpson, H., Robson, S. C., Bulmer, J. N., Barber, A. & Lyall, F. Transforming growth factor beta expression in human placenta and placental bed during early pregnancy. Placenta 23, 44-58, doi:10.1053/plac.2001.0746 (2002). 28 von Wolff, M., Wang, X., Gabius, H. J. & Strowitzki, T. Galectin fingerprinting in human endometrium and decidua during the menstrual cycle and in early gestation. Mol Hum Reprod 11, 189-194, doi:10.1093/molehr/gah144 (2005). 29 Phillips, B. et al. Differential expression of two beta-galactoside-binding lectins in the reproductive tracts of pregnant mice. Biology of reproduction 55, 548-558 (1996). 30 Eriksson, M., Meadows, S. K., Wira, C. R. & Sentman, C. L. Unique phenotype of human uterine NK cells and their regulation by endogenous TGF-beta. J Leukoc Biol 76, 667-675, doi:10.1189/jlb.0204090 (2004). 31 Perrier d'Hauterive, S. et al. Human endometrial leukemia inhibitory factor and interleukin-6: control of secretion by transforming growth factor-beta-related members. Neuroimmunomodulation 12, 157-163, doi:10.1159/000084848 (2005). 32 Cheng, S. K., Olale, F., Brivanlou, A. H. & Schier, A. F. Lefty blocks a subset of TGFbeta signals by antagonizing EGF-CFC coreceptors. PLoS biology 2, E30, doi:10.1371/journal.pbio.0020030 (2004). 33 Tang, M., Taylor, H. S. & Tabibzadeh, S. In vivo gene transfer of lefty leads to implantation failure in mice. Hum Reprod 20, 1772-1778, doi:10.1093/humrep/deh849 (2005). 34 Maquoi, E., van den Brule, F. A., Castronovo, V. & Foidart, J. M. Changes in the distribution pattern of galectin-1 and galectin-3 in human placenta correlates with the differentiation pathways of trophoblasts. Placenta 18, 433-439 (1997).

27 35 Ilarregui, J. M. et al. Tolerogenic signals delivered by dendritic cells to T cells through a galectin-1-driven immunoregulatory circuit involving interleukin 27 and interleukin 10. Nat Immunol 10, 981-991, doi:10.1038/ni.1772 (2009). 36 Chung, C. D., Patel, V. P., Moran, M., Lewis, L. A. & Miceli, M. C. Galectin-1 induces partial TCR zeta-chain phosphorylation and antagonizes processive TCR signal transduction. Journal of immunology 165, 3722-3729 (2000). 37 Bozic, M. et al. Galectin-1 and galectin-3 in the trophoblast of the gestational trophoblastic disease. Placenta 25, 797-802, doi:10.1016/j.placenta.2004.03.006 (2004). 38 Blois, S. et al. A pivotal role for galectin-1 in fetomaternal tolerance. Nature Medicine 13, 1450-1457 (2007). 39 Confavreux, C., Hutchinson, M., Hours, M. M., Cortinovis-Tourniaire, P. & Moreau, T. Rate of pregnancy-related relapse in multiple sclerosis. Pregnancy in Multiple Sclerosis Group. The New England journal of medicine 339, 285-291, doi:10.1056/NEJM199807303390501 (1998). 40 Hellwig, K., Haghikia, A., Rockhoff, M. & Gold, R. Multiple sclerosis and pregnancy: experience from a nationwide database in Germany. Ther Adv Neurol Disord 5, 247- 253, doi:10.1177/1756285612453192 (2012). 41 Patas, K., Engler, J. B., Friese, M. A. & Gold, S. M. Pregnancy and multiple sclerosis: feto-maternal immune cross talk and its implications for disease activity. J Reprod Immunol 97, 140-146, doi:10.1016/j.jri.2012.10.005 (2013). 42 Ostensen, M. & Villiger, P. M. Immunology of pregnancy-pregnancy as a remission inducing agent in rheumatoid arthritis. Transpl Immunol 9, 155-160 (2002). 43 Ostensen, M. & Villiger, P. M. The remission of rheumatoid arthritis during pregnancy. Semin Immunopathol 29, 185-191, doi:10.1007/s00281-007-0072-5 (2007). 44 Gatson, N. N. et al. Induction of pregnancy during established EAE halts progression of CNS autoimmune injury via pregnancy-specific serum factors. J Neuroimmunol 230, 105-113, doi:10.1016/j.jneuroim.2010.09.010 (2011). 45 Langer-Gould, A., Garren, H., Slansky, A., Ruiz, P. J. & Steinman, L. Late pregnancy suppresses relapses in experimental autoimmune encephalomyelitis: evidence for a suppressive pregnancy-related serum factor. Journal of immunology 169, 1084- 1091 (2002). 46 McClain, M. A. et al. Pregnancy suppresses experimental autoimmune encephalomyelitis through immunoregulatory cytokine production. Journal of immunology 179, 8146-8152 (2007). 47 Munoz-Suano, A., Kallikourdis, M., Sarris, M. & Betz, A. G. Regulatory T cells protect from autoimmune arthritis during pregnancy. Journal of autoimmunity 38, J103-108, doi:10.1016/j.jaut.2011.09.007 (2012). 48 Finkelsztejn, A., Brooks, J. B., Paschoal, F. M., Jr. & Fragoso, Y. D. What can we really tell women with multiple sclerosis regarding pregnancy? A systematic review and meta-analysis of the literature. BJOG 118, 790-797, doi:10.1111/j.1471- 0528.2011.02931.x (2011). 49 Fontenot, J. D., Gavin, M. A. & Rudensky, A. Y. Foxp3 programs the development and function of CD4+CD25+ regulatory T cells. Nat Immunol 4, 330-336, doi:10.1038/ni904 (2003).

28 50 Sakaguchi, S., Sakaguchi, N., Asano, M., Itoh, M. & Toda, M. 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 (1995). 51 Hori, S., Nomura, T. & Sakaguchi, S. Control of regulatory T cell development by the transcription factor FoxP3. Science 299, 1057-1061 (2003). 52 Apostolou, I., Sarukhan, A., Klein, L. & von Boehmer, H. Origin of regulatory T cells with known specificity for antigen. Nat Immunol 3, 756-763, doi:10.1038/ni816 (2002). 53 Hsieh, C. S., Lee, H. M. & Lio, C. W. Selection of regulatory T cells in the thymus. Nature reviews. Immunology 12, 157-167, doi:10.1038/nri3155 (2012). 54 Jordan, M. S. et al. Thymic selection of CD4+CD25+ regulatory T cells induced by an agonist self-peptide. Nat Immunol 2, 301-306, doi:10.1038/86302 (2001). 55 Bennett, C., Christie, J., Ramsdell, F., Brunkow, M. & Ferguson, P. The immune dysregulation polyendicrinopath, X-linked syndrome (IPEX) is caused by mutations of FOXP3. Nature Genetics 27, 20-21 (2001). 56 Wildin, R., Ramsdell, F., Peake, J., Faravelli, F. & Casanova, J. X-linked neonatal diabetes mellitus, enteropathy and endocrinopathy syndrome is the human equivalent of mouse scurfy. Nature Genetics 27, 18-20 (2001). 57 Khattri, R., Cox, T., Yasayko, S. & Ramsdell, F. An essential role for Scurfin in CD4+CD25+ T regulatory cells. Nature Immunology 4 (2003). 58 Kim, J. M., Rasmussen, J. P. & Rudensky, A. Y. Regulatory T cells prevent catastrophic autoimmunity throughout the lifespan of mice. Nat Immunol 8, 191-197 (2007). 59 Godfrey, V. L., Wilkinson, J. E., Rinchik, E. M. & Russell, L. B. Fatal lymphoreticular disease in the scurfy (sf) mouse requires T cells that mature in a sf thymic environment: potential model for thymic education. Proceedings of the National Academy of Sciences of the United States of America 88, 5528-5532 (1991). 60 Chen, W. et al. 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, doi:10.1084/jem.20030152 (2003). 61 Tone, Y. et al. Smad3 and NFAT cooperate to induce Foxp3 expression through its enhancer. Nat Immunol 9, 194-202, doi:10.1038/ni1549 (2008). 62 Zheng, Y. et al. Role of conserved non-coding DNA elements in the Foxp3 gene in regulatory T-cell fate. Nature 463, 808-812, doi:10.1038/nature08750 (2010). 63 Josefowicz, S. Z. et al. Extrathymically generated regulatory T cells control mucosal TH2 inflammation. Nature 482, 395-399, doi:10.1038/nature10772 (2012). 64 Heikkinen, J., Mottonen, M., Alanen, A. & Lassila, O. Phenotypic characterization of regulatory T cells in the human decidua. Clinical and experimental immunology 136, 373-378, doi:10.1111/j.1365-2249.2004.02441.x (2004). 65 Jiang, T. T. et al. Regulatory T cells: new keys for further unlocking the enigma of fetal tolerance and pregnancy complications. Journal of immunology 192, 4949- 4956, doi:10.4049/jimmunol.1400498 (2014). 66 Somerset, D., Zheng, Y., Kilby, M., Sansom, D. & Drayson, M. Normal human pregnancy is associated with an elevation in the immune suppressive CD25+ CD4+ regulatory T-cell subset. Immunology 112, 38-43 (2004).

29 67 Aluvihare, V., Kallikourdis, M. & Betz, A. Regulatory T cells mediate maternal tolerance to the fetus. Nat Immunol 5, 266-271 (2004). 68 Rowe, J. H., Ertelt, J. M., Xin, L. & Way, S. S. Pregnancy imprints regulatory memory that sustains anergy to fetal antigen. Nature 490, 102-106 (2012). 69 Shima, T. et al. Regulatory T cells are necessary for implantation and maintenance of early pregnancy but not late pregnancy in allogeneic mice. Journal of Reproductive Immunology 85, 121-129 (2010). 70 Thuere, C. et al. Kinetics of regulatory T cells during murine pregnancy. Am J Reprod Immunol 58, 514-523, doi:10.1111/j.1600-0897.2007.00538.x (2007). 71 Bonney, E. A. & Brown, S. A. To drive or be driven: the path of a mouse model of recurrent pregnancy loss. Reproduction 147, R153-167, doi:10.1530/REP-13-0583 (2014). 72 Zenclussen, A. C. et al. Regulatory T cells induce a privileged tolerant microenvironment at the fetal-maternal interface. European journal of immunology 36, 82-94, doi:10.1002/eji.200535428 (2006). 73 Zenclussen, A. C. et al. Abnormal T-cell reactivity against paternal antigens in spontaneous abortion: adoptive transfer of pregnancy-induced CD4+CD25+ T regulatory cells prevents fetal rejection in a murine abortion model. The American journal of pathology 166, 811-822, doi:10.1016/S0002-9440(10)62302-4 (2005). 74 Darrasse-Jeze, G., Klatzmann, D., Charlotte, F., Salomon, B. L. & Cohen, J. L. CD4+CD25+ regulatory/suppressor T cells prevent allogeneic fetus rejection in mice. Immunol Lett 102, 106-109, doi:10.1016/j.imlet.2005.07.002 (2006). 75 Rowe, J. H., Ertelt, J. M., Aguilera, M. N., Farrar, M. A. & Way, S. S. Foxp3(+) regulatory T cell expansion required for sustaining pregnancy compromises host defense against prenatal bacterial pathogens. Cell Host Microbe 10, 54-64 (2011). 76 Sasaki, Y. et al. Proportion of peripheral blood and decidual CD4+ CD25(bright) regulatory T cells in pre-eclampsia. Clinical and experimental immunology 149, 139- 145 (2007). 77 Steegers, E. A., von Dadelszen, P., Duvekot, J. J. & Pijnenborg, R. Pre-eclampsia. Lancet 376, 631-644, doi:10.1016/S0140-6736(10)60279-6 (2010). 78 Toldi, G. et al. Decreased number of FoxP3+ regulatory T cells in preeclampsia. Acta obstetricia et gynecologica Scandinavica 87, 1229-1233 (2008). 79 Sasaki, Y. et al. Decidual and peripheral blood CD4+CD25+ regulatory T cells in early pregnancy subjects and spontaneous abortion cases. Mol Hu Reprod 10, 347-353 (2004). 80 Jin, L. P., Chen, Q. Y., Zhang, T., Guo, P. F. & Li, D. J. The CD4+CD25 bright regulatory T cells and CTLA-4 expression in peripheral and decidual lymphocytes are down- regulated in human miscarriage. Clin Immunol 133, 402-410, doi:10.1016/j.clim.2009.08.009 (2009). 81 Bao, S. H. et al. Decidual CD4+CD25+CD127dim/- regulatory T cells in patients with unexplained recurrent spontaneous miscarriage. Eur J Obstet Gynecol Reprod Biol 155, 94-98, doi:10.1016/j.ejogrb.2010.11.007 (2011). 82 Schober, L. et al. Term and preterm labor: decreased suppressive activity and changes in composition of the regulatory T-cell pool. Immunology and cell biology 90, 935-944, doi:10.1038/icb.2012.33 (2012).

30 83 Levine, A. G., Arvey, A., Jin, W. & Rudensky, A. Y. Continuous requirement for the TCR in regulatory T cell function. Nat Immunol 15, 1070-1078, doi:10.1038/ni.3004 (2014). 84 Moon, J. et al. Naive CD4(+) T cell frequency varies for different epitopes and predicts repertoire diversity and response magnitude. Immunity 27, 203-213 (2007). 85 Moon, J. J. et al. Tracking epitope-specific T cells. Nat. Protoc. 4, 565-581 (2009). 86 Moon, J. J. et al. Quantitative impact of thymic selection on Foxp3+ and Foxp3- subsets of self-peptide/MHC class II-specific CD4+ T cells. Proc. Natl. Acad. Sci. U.S.A. 108, 14602-14607 (2011). 87 Erlebacher, A., Vencato, D., Price, K., Zhang, D. & Glimcher, L. Constraints in antigen presentation severely restric T cell recognition of allogeneic fetus. Journal of Clinical Investigation 117, 1399-1411 (2007). 88 Samstein, R. M., Josefowicz, S. Z., Arvey, A., Treuting, P. M. & Rudensky, A. Y. Extrathymic generation of regulatory T cells in placental mammals mitigates maternal-fetal conflict. Cell 150, 29-38, doi:10.1016/j.cell.2012.05.031 (2012). 89 Campbell, D., MacGillivray, I. & Carr-Hill, R. Pre-eclampsia in second pregnancy. Br J Obstet Gynaecol 92, 131-140 (1985). 90 Li, D. K. & Wi, S. Changing paternity and the risk of preeclampsia/eclampsia in the subsequent pregnancy. American journal of epidemiology 151, 57-62 (2000). 91 Belkaid, Y., Piccirillo, C. A., Mendez, S., Shevach, E. M. & Sacks, D. L. CD4+CD25+ regulatory T cells control Leishmania major persistence and immunity. Nature 420, 502-507, doi:10.1038/nature01152 (2002). 92 Homann, D., Teyton, L. & Oldstone, M. Differential regulation of antiviral T-cells immunity results stable CD8+ by declining CD4+ T-cell memory. Nat Med 7, 913-919 (2001). 93 Nelson, R. W., McLachlan, J. B., Kurtz, J. R. & Jenkins, M. K. CD4+ T cell persistence and function after infection are maintained by low-level peptide:MHC class II presentation. Journal of immunology 190, 2828-2834 (2013). 94 Uzonna, J. E., Wei, G., Yurkowski, D. & Bretscher, P. Immune elimination of Leishmania major in mice: implications for immune memory, vaccination, and reactivation disease. Journal of immunology 167, 6967-6974 (2001). 95 Mold, J. E. et al. Maternal alloantigens promote the development of tolerogenic fetal regulatory T cells in utero. Science 322, 1562-1565 (2008). 96 Burlingham, W. J. et al. The effect of tolerance to noninherited maternal HLA antigens on the survival of renal transplants from sibling donors. N. Engl. J. Med. 339, 1657-1664 (1998). 97 Ichinohe, T. et al. Feasibility of HLA-haploidentical hematopoietic stem cell transplantation between noninherited maternal antigen (NIMA)-mismatched family members linked with long-term fetomaternal microchimerism. Blood 104, 3821- 3828 (2004). 98 van Rood, J. J. et al. Effect of tolerance to noninherited maternal antigens on the occurrence of graft-versus-host disease after bone marrow transplantation from a parent or an HLA-haploidentical sibling. Blood 99, 1572-1577 (2002).

31 99 Owen, R. D., Wood, H. R., Foord, A. G., Sturgeon, P. & Baldwin, L. G. Evidence for actively acquired tolerance to Rh antigens. Proc. Natl. Acad. Sci. U.S.A. 40, 420-424 (1954). 100 Claas, F. H., Gijbels, Y., van der Velden-de Munck, J. & van Rood, J. J. Induction of B cell unresponsiveness to noninherited maternal HLA antigens during fetal life. Science 241, 1815-1817 (1988). 101 Andrassy, J. et al. Tolerance to noninherited maternal MHC antigens in mice. Journal of immunology 171, 5554-5561 (2003). 102 Matsuoka, K. et al. Fetal tolerance to maternal antigens improves the outcome of allogeneic bone marrow transplantation by a CD4+ CD25+ T-cell-dependent mechanism. Blood 107, 404-409, doi:10.1182/blood-2005-07-3045 (2006). 103 Akiyama, Y. et al. Transplantation tolerance to a single noninherited MHC class I maternal alloantigen studied in a TCR-transgenic mouse model. Journal of immunology 186, 1442-1449, doi:10.4049/jimmunol.1003023 (2011). 104 Araki, M. et al. Prediction of reactivity to noninherited maternal antigen in MHC- mismatched, minor histocompatibility antigen-matched stem cell transplantation in a mouse model. Journal of immunology 185, 7739-7745, doi:10.4049/jimmunol.1001226 (2010). 105 Mold, J. E. & McCune, J. M. Immunological tolerance during fetal development: from mouse to man. Adv. Immunol. 115, 73-111 (2012). 106 Schroder, J. & De la Chapelle, A. Fetal lymphocytes in the maternal blood. Blood 39, 153-162 (1972). 107 Schroder, J., Tiilikainen, A. & De la Chapelle, A. Fetal leukocytes in the maternal circulation after delivery. I. Cytological aspects. Transplantation 17, 346-354 (1974). 108 Tiilikainen, A., Schroder, J. & De la Chapelle, A. Fetal leukocytes in the maternal circulation after delivery. II. Masking of HL-A antigens. Transplantation 17, 355-360 (1974). 109 Walknowska, J., Conte, F. A. & Grumbach, M. M. Practical and theoretical implications of fetal-maternal lymphocyte transfer. Lancet 1, 1119-1122 (1969). 110 Bianchi, D. W., Flint, A. F., Pizzimenti, M. F., Knoll, J. H. & Latt, S. A. Isolation of fetal DNA from nucleated erythrocytes in maternal blood. Proceedings of the National Academy of Sciences of the United States of America 87, 3279-3283 (1990). 111 Ariga, H. et al. Kinetics of fetal cellular and cell-free DNA in the maternal circulation during and after pregnancy: implications for noninvasive prenatal diagnosis. Transfusion 41, 1524-1530 (2001). 112 Thomas, M. R., Williamson, R., Craft, I. & Rodeck, C. H. The time of appearance, and quantitation, of fetal DNA in the maternal circulation. Ann N Y Acad Sci 731, 217-225 (1994). 113 Bianchi, D., Zickwolf, G., Weil, G., Sylvester, S. & DeMaria, M. Male fetal progenitor cells persist in maternal blood for as long as 27 years postpartum. PNAS 93, 705- 708 (1996). 114 Khosrotehrani, K., Johnson, K. L., Cha, D. H., Salomon, R. N. & Bianchi, D. W. Transfer of fetal cells with multilineage potential to maternal tissue. JAMA : the journal of the American Medical Association 292, 75-80, doi:10.1001/jama.292.1.75 (2004). 115 Nelson, J. L. The otherness of self: microchimerism in health and disease. Trends Immunol. 33, 421-427 (2012).

32 116 Stevens, A. M. et al. Liver biopsies from human females contain male hepatocytes in the absence of transplantation. Lab Invest 84, 1603-1609, doi:10.1038/labinvest.3700193 (2004). 117 Gadi, V. K. Fetal microchimerism in breast from women with and without breast cancer. Breast Cancer Res Treat 121, 241-244, doi:10.1007/s10549-009-0548-1 (2010). 118 Artlett, C. M. et al. Increased microchimeric CD4+ T lymphocytes in peripheral blood from women with systemic sclerosis. Clin Immunol 103, 303-308 (2002). 119 Evans, P. C. et al. Long-term fetal microchimerism in peripheral blood mononuclear cell subsets in healthy women and women with scleroderma. Blood 93, 2033-2037 (1999). 120 Guetta, E., Gordon, D., Simchen, M. J., Goldman, B. & Barkai, G. Hematopoietic progenitor cells as targets for non-invasive prenatal diagnosis: detection of fetal CD34+ cells and assessment of post-delivery persistence in the maternal circulation. Blood Cells Mol Dis 30, 13-21 (2003). 121 O'Donoghue, K. et al. Identification of fetal mesenchymal stem cells in maternal blood: implications for non-invasive prenatal diagnosis. Mol Hum Reprod 9, 497-502 (2003). 122 Osada, H. et al. Detection of fetal HPCs in maternal circulation after delivery. Transfusion 41, 499-503 (2001). 123 Adams, K. M. et al. Male DNA in female donor apheresis and CD34-enriched products. Blood 102, 3845-3847, doi:10.1182/blood-2003-05-1570 (2003). 124 Bianchi, D. W., Zickwolf, G. K., Weil, G. J., Sylvester, S. & DeMaria, M. A. Male fetal progenitor cells persist in maternal blood for as long as 27 years postpartum. Proceedings of the National Academy of Sciences of the United States of America 93, 705-708 (1996). 125 Jimenez, D. F., Leapley, A. C., Lee, C. I., Ultsch, M. N. & Tarantal, A. F. Fetal CD34+ cells in the maternal circulation and long-term microchimerism in rhesus monkeys (Macaca mulatta). Transplantation 79, 142-146 (2005). 126 Jimenez, D. F. & Tarantal, A. F. Quantitative analysis of male fetal DNA in maternal serum of gravid rhesus monkeys (Macaca mulatta). Pediatric research 53, 18-23, doi:10.1203/00006450-200301000-00007 (2003). 127 Khosrotehrani, K., Johnson, K. L., Guegan, S., Stroh, H. & Bianchi, D. W. Natural history of fetal cell microchimerism during and following murine pregnancy. J Reprod Immunol 66, 1-12, doi:10.1016/j.jri.2005.02.001 (2005). 128 Karniychuk, U. U., Van Breedam, W., Van Roy, N., Rogel-Gaillard, C. & Nauwynck, H. J. Demonstration of microchimerism in pregnant sows and effects of congenital PRRSV infection. Vet Res 43, 19, doi:10.1186/1297-9716-43-19 (2012). 129 Liegeois, A., Gaillard, M., Ouvre, E. & Lewin, D. Microchimerism in pregnant mice. Transplant Proc 13, 1250-1252 (1981). 130 Turin, L. et al. Bovine fetal microchimerism in normal and embryo transfer pregnancies and its implications for biotechnology applications in cattle. Biotechnol J 2, 486-491, doi:10.1002/biot.200600218 (2007). 131 Wang, Y. et al. Fetal cells in mother rats contribute to the remodeling of liver and kidney after injury. Biochemical and biophysical research communications 325, 961- 967, doi:10.1016/j.bbrc.2004.10.105 (2004).

33 132 Kremer Hovinga, I. C. et al. Chimerism occurs twice as often in lupus nephritis as in normal kidneys. Arthritis and rheumatism 54, 2944-2950, doi:10.1002/art.22038 (2006). 133 Mosca, M. et al. Correlations of Y chromosome microchimerism with disease activity in patients with SLE: analysis of preliminary data. Annals of the rheumatic diseases 62, 651-654 (2003). 134 Nelson, J. L. Microchimerism and the pathogenesis of systemic sclerosis. Curr Opin Rheumatol 10, 564-571 (1998). 135 Lambert, N. C. et al. Male microchimerism in women with systemic sclerosis and healthy women who have never given birth to a son. Annals of the rheumatic diseases 64, 845-848, doi:10.1136/ard.2004.029314 (2005). 136 Johnson, K. L. et al. Fetal cell microchimerism in tissue from multiple sites in women with systemic sclerosis. Arthritis and rheumatism 44, 1848-1854, doi:10.1002/1529-0131(200108)44:8<1848::AID-ART323>3.0.CO;2-L (2001). 137 Srivatsa, B. et al. Microchimerism of presumed fetal origin in thyroid specimens from women: a case-control study. Lancet 358, 2034-2038, doi:10.1016/S0140- 6736(01)07099-4 (2001). 138 Gammill, H. & Nelson, J. Naturally acquired microchimerism. Int J Dev Biol 54, 531- 543 (2010). 139 Adams, K. M. & Nelson, J. L. Microchimerism: an investigative frontier in autoimmunity and transplantation. JAMA : the journal of the American Medical Association 291, 1127-1131, doi:10.1001/jama.291.9.1127 (2004). 140 Eikmans, M. et al. Naturally acquired microchimerism: implications for transplantation outcome and novel methodologies for detection. Chimerism 5, 24-39 (2014). 141 Geha, R. S. & Reinherz, E. Identification of circulating maternal T and B lymphocytes in uncomplicated severe combined immunodeficiency by HLA typing of subpopulations of T cells separated by the fluorescence-activated cell sorter and of Epstein Barr virus-derived B cell lines. Journal of immunology 130, 2493-2495 (1983). 142 Kadowaki, J. et al. XX-XY lymphoid chimaerism in congenital immunological deficiency syndrome with thymic alymphoplasia. Lancet 2, 1152-1156 (1965). 143 Pollack, M. S. et al. DR-positive maternal engrafted T cells in a severe combined immunodeficiency patient without graft-versus-host disease. Transplantation 30, 331-334 (1980). 144 Hall, J. M. et al. Detection of maternal cells in human umbilical cord blood using fluorescence in situ hybridization. Blood 86, 2829-2832 (1995). 145 Jonsson, A. M., Uzunel, M., Gotherstrom, C., Papadogiannakis, N. & Westgren, M. Maternal microchimerism in human fetal tissues. American journal of obstetrics and gynecology 198, 325 e321-326, doi:10.1016/j.ajog.2007.09.047 (2008). 146 Loubiere, L. S. et al. Maternal microchimerism in healthy adults in lymphocytes, monocyte/macrophages and NK cells. Lab Invest 86, 1185-1192, doi:10.1038/labinvest.3700471 (2006). 147 Maloney, S. et al. Microchimerism of maternal origin persists into adult life. The Journal of clinical investigation 104, 41-47, doi:10.1172/JCI6611 (1999).

34 148 Nelson, J. L. et al. Maternal microchimerism in peripheral blood in type 1 diabetes and pancreatic islet beta cell microchimerism. PNAS 104, 1637-1642, doi:10.1073/pnas.0606169104 (2007). 149 Artlett, C. M. et al. Chimeric cells of maternal origin in juvenile idiopathic inflammatory myopathies. Childhood Myositis Heterogeneity Collaborative Group. Lancet 356, 2155-2156 (2000). 150 Lambert, N. C. et al. Quantification of maternal microchimerism by HLA-specific real- time polymerase chain reaction: studies of healthy women and women with scleroderma. Arthritis and rheumatism 50, 906-914, doi:10.1002/art.20200 (2004). 151 Reed, A. M., Picornell, Y. J., Harwood, A. & Kredich, D. W. Chimerism in children with juvenile dermatomyositis. Lancet 356, 2156-2157, doi:10.1016/S0140- 6736(00)03500-5 (2000). 152 Stevens, A. M., Hermes, H. M., Rutledge, J. C., Buyon, J. P. & Nelson, J. L. Myocardial- tissue-specific phenotype of maternal microchimerism in neonatal lupus congenital heart block. Lancet 362, 1617-1623, doi:10.1016/S0140-6736(03)14795-2 (2003). 153 Suskind, D. L. et al. Maternal microchimerism in the livers of patients with biliary atresia. BMC gastroenterology 4, 14, doi:10.1186/1471-230X-4-14 (2004). 154 Gotherstrom, C., Johnsson, A. M., Mattsson, J., Papadogiannakis, N. & Westgren, M. Identification of maternal hematopoietic cells in a 2nd-trimester fetus. Fetal diagnosis and therapy 20, 355-358, doi:10.1159/000086812 (2005). 155 Bakkour, S. et al. Analysis of maternal microchimerism in rhesus monkeys (Macaca mulatta) using real-time quantitative PCR amplification of MHC polymorphisms. Chimerism 5, 6-15 (2014). 156 Dutta, P. & Burlingham, W. J. Stem cell microchimerism and tolerance to noninherited maternal antigens. Chimerism 1, 2-10, doi:10.4161/chim.1.1.12667 (2010). 157 Dutta, P. et al. Microchimerism is strongly correlated with tolerance to noninherited maternal antigens in mice. Blood 114, 3578-3587 (2009). 158 Campbell, D. A., Jr. et al. Breast feeding and maternal-donor renal allografts. Possibly the original donor-specific transfusion. Transplantation 37, 340-344 (1984). 159 Molitor, M. L., Haynes, L. D., Jankowska-Gan, E., Mulder, A. & Burlingham, W. J. HLA class I noninherited maternal antigens in cord blood and breast milk. Hum. Immunol. 65, 231-239 (2004). 160 Zhou, L. et al. Two independent pathways of maternal cell transmission to offspring: through placenta during pregnancy and by breast-feeding after birth. Immunology 101, 570-580 (2000).

Chapter 2: Pregnancy-induced maternal regulatory T cells, bona fide memory or maintenance by antigenic reminder from fetal cell microchimerism?

* This is an open-access article reprinted here under the terms of the Creative Commons Attribution Non-Commercial License. The manuscript was originally published 2014. Kinder JM. 2014. Chimerism 5(1):16-9

Long-term maintenance of immune components with defined specificity, without antigen is the hallmark feature of immunological memory. However, there are fundamental differences in how memory CD8+ compared with CD4+ T cells are maintained. After complete antigen elimination, CD8+ T cells can persist as a self- renewing numerically stable cell population, and therefore satisfy the most stringent definition of “memory”. Comparatively, CD4+ T cell maintenance is considerably less stable, often requiring low-level antigen persistence or antigenic reminders. Recent studies show these basic memory features, classically ascribed to effector CD8+ and CD4+ T cells, extend to immune suppressive Foxp3+ regulatory CD4+ T cells (Tregs). In particular, gestational expansion and postpartum retention of maternal Tregs with fetal specificity may explain the protective benefits of primary pregnancy on complications in subsequent pregnancy. Herein, the possibility of ongoing antigenic reminders from fetal cell microchimerism in postpartum maintenance of maternal Tregs with fetal specificity is considered.

The mammalian immune system is endowed not only with efficient self, non-self discrimination, but also the ability to “remember” antigenic encounters. For immunologically foreign antigens, prior stimulation has the potential to prime long- term retention of “memory” immune cells with specificity to the inciting antigen. In turn, establishing the molecular and cellular requirements for immunological memory has critical implications for developing more durable vaccines and other immune modulatory therapies.

Emerging studies highlight an interesting discordance in necessity for antigen persistence in maintaining long-term retention of CD8+ compared with CD4+ T cells with defined specificity.1-7 This is best illustrated by the dynamics of pathogen- specific CD8+ and CD4+ T cells after infection with viruses or other intracellular pathogens that do not cause persistence. While both T cell subsets expand robustly during acute infection, a numerically stable self-renewing pool of pathogen-specific

CD8+ T cells is maintained indefinitely despite complete antigen elimination. By contrast, CD4+ T cells responding to the same acute infection undergo protracted, but stable contraction with an estimated half-life of 15 to 40 days.8-10 This discordance may reflect the necessity for each T cell subset in host defense. For acute infection with viruses or other intra-cytoplasmic pathogens (e.g. influenza A, lymphocytic choriomeningitis virus, or Listeria monocytogenes) where protection is conferred by CD8+ T cells, these cells are chosen for selective retention.

Comparatively for pathogens that primarily cause persistent infection and reside within the phagocytic vacuole of infected cells thereby escaping detection or elimination by CD8+ T cells (e.g. Mycobacterium tuberculosis, Leishmania major, or

Salmonella spp.), pathogen-specific CD4+ T cells play a more dominant protective role.11-13 Importantly however, while CD8+ T cell mediated protection against secondary infection is maintained well after antigen elimination, protection by retained memory CD4+ T cells requires low-level antigen persistence. Accordingly for pathogens capable of establishing persistent infection, antigen elimination that occurs naturally or with adjunctive antimicrobials, accelerates contraction of

pathogen-specific CD4+ T cells and overrides the protective benefits of prior infection.14-16 Therefore, unlike CD8+ T cells, the long-term maintenance of CD4+ T cell memory appear to require more frequent, if not constant, antigenic reminders.

While the memory features of CD4+ T cells has been best characterized for IFN-γ producing Th1 cells, other CD4+ effector lineages (e.g. Th2 or Th17 cells) appear to share a similar potential for long-term retention.7, 10 By redirecting tools for tracking antigen-specific T cells, we and others have recently shown these memory features classically described for effector T cells also extend to immune suppressive regulatory CD4+ T cells (Tregs) identified by Foxp3 expression. Treg memory was first shown using transgenic mice where the model antigen, ovalbumin, could be inducibly expressed within the skin.17 Primary dermal stimulation with this surrogate self-antigen primed expansion and retention of ovalbumin specific Tregs that dampens the severity of localized autoimmune reactions when this antigen was re- expressed ~30 days later. Likewise, a complementary study tracking Tregs after influenza A infection showed accelerated accumulation of virus-specific Tregs after secondary, compared with primary infection, which may be important for limiting pathological airway inflammation from over-exuberant effector CD8+ T cell activation.18 Our own studies investigating maternal Tregs with specificity to the

b immune-dominant I-A :2W1S55-68 peptide expressed as a surrogate fetal antigen during allogeneic pregnancy, showed Foxp3+ CD4+ T cells with this specificity progressively expand throughout gestation.19 Interestingly after delivery of the fetus and other gross products of conception, maternal Tregs with fetal specificity were

maintained at markedly enriched levels; and these cells re-expand with accelerated tempo during secondary pregnancy upon encounter with the same paternal-fetal antigen. Considering the necessity for expanded maternal Tregs in maintaining fetal tolerance during pregnancy,19-24 these findings likely provide critical mechanistic insights for how primary pregnancy protects against complications stemming from fractured fetal tolerance in subsequent pregnancy.19, 20, 25, 26 In turn, applied to the basic biology of CD4+ T cells, these findings together establish Foxp3+ Tregs, like effector T cells, can persist as memory immune cells.

Given the discordance in necessity for antigen persistence in sustaining long-term retention of CD8+ compared with CD4+ effector T cells with defined specificity, these findings also open up exciting new questions regarding whether retained Tregs reflects bona fide memory or maintenance in response to antigen persistence. In the case of Tregs with specificity for surrogate-self ovalbumin antigen within the skin, ongoing stimulation is unlikely since naive ovalbumin specific T cells failed to proliferate after adoptive transfer without induced antigen expression.17 Similarly,

Tregs retained after influenza A infection are unlikely to reflect stimulation from residual viral antigen, since this pathogen is not known to cause persistent infection.18 However in each of these models, the longer-term durability of Tregs, with specificity to either self or pathogen, remain undefined since the impacts of secondary antigen challenge were reported at most ~35 days after silencing primary antigen stimulation.17, 18 In our studies tracking maternal Tregs with surrogate fetal-

2W1S specificity, enriched cells were maintained through at least 100 days

postpartum despite progressively diminishing cell numbers.19 In particular, the postpartum decay kinetics of maternal Tregs with fetal specificity (estimated t1/2 of 25 days) show striking similarity with effector CD4+ T cells primed by acute infection.

On the other hand and in sharp contrast to the tempo of antigen stimulation that occurs after acute infection conditions, retained maternal Tregs with fetal specificity are likely to have more frequent antigenic encounters from fetal cells that establish microchimerism, analogous to low-level antigen stimulation in the later stages of persistent infection. Fetal cell microchimerism initiated during pregnancy and sustained postpartum probably occurs ubiquitously, but this phenomenon has become only recently widely appreciated with the use of molecular tools that allow these rare (~1 in 106) cells to be consistently identified.27-29 Accordingly, antigenic reminder from fetal cell microchimerism may be pivotal for sustaining memory among pregnancy induced maternal Tregs. Moreover, if maternal CD4+ Treg memory is sustained by fetal cell microchimerism, it would be interesting to consider the necessity for comparable antigenic reminders in maintaining regulatory CD8+ T cells shown in other contexts.30-32 Along with the long-term maintenance of maternal cell microchimerism sustained by fetal Tregs in offspring,33 this emerging body of evidence highlight remarkably potent and long-lived immunological programming that occurs naturally with the bi-directional transfer of cells and antigens between mother and fetus through in utero exposure.

Based on these findings, we propose important next steps are to more meticulously dissect the physiological milieu of pregnancy and in utero development that primes immunological tolerance and Treg memory. Taking cues from effector CD4+ T cell memory,1-7, 34 this will likely include interrelated contributions from naive cell precursor frequency, primary expansion magnitude, antigen avidity, and response to cytokine growth factors, along with increased frequency of antigenic reminders.

Furthermore, given the potential for Treg conversion into inflammatory cytokine producing effector T cells with the same specificity,35, 36 microchimeric fetal cells also have the dangerous potential for sensitizing responses that may trigger autoimmunity.37-39 This is analogous to pathological responses to microchimeric maternal cells in offspring with various diverse autoimmune disorders including diabetes,40 biliary atresia,41 and dermatomyositis.42 Therefore, establishing the molecular signals that reinforce Treg differentiation stability are of equally high importance and priority.

Nevertheless, applied to the devastating complications in human pregnancy that stem from underlying defects in fetal tolerance (preeclampsia, prematurity, miscarriage), basic investigation on the fundamental biology of CD4+ T cells and memory features for protective regulatory subsets provides renewed hope for new, more efficacious therapeutic approaches. In turn, given the striking parallels between Treg and effector CD4+ T cell memory, unraveling how maternal Treg memory is sustained will likely also provide critical insights for priming more durable

effector T cells with pathogen specificity for augmenting host defense against infection.

Acknowledgements. This work was supported in part by the NIAID through awards

R01-AI087830, R01-AI100934, and R21-AI112186.3 SSW holds an Investigator in the Pathogenesis of Infectious Disease award from the Burroughs Wellcome Fund.

References

1. Seder R, Ahmed R. Similarities and differences in CD4+ and CD8 effector and memory T cells generation. Nat Immunol 2003; 4:835 - 42.

2. Williams MA, Holmes BJ, Sun JC, Bevan MJ. Developing and maintaining protective CD8+ memory T cells. Immunol Rev 2006; 211:146-53.

3. Prlic M, Williams M, Bevan M. Requirements for CD8 T-cell priming, memory generation and maintenance. Curr Opin Immunol 2007; 19:315-9.

4. Harty J, Badovinac V. Shaping and reshaping CD8+ T-cell memory. Nature reviews Immunology 2008; 8:107-19.

5. Sprent J, Surh CD. Normal T cell homeostasis: the conversion of naive cells into memory-phenotype cells. Nat Immunol 2011; 12:478-84.

6. van Leeuwen EM, Sprent J, Surh CD. Generation and maintenance of memory CD4(+) T Cells. Curr Opin Immunol 2009; 21:167-72.

7. Pepper M, Jenkins MK. Origins of CD4(+) effector and central memory T cells. Nat Immunol 2011; 12:467-71.

8. Homann D, Teyton L, Oldstone M. Differential regulation of antiviral T-cells immunity results stable CD8+ by declining CD4+ T-cell memory. Nat Med 2001; 7:913-9.

9. Harrington L, Janowski K, Oliver J, Zajac A, Weaver C. Memory CD4 T cells emerge from effector T-cell progenitors. Nature 2008; 452:356-61.

10. Pepper M, Linehan J, Pagan A, Zell T, Dileepan T, Cleary P, Jenkins M. Different routes of bacterial infection induce long-lived Th1 memory cells and short-lived Th17 cells. Nat Immunol 2010; 11:83-90.

11. Reiley WW, Shafiani S, Wittmer ST, Tucker-Heard G, Moon JJ, Jenkins MK, Urdahl KB, Winslow GM, Woodland DL. Distinct functions of antigen-specific CD4 T cells during murine Mycobacterium tuberculosis infection. PNAS 2010; 107:19408-13.

12. Winslow GM, Roberts AD, Blackman MA, Woodland DL. Persistence and turnover of antigen-specific CD4 T cells during chronic tuberculosis infection in the mouse. J Immunol 2003; 170:2046-52.

13. McSorley SJ, Cookson BT, Jenkins MK. Characterization of CD4+ T cell responses during natural infection with Salmonella typhimurium. J Immunol 2000; 164:986-93.

14. Uzonna JE, Wei G, Yurkowski D, Bretscher P. Immune elimination of Leishmania major in mice: implications for immune memory, vaccination, and reactivation disease. J Immunol 2001; 167:6967-74.

15. Belkaid Y, Piccirillo CA, Mendez S, Shevach EM, Sacks DL. CD4+CD25+ regulatory T cells control Leishmania major persistence and immunity. Nature 2002; 420:502-7.

16. Nelson RW, McLachlan JB, Kurtz JR, Jenkins MK. CD4+ T cell persistence and function after infection are maintained by low-level peptide:MHC class II presentation. J Immunol 2013; 190:2828-34.

17. Rosenblum M, Gratz I, Paw J, Lee K, Marshak-Rothstein A, Abbas A. Response to self antigen imprints regulatory memory in tissues. Nature 2011; 480:538-42.

18. Brincks E, Roberts A, Cookenham T, Sell S, Kohlmeier J, Blackman M, Woodland D. Antigen-specific memory regulatory CD4+FoxP3+ T cells control memory responses to influenza virus infection. J Immunol 2013; 190:3438-46.

19. Rowe JH, Ertelt JM, Xin L, Way SS. Pregnancy imprints regulatory memory that sustains anergy to fetal antigen. Nature 2012; 490:102-6.

20. Aluvihare V, Kallikourdis M, Betz A. Regulatory T cells mediate maternal tolerance to the fetus. Nat Immunol 2004; 5:266-71.

21. Rowe JH, Ertelt JM, Aguilera MN, Farrar MA, Way SS. Foxp3(+) regulatory T cell expansion required for sustaining pregnancy compromises host defense against prenatal bacterial pathogens. Cell Host Microbe 2011; 10:54-64.

22. Somerset D, Zheng Y, Kilby M, Sansom D, Drayson M. Normal human pregnancy is associated with an elevation in the immune suppressive CD25+ CD4+ regulatory T-cell subset. Immunology 2004; 112:38-43.

23. Sasaki Y, Darmochwal-Kolarz D, Suzuki D, Sakai M, Ito M, Shima T, Shiozaki A, Rolinksi J, Saito S. Proportion of peripheral blood and decidual CD4+ CD25(bright) regulatory T cells in pre-eclampsia. Clin Exp Immunol 2007; 149:139-45.

24. Toldi G, Svec P, Vasarhelyi B, Meszaros G, Rigo J, Tulassay T, Treszl A. Decreased number of FoxP3+ regulatory T cells in preeclampsia. Acta Obstetricia et Gynecologica Scandinavica 2008; 87:1229-33.

25. Campbell D, MacGillivray I, Carr-Hill R. Pre-eclampsia in second pregnancy. Br J Obstet Gynaecol 1985; 92:131-40.

26. Trupin L, Simon L, Eskenazi B. Change in paternity: a risk factor for preeclampsia in multiparas. Epidemiology 1996; 7:240-4.

27. Bianchi D, Zickwolf G, Weil G, Sylvester S, DeMaria M. Male fetal progenitor cells persist in maternal blood for as long as 27 years postpartum. PNAS 1996; 93:705- 8.

28. Gammill H, Nelson J. Naturally acquired microchimerism. Int J Dev Biol 2010; 54:531-43.

29. Khosrotehrani K, Johnson KL, Cha DH, Salomon RN, Bianchi DW. Transfer of fetal cells with multilineage potential to maternal tissue. JAMA : the journal of the American Medical Association 2004; 292:75-80.

30. Robb RJ, Lineburg KE, Kuns RD, Wilson YA, Raffelt NC, Olver SD, Varelias A, Alexander KA, Teal BE, Sparwasser T, et al. Identification and expansion of highly suppressive CD8(+)FoxP3(+) regulatory T cells after experimental allogeneic bone marrow transplantation. Blood 2012; 119:5898-908.

31. Fleissner D, Hansen W, Geffers R, Buer J, Westendorf AM. Local induction of immunosuppressive CD8+ T cells in the gut-associated lymphoid tissues. PloS one 2010; 5:e15373.

32. Uss E, Rowshani AT, Hooibrink B, Lardy NM, van Lier RA, ten Berge IJ. CD103 is a marker for alloantigen-induced regulatory CD8+ T cells. J Immunol 2006; 177:2775-83.

33. Mold JE, Michaelsson J, Burt TD, Muench MO, Beckerman KP, Busch MP, Lee TH, Nixon DF, McCune JM. Maternal alloantigens promote the development of tolerogenic fetal regulatory T cells in utero. Science 2008; 322:1562-5.

34. Gratz I, Truong H, Yang S, Maurano M, Lee K, Abbas A, Rosenblum M. Cutting Edge: memory regulatory t cells require IL-7 and not IL-2 for their maintenance in peripheral tissues. J Immunol 2013; 190:4483-7.

35. Bailey-Bucktrout SL, Martinez-Llordella M, Zhou X, Anthony B, Rosenthal W, Luche H, Fehling HJ, Bluestone JA. Self-antigen-driven activation induces instability of regulatory T cells during an inflammatory autoimmune response. Immunity 2013; 39:949-62.

36. Komatsu N, Okamoto K, Sawa S, Nakashima T, Oh-Hora M, Kodama T, Tanaka S, Bluestone JA, Takayanagi H. Pathogenic conversion of Foxp3(+) T cells into TH17 cells in autoimmune arthritis. Nat Med 2014; 20:62-8.

37. Adams KM, Nelson JL. Microchimerism: an investigative frontier in autoimmunity and transplantation. JAMA : the journal of the American Medical Association 2004; 291:1127-31.

38. Stevens AM. Microchimeric cells in systemic lupus erythematosus: targets or innocent bystanders? Lupus 2006; 15:820-6.

39. Mosca M, Curcio M, Lapi S, Valentini G, D'Angelo S, Rizzo G, Bombardieri S. Correlations of Y chromosome microchimerism with disease activity in patients with SLE: analysis of preliminary data. Annals of the rheumatic diseases 2003; 62:651-4.

40. Nelson JL, Gillespie KM, Lambert NC, Stevens AM, Loubiere LS, Rutledge JC, Leisenring WM, Erickson TD, Yan Z, Mullarkey ME, et al. Maternal microchimerism in peripheral blood in type 1 diabetes and pancreatic islet beta cell microchimerism. PNAS 2007; 104:1637-42.

41. Suskind DL, Rosenthal P, Heyman MB, Kong D, Magrane G, Baxter-Lowe LA, Muench MO. Maternal microchimerism in the livers of patients with biliary atresia. BMC gastroenterology 2004; 4:14.

42. Reed AM, Picornell YJ, Harwood A, Kredich DW. Chimerism in children with juvenile dermatomyositis. Lancet 2000; 356:2156-7.

Chapter 3: Tolerance to noninherited maternal antigens, reproductive microchimerism and regulatory T cell memory: 60 years after ‘Evidence for actively acquired tolerance to Rh antigens’

* This is an open-access article reprinted here under the terms of the Creative Commons Attribution Non-Commercial License. The manuscript was originally published in 2015. Kinder J. 2015. Chimerism 6(1-2):8-20

Abstract

Compulsory exposure to genetically foreign maternal tissue imprints in offspring sustained tolerance to noninherited maternal antigens (NIMA). Immunological tolerance to NIMA was first described by Dr. Ray D. Owen for women genetically negative for erythrocyte rhesus (Rh) antigen with reduced sensitization from developmental Rh exposure by their mothers. Extending this analysis to HLA haplotypes has uncovered the exciting potential for therapeutically exploiting NIMA-specific tolerance naturally engrained in mammalian reproduction for improved clinical outcomes after allogeneic transplantation. Herein, we summarize emerging scientific concepts stemming from tolerance to NIMA that includes postnatal maintenance of microchimeric maternal origin cells in offspring, expanded accumulation of immune suppressive regulatory T cells with

NIMA-specificity, along with teleological benefits and immunological consequences of

NIMA-specific tolerance conserved across mammalian species.

Introduction, pioneering observations on immunological tolerance by Dr. Ray

Owen

Each individual among outbred populations is immunologically unique – defined by inherited maternal and paternal genes that encode distinctive MHC haplotype alleles along with other minor alloantigens. This established definition of immunological identity, with ensuing implications for tolerance based on binary self versus non-self antigen distinction is based on pioneering observations by Dr. Ray D. Owen comparing antigen diversity among fraternal twin cattle. In a seminal paper published in 1945, only five paragraphs were needed to articulate and re-conceptualize foundational concepts regarding immunological identity and tolerance 1.

Dr. Owen’s roots were in farming dairy cattle. In this context, he made the intriguing observation that a majority of fraternal twin cattle had compatible blood types despite a diversity of at least 40 distinct genetically controlled antigens known for this species 1.

This unexpected finding persisted even in cases of superfecundation, involving twins in the same pregnancy sired by genetically distinct fathers. Owen also noted that one bull derived from a fraternal twin litter failed to transmit his phenotypic blood group antigens in up to twenty of his sired next generation progeny. Reflecting on anatomical vascular anastomoses between bovine twin embryos 2, these observations were pieced together to postulate co-existence of shared blood cells between genetically non-identical twin cattle throughout adult life 1. More importantly, Owen recognized the revolutionary cross-disciplinary implications of these findings for immunology and genetics,

articulating arguably the first definitive example of persistent immunological tolerance to genetically foreign antigens.

Although this seminal characterization of tolerance between fraternal twin cattle establishing the existence of acquired immunological tolerance is most widely recognized, other related contributions have had equally sustained impacts shaping research on immunological responsiveness to developmentally pertinent genetically foreign antigens. This brief review written to commemorate Dr. Owen’s 100th birthday contains a snapshot of past and ongoing work on immune tolerance to noninherited maternal antigen directly stemming from Evidence for actively acquired tolerance to Rh antigens reported by Owens and colleagues 60 years ago 3.

Human immunological tolerance to noninherited maternal antigens

Despite the pervasive immunological implications stemming from tolerance to discordant cells in the somewhat obscure setting of fraternal twinning, far more common is compulsory exposure of each individual during in utero fetal development to genetically foreign maternal cells and tissues that express noninherited maternal antigens (NIMA). Here, Owen was the first to recognize that physiological exposure to discordant maternal antigens in this developmental context can confer sustained immunological tolerance to NIMA in offspring 3. Investigating the heterogeneity of sensitization to erythrocyte rhesus (Rh) antigen among Rh-negative women during pregnancies with Rh-positive male partners, Owen postulated that early developmental

stimulation by Rh antigen among women born to Rh-positive mothers might confer persistent tolerance to Rh sensitization. In other words, exposure to noninherited antigens expressed by the maternal grandmother may have beneficial impacts in women on the outcome of next-generation pregnancies. This hypothesis was addressed by comparing the maternal Rh status of women categorized as either Rh-tolerant or Rh- intolerant following repeated stimulation by concepti bearing this genetically foreign paternal antigen. Remarkably, a significant majority (78%; 32 of 41) of Rh-tolerant women were shown to have Rh-positive mothers, whereas this maternal Rh skewing was eliminated among Rh-intolerant women (48%; 27 of 56 born to Rh-positive mothers) 3. Thus, a critical association between early developmental exposure to Rh antigen for reproductive age women from their own mothers, and protection against Rh sensitization during next generation pregnancies was recognized. Interestingly however, the incidence of erythroblastosis fetalis or newborn hemolytic disease remained similar among Rh-tolerant and Rh-intolerant mothers suggesting NIMA-specific tolerance to this single alloantigen alone was not sufficient to confer survival benefits to next- generation offspring 3, 4.

As the need for immunological tolerance to genetically foreign antigens became increasingly recognized in transfusion medicine and transplantation over the next thirty years, applicability of NIMA-specific tolerance in these clinical contexts has been further evaluated. Functional consequences of NIMA-specific tolerance were demonstrated in individuals who received multiple blood transfusions and normally developed antibodies against almost all HLA alloantigens. Interestingly however, a majority of transfusion dependent individuals broadly exposed to foreign HLA alleles were found to selectively

lack antibodies with specificity to NIMA, compared with noninherited paternal antigen

(NIPA) HLA haplotypes 5. Along with critical implications these data have on organ donor selection prior to transplantation, NIMA-specific tolerance was also recognized to protect against allograft rejection following transplantation. In a landmark retrospective analysis of 205 kidney transplant recipients from HLA mis-matched sibling donors, long- term allograft survival was markedly improved if mismatched for NIMA compared to

NIPA HLA haplotypes 6. In fact, 5 and 10 year survival of NIMA-mismatched kidneys was nearly identical to survival rates of HLA-identical grafts. Together, these classical studies not only reaffirm Owen’s initial observation of persistent postnatal NIMA-specific tolerance, but also highlight exciting translational opportunities for exploiting tolerance to

NIMA naturally imprinted during early development for improved outcomes after transplantation.

More variable degrees of protection have been described after hematopoietic stem cell transplantation exploiting NIMA-specific tolerance to protect against graft-versus-host disease (GVHD) in HLA-discordant donor-recipient pairings 7, 8. Among HLA- haploidentical sibling-to-sibling donor-recipient pairs, van Rood and colleagues reported a 1.9-fold reduced risk of acute GVHD in NIMA-mismatched compared with NIPA- mismatched bone marrow transplants 9. Ichinohe and colleagues described 9.9-fold reduced rates of severe grade III/IV acute GVHD after hematopoietic stem cell transplantation between NIMA-mismatched family members 10. In addition, long-term follow up of the same cohort by Kanda and colleagues revealed immunosuppressive therapy could be successfully withdrawn for NIMA-mismatched transplant recipients with mild GVHD 11.

Despite these remarkable protective benefits from initial retrospective analysis, individual case reports of prospectively selected NIMA-mismatched donor-recipient pairs showed less promising results. In one case series of three patients selected to receive NIMA-mismatched donor hematopoietic stem cells, acute GVHD occurred in two recipients and graft rejection occurred in the third 12. For these recipients of allogeneic donor stem cells, the hypothesis that more intense immune suppressive conditioning therapies may override protection conferred by NIMA-specific tolerance is consistent with comparable survival rates of NIMA-mismatched solid organ allografts in retrospective cohorts when increased potency immune suppression with cyclosporine is used to avert rejection 6, 13. However, the lack of significant protective benefits in these isolated contexts does not negate the need for further investigating how naturally engrained NIMA-specific tolerance can be therapeutically exploited in transplantation and other clinical areas that require more stringent immunological regulation (e.g. allergy, autoimmunity, maternal-fetal tolerance).

Another more intriguing explanation for incomplete and discordant phenotypes of NIMA- specific tolerance may be related to variations in postnatal exposure to maternal cells bearing NIMA through breastfeeding, and ensuing differences in levels of maternal cell microchimerism 14-17. Postnatal persistence of genetically foreign chimeric maternal cells in offspring was originally described in infants with severe combined immune deficiency

18-23. In a study of 121 infants with defective T and B lymphocyte development, 40% had engrafted maternal T cells and a similar proportion developed clinically apparent GVHD caused by anti-fetal allo-immunity 24. Similarly in cases of maternal malignancy during pregnancy, transplacental metastases of immune evasive tumor cells has been

described in melanoma, lymphoblastic leukemia, and lung adenocarcinoma 25-27. With more sensitive techniques allowing detection of potentially rarer maternal cells or their

DNA (e.g. fluorescence in situ hybridization, quantitative polymerase chain reaction), microchimerism of maternal origin is increasingly recognized to occur near ubiquitously among offspring 28-30. Maternal cell-specific DNA can be detected in up to 30% of cord blood specimens at a median concentration of 0.3% among fetal cells 31, whereas DNA encoding maternal MHC haplotype alleles are found in 22% to 55% of healthy adult individuals after analysis of less than 2 µg of peripheral blood DNA 28, 29. Thus, vertical transfer and engraftment of maternal cells in offspring is likely an unavoidable by- product of in utero development through a porous placental interface.

The additional immune modulatory properties of breastfeeding on NIMA-specific tolerance are most definitively illustrated by improved outcomes among breast-fed individuals receiving maternal donor kidney allografts 16. In retrospective inquires on breastfeeding, functional survival of the maternal allograft was significantly increased among breast-fed compared with non-breast fed renal transplant recipient offspring.

These protective benefits of breastfeeding were specific to maternal allograft tissue bearing NIMA, since no differences in paternal donor allograft survival were identified 16.

These remarkable benefits in human transplantation strongly implicate potent antigen- specific immune modulatory properties of soluble maternal-HLA in breast milk 32, 33. In turn, direct associations between breastfeeding, increased postnatal persistence of microchimeric maternal cells and NIMA-specific tolerance shown in complementary animal studies suggest breast milk may contain a critical source of maternal cells that establish microchimerism in offspring 14, 15, 17, 29, 33. Alternatively, another interpretation

of increased maternal cell microchimerism in breast fed individuals is that NIMA-specific tolerance prevents rejection of genetically foreign cells transferred through breast milk.

Additional clues on the developmental ontogeny and adaptive immune components responsible for human NIMA-specific tolerance were unveiled in pioneering analysis of latent anti-maternal immunity in T cells of fetal and adult offspring 34. Recognizing the need for active suppression among maturing fetal T cells during in utero development to avert potentially harmful anti-maternal immunity, Mold, McCune and colleagues showed tolerogenic fetal immune suppressive regulatory T cells develop from exposure to genetically foreign maternal alloantigens 35. This dedicated fetal CD4+ T cell subset identified by expression of the Foxp3 transcriptional regulator or high-affinity IL-2 cytokine receptor, CD25, was shown to selectively suppress anti-maternal responses. In elegant co-culture assays between purified fetal T cells and antigen presenting cells from the biological mother or non-related adult donors, selective suppression of anti- maternal T cell proliferation by fetal CD25+ T cells was demonstrated 35. Expansion of

CD25+ or Foxp3+ regulatory T cells that suppress anti-maternal immunity also paralleled microchimeric maternal cells in fetal lymph node tissue. Thus for human infants with numerically replete T cells at the time of birth, NIMA-specific tolerance is likely essential for restraining harmful anti-maternal immunity during in utero and early postnatal development when exposure to foreign maternal antigens is unavoidable 34, 35. With this reasoning, postnatal persistence of NIMA-specific tolerance through adulthood can be viewed as a developmental remnant of immune suppressive pathways essential for in utero fetal-maternal co-habitation. Together, this emerging body of human retrospective and experimental data sparked by Owen’s initial characterization of maternal Rh

ancestral phenotype highlight profound immune modulatory properties engrained through developmental NIMA exposure. However, this intriguing immunological association between NIMA-specific tolerance, maternal cell microchimerism, and expanded fetal regulatory T cells that actively suppress anti-maternal immunity also raises provocative new questions with regards to why this biological phenomenon is programmed to persist through adulthood.

Animal models of immune tolerance with early developmental antigen exposure

Nearly 10 years after Owen’s seminal description of immunological tolerance between genetically discordant fraternal twin cattle, Dr. Peter Medawar’s analysis of skin graft survival provided pivotally important experimental evidence supporting the existence of actively acquired tolerance to genetically foreign tissues 36. Employing unique strains of highly inbred genetically identical mice, Medawar’s classical experiments showed in utero exposure to cells from discordant mouse strains can confer tolerance to skin grafts that persists through adulthood. Tolerance to skin grafting in chickens was similarly observed after embryonic cell transfer between unique inbred strains identified by distinctive feather coloration 36. Although the use of genetically homozygous inbred animals in these studies precludes analysis of NIMA-specific tolerance, these results nonetheless clearly established antigen-specific tolerogenic properties stemming from in utero and early developmental antigen exposure. Considering immunological tolerance to developmentally irrelevant alloantigens can be primed by in utero stimulation,

physiological exposure of the fetus to semi-allogeneic maternal tissues would be expected to confer similar, if not more profound, immunological tolerance to NIMA.

There is now definitive evidence that NIMA-specific tolerance initially described in retrospective analysis of human transplantation outcomes can be faithfully reproduced and further dissected in animals. Using an elegant F1 backcross breeding strategy to generate genetically identical mice discordantly exposed to defined MHC haplotype alleles as surrogate NIMA or NIPA, Burlingham and colleagues showed remarkable protective benefits against rejection of fully allogeneic NIMA-compatible donor heart grafts 14, 37. In parallel with the aforementioned improved survival of human maternal kidney allografts in breast-fed compared with non-breast fed recipient offspring 16, complementary cross fostering nursing studies in mice show maternal antigen exposure both in utero and through oral breast milk ingestion are simultaneously essential for improved survival of NIMA-mismatched cardiac allografts 14.

Similarly in animal models of hematopoietic stem cell transplantation, GVHD was attenuated in irradiated recipient mice reconstituted with allogeneic NIMA-mismatched donor splenocytes 38. Postnatal exposure to NIMA through breastfeeding also enhances protection against GVHD for immune progenitor cells transferred into NIMA-mismatched irradiated recipient mice 15, and these beneficial impacts are directly linked with postnatal persistence of microchimeric maternal cells in offspring 17. In turn, analysis of individual mice after solid organ or hematopoietic stem cell transplantation have identified direct associations between NIMA-tolerant phenotypes, levels of maternal cell microchimerism and expanded accumulation of CD25+, Foxp3+, or transforming growth

factor-β producing regulatory CD4+ T cells 15, 17, 37, 39. Thus, animal models of NIMA- specific tolerance amenable to experimental investigation have been instrumental in verifying, as well as further establishing the immunological cellular and molecular mechanisms responsible for NIMA-specific tolerance 39.

The study of in utero transplantation of genetically foreign cells also exploits the tolerogenic properties unique to fetal development and provides important mechanistic clues on NIMA-specific tolerance 40-42. The theoretical advantage of in utero transplantation is that therapeutic introduction of genetically foreign cells into the fetal recipient prior to maturation of adaptive immune components can induce long-term donor-specific tolerance without the need for toxic myeloablative conditioning. Animal models of in utero hematopoietic cell transplantation highlight the critical importance of a minimum threshold of antigen exposure necessary to establish and maintain allospecific tolerance 40, 43-48. Therefore, tolerance to NIMA that parallels persistence of maternal origin microchimeric cells also likely hinges on a minimum level of exposure to microchimeric maternal cells 17, 44. Further study is needed however, to determine how the level of maternal microchimerism may dictate alternate outcomes of autoimmunity or

NIMA-specific tolerance. Additionally, given that in utero transplantation of genetically foreign cells to the fetus does not occur in isolation from the immunologically competent mother, maternal allo-sensitization can result from the introduction of discordant third- party alloantigens into the fetus 49, 50. Thus, discordance in protective benefits of NIMA- specific tolerance after transplantation may also reflect transfer of maternal adaptive immune components that have undergone sensitization to fetal-specific antigen 6, 9-13.

To more definitively identify the specificity of immune suppressive regulatory T cells that expand with developmental NIMA exposure, we recently developed a breeding strategy that uniquely transforms defined model antigens into surrogate NIMA 51. Specifically, female mice heterozygous for a previously defined transgene that encodes cell surface expression of a recombinant protein containing ovalbumin plus the 2W1S variant of mouse I-Eα peptide were used for mating with non-transgenic males 52, 53. This

b approach that simultaneously transforms the MHC class II I-A :2W1S55-68 peptide plus ovalbumin into surrogate NIMA in half the offspring, combined with tetramer staining and bead enrichment tools for precisely identifying rare I-Ab:2W1S-specific CD4+ T cells, provided a unique opportunity to investigate the differentiation of endogenous

NIMA-specific cells 54. We found CD4+ T cells with surrogate-NIMA specificity in NIMA- exposed adult mice became highly enriched (~50%) for Foxp3 expression compared with CD4+ T cells of the same specificity in NIPA exposed or control mice without developmental 2W1S exposure 51. In agreement with aforementioned human and mouse studies highlighting the importance of postnatal stimulation by genetically foreign maternal cells through breastfeeding 15-17, expanded accumulation of NIMA-specific

Foxp3+ regulatory T cells declined sharply in cross fostered mice exposed to maternal tissues bearing NIMA-2W1S during in utero development or through breastfeeding in isolation. On the other hand, comparable absolute numbers of I-Ab:2W1S-specific CD4+

T cells, and their similar avidity for cognate I-Ab:2W1S peptide, between NIMA exposed and naive control mice suggest thymic deletion of NIMA responsive cells play less important roles in immunological tolerance to NIMA (51 and unpublished data). Thus,

early developmental exposure to NIMA primes in offspring an expanded pool of peripherally induced immune suppressive regulatory T cells with NIMA-specificity.

Sustained expansion of NIMA-specific regulatory T cells in offspring also paralleled postnatal retention of microchimeric maternal cells. OVA encoding DNA specific to genetically discordant maternal cells was identified in vital organs (e.g. liver, heart) of

NIMA exposed mice, at levels corresponding to 1 maternal cell in 105 to 106 offspring cells in agreement with other studies using complementary tools for estimating levels of maternal microchimerism in mice and non-human primates 17, 47, 48, 51, 55, 56. One interpretation of these data is that postnatal persistence of NIMA-specific tolerance is a developmental remnant that protects genetically foreign microchimeric maternal cells from rejection in offspring. Alternatively, retained microchimeric maternal cells may have themselves adapted tolerogenic properties required for driving expanded accumulation of NIMA-specific regulatory T cells and therefore promote their own survival.

To definitively investigate the cause and effect relationship between the interrelated phenomena of expanded accumulation of NIMA-specific regulatory T cells and microchimeric maternal cells simultaneously retained in offspring, the impacts of selectively depleting microchimeric maternal cells based on co-expression of ovalbumin protein with 2W1S55-58 peptide in NIMA exposed mice was evaluated. Remarkably, expanded accumulation of NIMA-specific regulatory T cells declined to background levels found in NIPA or naive control mice within the first 2 weeks after depleting microchimeric 2W1S-OVA+ maternal cells with anti-ovalbumin antibodies 51. These results are consistent with the hypothesis that microchimeric cells provide an essential

source of cognate maternal antigen required for maintaining expanded accumulation of

NIMA-specific regulatory T cells. In this regard, numerical retention of memory regulatory T cells with NIMA-specificity appear to share with effector CD4+ T cells of foreign microbial specificity the necessity for frequent, if not constant, cognate antigen exposure reminders. Based on these findings, it may be worthwhile to investigate if memory regulatory T cells described in other contexts (e.g. transient expression in the skin or after acute infection with viral pathogens) 57-60 represent bona fide memory like activated CD8+ T cells, or alternatively share a requirement for low-level exposure to cognate antigen 61-65. In the broader scientific context, these results illustrate how dissecting the fundamental immunology responsible for NIMA-specific tolerance can continue to reveal hidden immunological secrets engrained in mammalian reproduction.

Teleological benefits and immunological consequences of NIMA-specific tolerance

Despite the primary use of animal models to verify the existence and further establish immunological mechanisms responsible for NIMA-specific tolerance, comparison of

NIMA-specific tolerance across mammalian species can also provide critical insights on the evolutionary ontogeny of this highly engrained immunological phenomena. For human infants with numerically replete adaptive immune components at the time of birth

34, 35 tolerance to genetically foreign maternal cells and tissues begins in utero with suppressed activation of maturing immune cells with NIMA-specificity. However, this reasoning does not explain why tolerance imprinted by exposure to foreign antigens during early development is widely conserved across mammalian species (e.g. non-

human primates, ruminants, rodents) with sharply delayed adaptive immune cell maturation relative to parturition 34, 66. For example, prolonged survival of NIMA- matched allografts and expanded NIMA-specific regulatory T cells in human adults is consistently reproduced in adult mice despite the absence of peripheral T cells at the time of birth for this species 14, 34, 37. The preservation of NIMA-specific tolerance in mammalian species born without functional adaptive immune components suggests the existence of more universal biological benefits driving conserved tolerance to NIMA in placental mammals.

An important clue is postnatal persistence of NIMA-specific tolerance through adulthood that is actively maintained by maternal cells that established microchimerism in offspring

51. In turn, given the necessity for expanded tolerance that encompasses immunologically foreign paternal-fetal antigens in successful pregnancy shared by all eutherian placental mammals 67-69, we reasoned reinforced fetal tolerance during next- generation pregnancies may represent a more universal explanation for evolutionarily conserved NIMA-specific tolerance. This notion is supported by our recent demonstration of expanded NIMA-specific regulatory T cell accumulation in female compared with male NIMA-2W1S littermate offspring, and correspondingly enriched microchimeric maternal cell retained in female gender specific reproductive tissue 51.

To further investigate this hypothesis, susceptibility to complications during allogeneic pregnancy stemming from disruptions in fetal tolerance were evaluated in genetically identical female mice developmentally exposed to discordant MHC haplotypes as surrogate NIMA. Remarkably, this analysis showed NIMA-specific tolerance confers profound resiliency against fetal wastage normally triggered by infection with the

prenatal bacterial pathogen Listeria monocytogenes or partial transient depletion of bulk maternal regulatory T cells 51, 70-72. These protective benefits occurred in an antigen- specific fashion requiring commonality between NIMA and paternal-fetal expressed antigens since susceptibility to fetal wastage rebounded when third-party male mice bearing irrelevant MHC haplotype alleles were used for mating with NIMA-exposed female mice. Thus, expanded accumulation of regulatory T cells with fetal specificity, primed by either developmental NIMA stimulation or prior pregnancy, efficiently overrides susceptibility to invasive infection with prenatal pathogens like Listeria monocytogenes conferred by increased non-specific immune suppression from accumulation of bulk maternal regulatory T cells 71, 73. Given the pivotal importance of decidual infiltration by activated fetal-specific CD8+ T cells in the immune-pathogenesis of fetal wastage that occurs with prenatal infection 70, dissecting the anatomical and molecular details whereby fetal-specific regulatory CD4+ T cells efficiently reinforce fetal tolerance are critically important areas for future investigation with direct translational implications for improving human pregnancy outcomes.

Cross-generational protection against fetal wastage in animal pregnancy models are also in agreement with the ‘grandmother effect’ reported by Gammill, Nelson and colleagues where reduced rates of pregnancy complications stemming from disruptions in fetal tolerance (e.g. preeclampsia, recurrent miscarriage) in women parallel increased levels of microchimeric cells retained from their mothers 74-76. Given the necessity for overlap between NIMA and fetal expressed antigen during next-generation pregnancies for reinforced fetal tolerance in mice 51, important next-step are to investigate if similar overlap augments resiliency against complications in human pregnancy. In this regard,

while developmental exposure to the single minor Rh alloantigen initially described by

Owen was insufficient to prevent hemolytic disease of the newborn 3, 4, these data suggest broader non-inherited antigenic overlap between maternal grandmother and offspring that encompasses MHC haplotype alleles may more efficiently confer cross- generational reproductive benefits.

These findings also highlight striking commonality between reproductive benefits conferred by expanded regulatory T cells with NIMA-specificity retained in females from developmental exposure to genetically foreign maternal antigens, and regulatory T cells with fetal specificity retained in mothers after prior pregnancy 73. In each case, memory regulatory T cells in reproductive age females re-accumulate with sharply accelerated tempo in response to cognate fetal antigen stimulation that protects against disruptions in fetal tolerance. The actively retained enriched pool of memory maternal regulatory T cells with specificity to pre-existing fetal antigen provides an intriguing scientific framework that explains human partner-specific protective benefits of prior pregnancy against complications in subsequent pregnancies 77, 78. Given the necessity for cognate antigen reminders in the form of microchimeric maternal cells in maintaining expanded accumulation of NIMA-specific regulatory T cells 51, it is tantalizing to hypothesize that the protective subset of mother’s little helpers in the form of maternal regulatory T cells with pre-existing fetal specificity are similarly maintained by fetal cells that establish persistent microchimerism in mothers after parturition 79-83. In other words, microchimeric maternal cells that promote cross-generational reproductive fitness, and fetal cells that establish microchimerism in mothers after pregnancy can each be more accurately viewed as mother’s little genetic helpers. Thus, establishing functional

similarities and potential differences between how maternal compared with fetal microchimeric cells prime and sustain expanded memory regulatory T cells represent critically important areas for future investigation.

In the larger biological context, reproductive fitness during next generation pregnancies conferred by NIMA-specific tolerance underscores the remarkably engrained drive for genetic fitness in each female individual. In addition to transmitting half of homologous chromosomes through Mendelian inheritance, vertically transferred maternal cells that establish microchimerism in offspring selectively enforce fetal tolerance during next generation pregnancies that promotes conservation of NIMA 51. However in nature,

NIMA-specific tolerance among dominant female individuals is likely counterbalanced by pathogen-mediated selection for MHC diversity amongst homologous chromosomes within individuals, and between individuals across outbred populations 30, 84, 85.

Together, these findings highlight the need for more extended cross-generational analysis to resolve the ongoing controversy regarding how MHC haplotype similarity impacts mate selection and pregnancy outcomes 86-88. Here, the efficiency whereby

NIMA-specific tolerance retained in adult female mice protects against disruptions in fetal tolerance during next generation pregnancies strongly suggests enhanced protection against rejecting the fetal allograft in addition to other selective benefits promoting pathogen resistance drives preservation of this immunological phenomena at least in some placental mammalian species.

On the other hand, cross-generational reproductive advantages that preserve postnatal retention of microchimeric maternal cells may also perpetuate auto-inflammatory or autoimmune diseases in offspring 89, 90. These potentially harmful consequences of

retained microchimeric maternal cells have been best characterized in individuals with scleroderma where increased levels of maternal microchimerism have been identified 28,

29. Enriched microchimeric maternal cells are found in the peripheral blood and pancreatic tissue of individuals with type 1 diabetes 91. For individuals with rheumatoid arthritis, remarkable links between noninherited maternal HLA-DR alleles associated with disease susceptibility and resistance have each been described 92-94. Similarly, maternal cell microchimerism has also been recently shown to be increased among premature offspring in animal models of inflammation-induced preterm birth 95.

Sharply increased levels of microchimeric cells have also been described in the target tissue of infants and children with various autoimmune disorders. For example, significantly increased levels of female chimeric cells, presumably of maternal origin, were identified in muscle biopsy specimens of children with juvenile dermatomyositis or other idiopathic inflammatory myopathies 96, 97. Maternal chimeric cells were also uniformly identified in 15 cardiac biopsy samples from infants with neonatal lupus syndrome 98, and two independent case series of liver biopsy specimens from infants with biliary atresia 94, 99. Together, these clinical observations support the intriguing possibility that allo-reactivity either against or initiated by genetically foreign microchimeric cell could represent an important trigger for human autoimmune and autoinflammatory disorders 89, 100.

Interestingly however, there is also compelling data for improved survival of maternal compared with paternal hepatic allografts for infants with liver failure secondary to biliary atresia 96, 97, 101. Therefore, a definitive pathological role for microchimeric maternal cells in triggering autoimmunity will require additional investigation since enriched chimeric

cells in damaged or diseased tissues may also reflect their participation in tissue repair and regeneration. Furthermore, qualitative shifts in the molecular phenotype of microchimeric cells, timing and inflammatory context of developmental exposure to maternal tissues are each also likely to play profound roles in controlling whether tolerance or sensitization to NIMA develops 39, 40, 44, 102. Given the conserved nature of

NIMA-specific tolerance and long-term retention of microchimeric maternal cells in humans and mice, further studies using representative animal models of autoimmunity that bypass limitations in human tissue availability are likely to be highly informative in dissecting the beneficial and detrimental impacts of increasingly recognized constitutive chimerism among individuals.

Concluding perspectives

Sixty years since the initial description of actively acquired tolerance to Rh antigens by

Dr. Ray Owen we have witnessed a dramatic explosion of new data highlighting not only the existence of NIMA-specific tolerance, but also the translational applicability of this engrained immunological phenomenon in human solid organ and stem cell transplantation. With new technology for identifying exceptionally rare microchimeric maternal cells and transgenic mouse tools for tracking NIMA-specific immune components, NIMA-specific tolerance is now recognized to occur with vertical transmission of maternal cells that establish persistent microchimerism in offspring. In turn, the recognition of individuals in outbred populations as being constitutively chimeric, with non-inherited legacy of tolerogenic microchimeric cells, forces

reconsideration of immunological identity beyond binary definitions of self versus non- self antigen distinction – that incorporates transmission of maternal attributes through matrilineal non-Mendelian heredity 103. Along with improved outcomes after transplantation, further mining immunological secrets engrained within mammalian reproduction initially recognized by Owen may hold exciting new keys for more effective therapeutic strategies for preventing pregnancy complications and reversing autoimmunity.

References

1. Owen RD. Immunogenetic Consequences of Vascular Anastomoses between Bovine Twins. Science 1945; 102:400-1.

2. Lillie FR. The Theory of the Free-Martin. Science 1916; 43:611-3.

3. Owen RD, Wood HR, Foord AG, Sturgeon P, Baldwin LG. Evidence for actively acquired tolerance to Rh antigens. Proc Natl Acad Sci USA 1954; 40:420-4.

4. Booth PB, Dunsford I, Grant J, Murray S. Haemolytic disease in first-born infants. British medical journal 1953; 2:41-2.

5. Claas FH, Gijbels Y, van der Velden-de Munck J, van Rood JJ. Induction of B cell unresponsiveness to noninherited maternal HLA antigens during fetal life. Science 1988; 241:1815-7.

6. Burlingham WJ, Grailer AP, Heisey DM, Claas FH, Norman D, Mohanakumar T, Brennan DC, de Fijter H, van Gelder T, Pirsch JD, et al. The effect of tolerance to noninherited maternal HLA antigens on the survival of renal transplants from sibling donors. N Engl J Med 1998; 339:1657-64.

7. Eikmans M, van Halteren AG, van Besien K, van Rood JJ, Drabbels JJ, Claas FH. Naturally acquired microchimerism: implications for transplantation outcome and novel methodologies for detection. Chimerism 2014; 5:24-39.

8. Hirayama M, Azuma E, Komada Y. Tolerogenic effect of non-inherited maternal antigens in hematopoietic stem cell transplantation. Front Immunol 2012; 3:135.

9. van Rood JJ, Loberiza FR, Jr., Zhang MJ, Oudshoorn M, Claas F, Cairo MS, Champlin RE, Gale RP, Ringden O, Hows JM, et al. Effect of tolerance to noninherited maternal antigens on the occurrence of graft-versus-host disease after bone marrow transplantation from a parent or an HLA-haploidentical sibling. Blood 2002; 99:1572-7.

10. Ichinohe T, Uchiyama T, Shimazaki C, Matsuo K, Tamaki S, Hino M, Watanabe A, Hamaguchi M, Adachi S, Gondo H, et al. Feasibility of HLA-haploidentical hematopoietic stem cell transplantation between noninherited maternal antigen (NIMA)-mismatched family members linked with long-term fetomaternal microchimerism. Blood 2004; 104:3821-8.

11. Kanda J, Ichinohe T, Shimazaki C, Hamaguchi M, Watanabe A, Ishida H, Yoshihara T, Morimoto A, Uoshima N, Adachi S, et al. Long-term survival after HLA-haploidentical SCT from noninherited maternal antigen-mismatched family donors: impact of chronic GVHD. Bone marrow transplantation 2009; 44:327-9.

12. Okumura H, Yamaguchi M, Kotani T, Sugimori N, Sugimori C, Ozaki J, Kondo Y, Yamazaki H, Chuhjo T, Takami A, et al. Graft rejection and hyperacute graft- versus-host disease in stem cell transplantation from non-inherited maternal- antigen-complementary HLA-mismatched siblings. European journal of haematology 2007; 78:157-60.

13. Opelz G. The effect of tolerance to noninherited maternal HLA antigens on the survival of renal transplants from sibling donors. The New England journal of medicine 1999; 340:1369-70.

14. Andrassy J, Kusaka S, Jankowska-Gan E, Torrealba JR, Haynes LD, Marthaler BR, Tam RC, Illigens BM, Anosova N, Benichou G, et al. Tolerance to noninherited maternal MHC antigens in mice. Journal of immunology 2003; 171:5554-61.

15. Aoyama K, Koyama M, Matsuoka K, Hashimoto D, Ichinohe T, Harada M, Akashi K, Tanimoto M, Teshima T. Improved outcome of allogeneic bone marrow transplantation due to breastfeeding-induced tolerance to maternal antigens. Blood 2009; 113:1829-33.

16. Campbell DA, Jr., Lorber MI, Sweeton JC, Turcotte JG, Niederhuber JE, Beer AE. Breast feeding and maternal-donor renal allografts. Possibly the original donor- specific transfusion. Transplantation 1984; 37:340-4.

17. Dutta P, Molitor-Dart M, Bobadilla JL, Roenneburg DA, Yan Z, Torrealba JR, Burlingham WJ. Microchimerism is strongly correlated with tolerance to noninherited maternal antigens in mice. Blood 2009; 114:3578-87.

18. Flomenberg N, Dupont B, O'Reilly RJ, Hayward A, Pollack MS. The use of T cell culture techniques to establish the presence of an intrauterine-derived maternal T cell graft in a patient with severe combined immunodeficiency (SCID). Transplantation 1983; 36:733-5.

19. Geha RS, Reinherz E. Identification of circulating maternal T and B lymphocytes in uncomplicated severe combined immunodeficiency by HLA typing of subpopulations of T cells separated by the fluorescence-activated cell sorter and of Epstein Barr virus-derived B cell lines. Journal of immunology 1983; 130:2493-5.

20. Kadowaki J, Thompson RI, Zuelzer WW, Woolley PV, Jr., Brough AJ, Gruber D. XX-XY lymphoid chimaerism in congenital immunological deficiency syndrome with thymic alymphoplasia. Lancet 1965; 2:1152-6.

21. Pollack MS, Kapoor N, Sorell M, Kim SJ, Christiansen FT, Silver DM, Dupont B, O'Reilly RJ. DR-positive maternal engrafted T cells in a severe combined immunodeficiency patient without graft-versus-host disease. Transplantation 1980; 30:331-4.

22. Pollack MS, Kirkpatrick D, Kapoor N, Dupont B, O'Reilly RJ. Identification by HLA typing of intrauterine-derived maternal T cells in four patients with severe combined immunodeficiency. The New England journal of medicine 1982; 307:662-6.

23. Thompson LF, O'Connor RD, Bastian JF. Phenotype and function of engrafted maternal T cells in patients with severe combined immunodeficiency. Journal of immunology 1984; 133:2513-7.

24. Muller SM, Ege M, Pottharst A, Schulz AS, Schwarz K, Friedrich W. Transplacentally acquired maternal T lymphocytes in severe combined immunodeficiency: a study of 121 patients. Blood 2001; 98:1847-51.

25. Alexander A, Samlowski WE, Grossman D, Bruggers CS, Harris RM, Zone JJ, Noyes RD, Bowen GM, Leachman SA. Metastatic melanoma in pregnancy: risk of transplacental metastases in the infant. Journal of clinical oncology : official journal of the American Society of Clinical Oncology 2003; 21:2179-86.

26. Isoda T, Ford AM, Tomizawa D, van Delft FW, De Castro DG, Mitsuiki N, Score J, Taki T, Morio T, Takagi M, et al. Immunologically silent cancer clone transmission from mother to offspring. Proceedings of the National Academy of Sciences of the United States of America 2009; 106:17882-5.

27. Walker JW, Reinisch JF, Monforte HL. Maternal pulmonary adenocarcinoma metastatic to the fetus: first recorded case report and literature review. Pediatric pathology & molecular medicine 2002; 21:57-69.

28. Lambert NC, Erickson TD, Yan Z, Pang JM, Guthrie KA, Furst DE, Nelson JL. Quantification of maternal microchimerism by HLA-specific real-time polymerase chain reaction: studies of healthy women and women with scleroderma. Arthritis and rheumatism 2004; 50:906-14.

29. Maloney S, Smith A, Furst DE, Myerson D, Rupert K, Evans PC, Nelson JL. Microchimerism of maternal origin persists into adult life. The Journal of clinical investigation 1999; 104:41-7.

30. Spurgin LG, Richardson DS. How pathogens drive genetic diversity: MHC, mechanisms and misunderstandings. Proc Biol Sci 2010; 277:979-88.

31. Lo YM, Lau TK, Chan LY, Leung TN, Chang AM. Quantitative analysis of the bidirectional fetomaternal transfer of nucleated cells and plasma DNA. Clinical chemistry 2000; 46:1301-9.

32. Molitor ML, Haynes LD, Jankowska-Gan E, Mulder A, Burlingham WJ. HLA class I noninherited maternal antigens in cord blood and breast milk. Hum Immunol 2004; 65:231-9.

33. Zhou L, Yoshimura Y, Huang Y, Suzuki R, Yokoyama M, Okabe M, Shimamura M. Two independent pathways of maternal cell transmission to offspring: through

placenta during pregnancy and by breast-feeding after birth. Immunology 2000; 101:570-80.

34. Mold JE, McCune JM. Immunological tolerance during fetal development: from mouse to man. Adv Immunol 2012; 115:73-111.

35. Mold JE, Michaelsson J, Burt TD, Muench MO, Beckerman KP, Busch MP, Lee TH, Nixon DF, McCune JM. Maternal alloantigens promote the development of tolerogenic fetal regulatory T cells in utero. Science 2008; 322:1562-5.

36. Billingham RE, Brent L, Medawar PB. Actively acquired tolerance of foreign cells. Nature 1953; 172:603-6.

37. Molitor-Dart ML, Andrassy J, Kwun J, Kayaoglu HA, Roenneburg DA, Haynes LD, Torrealba JR, Bobadilla JL, Sollinger HW, Knechtle SJ, et al. Developmental exposure to noninherited maternal antigens induces CD4+ T regulatory cells: relevance to mechanism of heart allograft tolerance. Journal of immunology 2007; 179:6749-61.

38. Araki M, Hirayama M, Azuma E, Kumamoto T, Iwamoto S, Toyoda H, Ito M, Amano K, Komada Y. Prediction of reactivity to noninherited maternal antigen in MHC- mismatched, minor histocompatibility antigen-matched stem cell transplantation in a mouse model. Journal of immunology 2010; 185:7739-45.

39. Dutta P, Burlingham WJ. Microchimerism: tolerance vs. sensitization. Current opinion in organ transplantation 2011; 16:359-65.

40. Alhajjat AM, Lee AE, Strong BS, Shaaban AF. NK cell tolerance as the final endorsement of prenatal tolerance after in utero hematopoietic cellular transplantation. Frontiers in pharmacology 2015; 6:51.

41. Merianos D, Heaton T, Flake AW. In utero hematopoietic stem cell transplantation: progress toward clinical application. Biology of blood and marrow transplantation : journal of the American Society for Blood and Marrow Transplantation 2008; 14:729-40.

42. Derderian SC, Jeanty C, Walters MC, Vichinsky E, MacKenzie TC. In utero hematopoietic cell transplantation for hemoglobinopathies. Frontiers in pharmacology 2014; 5:278.

43. Alhajjat AM, Strong BS, Lee AE, Turner LE, Wadhwani RK, Ortaldo JR, Heusel JW, Shaaban AF. Prenatal Allospecific NK Cell Tolerance Hinges on Instructive Allorecognition through the Activating Receptor during Development. Journal of immunology 2015.

44. Durkin ET, Jones KA, Rajesh D, Shaaban AF. Early chimerism threshold predicts sustained engraftment and NK-cell tolerance in prenatal allogeneic chimeras. Blood 2008; 112:5245-53.

45. Kroemer A, Edtinger K, Li XC. The innate natural killer cells in transplant rejection and tolerance induction. Current opinion in organ transplantation 2008; 13:339-43.

46. Ashizuka S, Peranteau WH, Hayashi S, Flake AW. Busulfan-conditioned bone marrow transplantation results in high-level allogeneic chimerism in mice made tolerant by in utero hematopoietic cell transplantation. Experimental hematology 2006; 34:359-68.

47. Kim HB, Shaaban AF, Milner R, Fichter C, Flake AW. In utero bone marrow transplantation induces donor-specific tolerance by a combination of clonal deletion and clonal anergy. Journal of pediatric surgery 1999; 34:726-9; discussion 9-30.

48. Kim HB, Shaaban AF, Yang EY, Liechty KW, Flake AW. Microchimerism and tolerance after in utero bone marrow transplantation in mice. The Journal of surgical research 1998; 77:1-5.

49. Merianos DJ, Tiblad E, Santore MT, Todorow CA, Laje P, Endo M, Zoltick PW, Flake AW. Maternal alloantibodies induce a postnatal immune response that limits engraftment following in utero hematopoietic cell transplantation in mice. The Journal of clinical investigation 2009; 119:2590-600.

50. Jeanty C, Derderian SC, Mackenzie TC. Maternal-fetal cellular trafficking: clinical implications and consequences. Current opinion in pediatrics 2014; 26:377-82.

51. Kinder JM, Jiang TT, Ertelt JM, Xin L, Strong BS, Shaaban AF, Way SS. Cross- generational reproductive fitness enforced by microchimeric maternal cells. Cell 2015; 162:[Epub ahead of print, July 23 2015].

52. Moon JJ, Dash P, Oguin TH, 3rd, McClaren JL, Chu HH, Thomas PG, Jenkins MK. Quantitative impact of thymic selection on Foxp3+ and Foxp3- subsets of self- peptide/MHC class II-specific CD4+ T cells. Proc Natl Acad Sci USA 2011; 108:14602-7.

53. Rees W, Bender J, Teague TK, Kedl RM, Crawford F, Marrack P, Kappler J. An inverse relationship between T cell receptor affinity and antigen dose during CD4(+) T cell responses in vivo and in vitro. Proc Natl Acad Sci USA 1999; 96:9781-6.

54. Moon J, Chu H, Pepper M, McSorley S, Jameson S, Kedl R, Jenkins M. Naive CD4(+) T cell frequency varies for different epitopes and predicts repertoire diversity and response magnitude. Immunity 2007; 27:203-13.

55. Bakkour S, Baker CA, Tarantal AF, Wen L, Busch MP, Lee TH, McCune JM. Analysis of maternal microchimerism in rhesus monkeys (Macaca mulatta) using real-time quantitative PCR amplification of MHC polymorphisms. Chimerism 2014; 5:6-15.

56. Shaaban AF, Kim HB, Gaur L, Liechty KW, Flake AW. Prenatal transplantation of cytokine-stimulated marrow improves early chimerism in a resistant strain

combination but results in poor long-term engraftment. Experimental hematology 2006; 34:1278-87.

57. Brincks E, Roberts A, Cookenham T, Sell S, Kohlmeier J, Blackman M, Woodland D. Antigen-specific memory regulatory CD4+FoxP3+ T cells control memory responses to influenza virus infection. Journal of immunology 2013; 190:3438-46.

58. Rosenblum M, Gratz I, Paw J, Lee K, Marshak-Rothstein A, Abbas A. Response to self antigen imprints regulatory memory in tissues. Nature 2011; 480:538-42.

59. Gratz IK, Rosenblum MD, Maurano MM, Paw JS, Truong HA, Marshak-Rothstein A, Abbas AK. Cutting edge: Self-antigen controls the balance between effector and regulatory T cells in peripheral tissues. Journal of immunology 2014; 192:1351-5.

60. Sanchez AM, Zhu J, Huang X, Yang Y. The development and function of memory regulatory T cells after acute viral infections. Journal of immunology 2012; 189:2805-14.

61. Homann D, Teyton L, Oldstone M. Differential regulation of antiviral T-cells immunity results stable CD8+ by declining CD4+ T-cell memory. Nat Med 2001; 7:913-9.

62. Nelson RW, McLachlan JB, Kurtz JR, Jenkins MK. CD4+ T cell persistence and function after infection are maintained by low-level peptide:MHC class II presentation. Journal of immunology 2013; 190:2828-34.

63. Seder R, Ahmed R. Similarities and differences in CD4+ and CD8 effector and memory T cells generation. Nat Immunol 2003; 4:835 - 42.

64. Uzonna JE, Wei G, Yurkowski D, Bretscher P. Immune elimination of Leishmania major in mice: implications for immune memory, vaccination, and reactivation disease. Journal of immunology 2001; 167:6967-74.

65. Williams MA, Holmes BJ, Sun JC, Bevan MJ. Developing and maintaining protective CD8+ memory T cells. Immunol Rev 2006; 211:146-53.

66. Adkins B, Leclerc C, Marshall-Clarke S. Neonatal adaptive immunity comes of age. Nature reviews Immunology 2004; 4:553-64.

67. Erlebacher A. Mechanisms of T cell tolerance towards the allogeneic fetus. Nature reviews Immunology 2013; 13:23-33.

68. Munoz-Suano A, Hamilton AB, Betz AG. Gimme shelter: the immune system during pregnancy. Immunol Rev 2011; 241:20-38.

69. Samstein RM, Josefowicz SZ, Arvey A, Treuting PM, Rudensky AY. Extrathymic generation of regulatory T cells in placental mammals mitigates maternal-fetal conflict. Cell 2012; 150:29-38.

70. Chaturvedi V, Ertelt JM, Jiang TT, Kinder JM, Xin L, Owens KJ, Jones HN, Way SS. CXCR3 blockade protects against Listeria monocytogenes infection-induced fetal wastage. The Journal of clinical investigation 2015; 125:1713-25.

71. Rowe JH, Ertelt JM, Aguilera MN, Farrar MA, Way SS. Foxp3(+) regulatory T cell expansion required for sustaining pregnancy compromises host defense against prenatal bacterial pathogens. Cell Host Microbe 2011; 10:54-64.

72. Rowe JH, Ertelt JM, Xin L, Way SS. Listeria monocytogenes cytoplasmic entry induces fetal wastage by disrupting maternal FoxP3+ regulatory cell-sustained fetal tolerance. PLoS Pathogens 2012; 8.

73. Rowe JH, Ertelt JM, Xin L, Way SS. Pregnancy imprints regulatory memory that sustains anergy to fetal antigen. Nature 2012; 490:102-6.

74. Gammill HS, Adams Waldorf KM, Aydelotte TM, Lucas J, Leisenring WM, Lambert NC, Nelson JL. Pregnancy, microchimerism, and the maternal grandmother. PLoS One 2011; 6:e24101.

75. Gammill HS, Stephenson MD, Aydelotte TM, Nelson JL. Microchimerism in recurrent miscarriage. Cellular & molecular immunology 2014; 11:589-94.

76. Gammill HS, Stephenson MD, Aydelotte TM, Nelson JL. Microchimerism in women with recurrent miscarriage. Chimerism 2015:0.

77. Campbell D, MacGillivray I, Carr-Hill R. Pre-eclampsia in second pregnancy. Br J Obstet Gynaecol 1985; 92:131-40.

78. Trupin L, Simon L, Eskenazi B. Change in paternity: a risk factor for preeclampsia in multiparas. Epidemiology 1996; 7:240-4.

79. Kinder JM, Jiang TT, Clark DR, Chaturvedi V, Xin L, Ertelt JM, Way SS. Pregnancy- induced maternal regulatory T cells, bona fide memory or maintenance by antigenic reminder from fetal cell microchimerism? Chimerism 2014; 5:16-9.

80. Bianchi D, Zickwolf G, Weil G, Sylvester S, DeMaria M. Male fetal progenitor cells persist in maternal blood for as long as 27 years postpartum. PNAS 1996; 93:705- 8.

81. Trowsdale J, Betz A. Mother's little helpers: mechanisms of maternal-fetal tolerance. Nature Immunology 2006; 7:241-6.

82. Adams KM, Lambert NC, Heimfeld S, Tylee TS, Pang JM, Erickson TD, Nelson JL. Male DNA in female donor apheresis and CD34-enriched products. Blood 2003; 102:3845-7.

83. O'Donoghue K, Chan J, de la Fuente J, Kennea N, Sandison A, Anderson JR, Roberts IA, Fisk NM. Microchimerism in female bone marrow and bone decades

after fetal mesenchymal stem-cell trafficking in pregnancy. Lancet 2004; 364:179- 82.

84. Goulder PJ, Watkins DI. Impact of MHC class I diversity on immune control of immunodeficiency virus replication. Nature reviews Immunology 2008; 8:619-30.

85. Oliver MK, Telfer S, Piertney SB. Major histocompatibility complex (MHC) heterozygote superiority to natural multi-parasite infections in the water vole (Arvicola terrestris). Proc Biol Sci 2009; 276:1119-28.

86. Israeli M, Kristt D, Nardi Y, Klein T. Genetic Considerations in Human Sex-Mate Selection: Partners Share Human Leukocyte Antigen but not Short-Tandem-Repeat Identity Markers. Am J Reprod Immunol 2014; 71:467-71.

87. Chaix R, Cao C, Donnelly P. Is mate choice in humans MHC-dependent? PLoS Genet 2008; 4:e1000184.

88. Ober C, Weitkamp LR, Cox N, Dytch H, Kostyu D, Elias S. HLA and mate choice in humans. Am J Hum Genet 1997; 61:497-504.

89. Nelson JL. The otherness of self: microchimerism in health and disease. Trends Immunol 2012; 33:421-7.

90. Stevens AM. Microchimeric cells in systemic lupus erythematosus: targets or innocent bystanders? Lupus 2006; 15:820-6.

91. Nelson JL, Gillespie KM, Lambert NC, Stevens AM, Loubiere LS, Rutledge JC, Leisenring WM, Erickson TD, Yan Z, Mullarkey ME, et al. Maternal microchimerism in peripheral blood in type 1 diabetes and pancreatic islet beta cell microchimerism. PNAS 2007; 104:1637-42.

92. Feitsma AL, Worthington J, van der Helm-van Mil AH, Plant D, Thomson W, Ursum J, van Schaardenburg D, van der Horst-Bruinsma IE, van Rood JJ, Huizinga TW, et al. Protective effect of noninherited maternal HLA-DR antigens on rheumatoid arthritis development. Proceedings of the National Academy of Sciences of the United States of America 2007; 104:19966-70.

93. Rak JM, Maestroni L, Balandraud N, Guis S, Boudinet H, Guzian MC, Yan Z, Azzouz D, Auger I, Roudier C, et al. Transfer of the shared epitope through microchimerism in women with rheumatoid arthritis. Arthritis and rheumatism 2009; 60:73-80.

94. van der Horst-Bruinsma IE, Hazes JM, Schreuder GM, Radstake TR, Barrera P, van de Putte LB, Mustamu D, van Schaardenburg D, Breedveld FC, de Vries RR. Influence of non-inherited maternal HLA-DR antigens on susceptibility to rheumatoid arthritis. Annals of the rheumatic diseases 1998; 57:672-5.

95. Wegorzewska M, Le T, Tang Q, MacKenzie TC. Increased maternal T cell microchimerism in the allogeneic fetus during LPS-induced preterm labor in mice. Chimerism 2015:1-6.

96. Reed AM, Picornell YJ, Harwood A, Kredich DW. Chimerism in children with juvenile dermatomyositis. Lancet 2000; 356:2156-7.

97. Artlett CM, Ramos R, Jiminez SA, Patterson K, Miller FW, Rider LG. Chimeric cells of maternal origin in juvenile idiopathic inflammatory myopathies. Childhood Myositis Heterogeneity Collaborative Group. Lancet 2000; 356:2155-6.

98. Stevens AM, Hermes HM, Rutledge JC, Buyon JP, Nelson JL. Myocardial-tissue- specific phenotype of maternal microchimerism in neonatal lupus congenital heart block. Lancet 2003; 362:1617-23.

99. Suskind DL, Rosenthal P, Heyman MB, Kong D, Magrane G, Baxter-Lowe LA, Muench MO. Maternal microchimerism in the livers of patients with biliary atresia. BMC gastroenterology 2004; 4:14.

100. Gammill H, Nelson J. Naturally acquired microchimerism. Int J Dev Biol 2010; 54:531-43.

101. Nijagal A, Fleck S, Hills NK, Feng S, Tang Q, Kang SM, Rosenthal P, MacKenzie TC. Decreased risk of graft failure with maternal liver transplantation in patients with biliary atresia. American journal of transplantation : official journal of the American Society of Transplantation and the American Society of Transplant Surgeons 2012; 12:409-19.

102. Dutta P, Dart M, Roenneburg DA, Torrealba JR, Burlingham WJ. Pretransplant immune-regulation predicts allograft tolerance. American journal of transplantation : official journal of the American Society of Transplantation and the American Society of Transplant Surgeons 2011; 11:1296-301.

103. Maurel MC, Kanellopoulos-Langevin C. Heredity--venturing beyond genetics. Biology of reproduction 2008; 79:2-8.

78 Chapter 4: Offspring’s Tolerance of Mother Goes Viral

The primary focus of my dissertation has been investigating how immune tolerance to reproductive antigens encountered during pregnancy is achieved and sustained postpartum. Here we highlight a study that suggests a potential downside to favoring tolerance induction to foreign antigens in utero in the context of maternal hepatitis B infection that increases offspring susceptibility to persistent infection postnatally.

* This article is reprinted here with permission. The manuscript was originally published in 2016: Kinder J. 2016. Immunity 44(5):1085-7

79 An elaborate assortment of immunological shifts takes place during pregnancy to accommodate the intimate approximation of genetically discordant maternal and fetal tissues. While this is typically studied from the viewpoint of expanded maternal tolerance to foreign paternal antigens expressed by the developing fetus, the fetus is reciprocally exposed to an equally vast array of genetically foreign non-inherited maternal antigens in this unique developmental window. Given the predominant importance of reproduction for species survival, it is perhaps not surprising that many non-overlapping mechanisms are in place to sustain this bi-directional need for expanded immunological tolerance. However, this assortment of immunological shifts designed to ensure tolerance to foreign beneficial antigens can also create holes in host defense, which can be exploited by microbial pathogens to evade immunity allowing for potentially more severe or persistent infections. Does tolerance to innocuous noninherited maternal antigens in offspring extend to foreign microbes transmitted from mother? Could these tolerogenic pathways promote susceptibility in offspring to vertically transmitted microbial invaders? In this issue of Immunity, Tian and colleagues1 uncover one mechanism whereby perinatal exposure to hepatitis B virus (HBV) may potentiate persistent infection in offspring through immunological shifts driven by the conserved viral e antigen (HBeAg).

Acquisition of HBV in adults usually causes only short-lived acute infection. However, perinatally acquired HBV carries a 90% risk of persistent infection, which in large part, accounts for the majority of the estimated 250 million chronically infected individuals2.

To investigate whether this discordance in viral control stems from bystander immune

80 tolerance engendered during pregnancy, the authors developed an innovative preclinical approach for exposing offspring to HBV infected mothers. By mating female mice encoding HBV genomic DNA on only one of two homologous chromosomes with non-transgenic males, half of the offspring do not inherit HBV encoding DNA, but were exposed to a full complement of HBV antigens from their mothers during in utero development. Remarkably, prenatal exposure to HBV in this context made offspring drastically more susceptible to postnatal HBV infection, mimicked by injection of HBV genomic DNA – suggesting that in utero sensitization drives immunological shifts that favor long-term HBV persistence.

By contrast to suppressed responsiveness to non-inherited maternal cellular antigens that occurs with accumulation of immune-suppressive regulatory CD4+ T cells3,4, in utero sensitization to HBV in offspring was associated with diminished expansion of

HBV-specific CD8+ T cells, and their production of molecules associated with cytolytic activity such as IFN-γ and granzyme B. Functional quiescence of HBV-specific CD8+ T cells also paralleled sharply increased expression of PD-1, a molecule associated with an exhausted phenotype among T cells during persistent infection and cancer5.

Additionally, an increased frequency of hepatic macrophage cells, known as Kupffer cells, up-regulate expression of the primary PD-1 ligand, PD-L1. This interaction proved to be essential, because either blockade of PD-1/PD-L1 interactions with neutralizing antibody or bulk depletion of macrophages each efficiently overturned the susceptibility to persistent infection of offspring sensitized by in utero HBV exposure. Together these findings highlight how in utero sensitization with microbial antigens can drive drastic

81 shifts in the differentiation and co-stimulatory potential of macrophage cells within target tissues, which ultimately dictates the outcome of infection. Along with adding, “bad macrophage cells” to the list of potential therapeutic targets for re-invigorating the activation of exhausted CD8 T cells during persistent HBV infection, these results demonstrating functional exhaustion of microbe-specific CD8+ T cells controlled by tissue resident macrophage cells also adds an exciting new dimension whereby antigenic exposure in early development engenders immunological tolerance.

While this unfortunate outcome may reflect engagement of bystander tolerance in place to suppress harmful neonatal responses to genetically foreign maternal antigens, Tian and colleagues demonstrate a far more deliberate and sinister plot by showing HBV intentionally exploits these immune tolerance pathways for its benefit. Specifically, the authors show shifts in Kupffer cell polarization with in utero sensitization is mediated by the conserved HBV specific protein, HBeAg, whose biological role had been previously poorly defined. Susceptibility to persistent viral infection was overturned when HBeAg exposure was eliminated either during in utero sensitization or postnatal infection. In turn, protective immunity against persistent infection caused by HBV strains lacking only the HBeAg paralleled restored production of pro-inflammatory molecules (e.g., nitric oxide synthase, TNF-α, and IL-1β), with reciprocally blunted expression of anti- inflammatory molecules (e.g., arginase-1, mannose receptor-1, IL-10) by hepatic macrophage cells. Together, these results implicate the conservation of HBeAg is to precondition a more tolerant-inducing phenotype for neonatal macrophage cells that promotes persistent infection following perinatal HBV acquisition.

An interesting extension of these findings is the memory-like tolerant phenotype of classical innate immune components such as macrophage cells primed by HBeAg.

While the ability to “remember” has traditionally been thought to be a unique property of adaptive immune components (T and B cells), there is emerging evidence now that non- antigen-specific immune components can also be similarly “trained”6. This education or training of non-antigen-specific immune components is likely to play especially prominent roles during in utero development and early infancy when adaptive immune cells are under-developed or intentionally biased toward a more tolerant phenotype to handle the barrage of foreign innocuous maternal and commensal antigens6,7.

Therefore, to establish persistent infection, HBV has apparently learned how to co-op this tolerogenic phenotype of neonatal macrophage cells and their ability to be trained.

However, since susceptibility to persistent infection fades when HBV is administered to older mice, at time points further removed from in utero sensitization1, trained immunity in this context is also not permanent – potentially reflecting a physiological shift to other immune component layers with the progression of postnatal development8.

Furthermore, considering the efficiency whereby vaccination in the early newborn period protects against perinatal HBV transmission9, this tolerance may alternatively be disrupted with cognate antigen encounter in a highly inflammatory milieu. Nonetheless, these findings underscore the remarkable ability whereby prenatal stimulation can potently shape the functional repertoire of both antigen-specific adaptive and non- antigen-specific innate immune components.

Tian et al. highlight a mechanism whereby maternal HBV infection actively influences the development of immune cells within offspring to promote its persistence. Although it is clear HBV, through the conserved expression of HBeAg, has evolved to exploit these natural tolerogenic pathways, do other microbial pathogens take advantage of these mechanisms to promote persistence after congenital or perinatal infection? In addition to HBV, maternal infection with several other viral pathogens (e.g., herpes simplex virus, cytomegalovirus, rubella, HIV) causes high risk of vertical transmission or perinatal infection with often devastating consequences for the child10. Therefore, investigating whether other perinatally acquired pathogens exploit the crosstalk between polarized macrophage cells and functionally exhausted viral-specific CD8+ T cells described by Tian et al.1, or potentially other tolerance pathways activated during pregnancy, have exciting potential to spur the development of host-directed therapies to combat this wide diversity of microbes that selectively cause more severe or persistent infections when acquired in early infancy. In turn, while Tian and colleagues show disruption of the PD-1/PD-L1 interaction or depletion of macrophages each promotes the resolution of persistent HBV infection, these results also highlight the remarkable detrimental potential when naturally existing immune tolerance during pregnancy goes viral.

84 References

1 Tian, Y., Kuo, C. F., Akbari, O. & Ou, J. H. Hepatitis B virus persistence in offspring after vertical transmission is driven by macrophages that are altered by viral e antigen in mother. Immunity (2016).

2 Schweitzer, A., Horn, J., Mikolajczyk, R. T., Krause, G. & Ott, J. J. Estimations of worldwide prevalence of chronic hepatitis B virus infection: a systematic review of data published between 1965 and 2013. Lancet 386, 1546-1555, doi:10.1016/S0140-6736(15)61412-X (2015).

3 Kinder, J. M. et al. Cross-generational reproductive fitness enforced by microchimeric maternal cells. Cell 162, 505-515, doi:10.1016/j.cell.2015.07.006 (2015).

4 Mold, J. E. et al. Maternal alloantigens promote the development of tolerogenic fetal regulatory T cells in utero. Science 322, 1562-1565 (2008).

5 Pauken, K. E. & Wherry, E. J. Overcoming T cell exhaustion in infection and cancer. Trends Immunol 36, 265-276, doi:10.1016/j.it.2015.02.008 (2015).

6 Levy, O. & Wynn, J. L. A prime time for trained immunity: innate immune memory in newborns and infants. Neonatology 105, 136-141, doi:10.1159/000356035 (2014).

7 Elahi, S. et al. Immunosuppressive CD71+ erythroid cells compromise neonatal host defence against infection. Nature 504, 158-162, doi:10.1038/nature12675 (2013).

8 Mold, J. E. & McCune, J. M. Immunological tolerance during fetal development: from mouse to man. Adv. Immunol. 115, 73-111 (2012).

9 Poland, G. A. & Jacobson, R. M. Clinical practice: prevention of hepatitis B with the hepatitis B vaccine. The New England journal of medicine 351, 2832-2838, doi:10.1056/NEJMcp041507 (2004).

10 Silasi, M. et al. Viral infections during pregnancy. Am J Reprod Immunol 73, 199- 213, doi:10.1111/aji.12355 (2015).

In utero hepatitis B exposure No congenital exposure TNF-α Macrophage IL-10 Macrophage IL-1β iNOS Arg1 Mrc1 HBV HBV

pMHCI PD-L1 pMHCI PD-1 TCR TCR CD8+ T cell

CD8+ T cell IFN-γ

IFN-γ

Granzyme B Granzyme B

Persistent viral infection Self-resolving acute infection

Figure 1. Prenatal exposure to hepatitis B virus infected mothers promotes viral persistence in offspring. In utero hepatitis B virus exposure from infected mothers dampens the anti-viral response in offspring, resulting in persistent viral infection. This is accomplished by skewing neonatal macrophages towards a tolerant phenotype (IL-10 production, arginase-1 [Arg1] and mannose receptor 1 [Mrc1] expression) and expression of PD-L1 to maintain a functionally exhausted phenotype for HBV-specific CD8+ T cells (left-sided panel). By contrast, when congenital hepatitis B virus exposure is eliminated in genetically identical offspring, a robust innate inflammatory response and T-cell receptor (TCR) engagement by peptide + MHC (pMHCI) drives expansion of activated viral-specific CD8+ T cells resulting in self-resolving acute infection (right-sided panel).

Chapter 5: Cross-generational reproductive fitness enforced by microchimeric maternal cells

* This article is reprinted here with permission. The manuscript was originally published in 2015: Kinder J. 2015. Cell 162(3):505-15

SUMMARY

Compulsory exposure to maternal tissue during in utero development imprints persistent tolerance to immunologically foreign non-inherited maternal antigens (NIMA). Despite prolonged survival of NIMA-matched allografts after transplantation, the biological advantages preserving conserved postnatal tolerance to NIMA across mammalian species remain undefined. Here we show reproductive success in females is selectively enhanced during pregnancies sired by males expressing alloantigens with overlapping

NIMA specificity. Vertically transferred maternal cells that selectively establish microchimerism in female reproductive tissue drive expanded accumulation of immune suppressive regulatory T cells (Tregs) with NIMA specificity. In turn, NIMA-specific

Tregs avert fetal wastage triggered by prenatal infection and non-infectious disruptions in fetal tolerance. These findings demonstrate genetic fitness, canonically thought to be restricted to transmitting only half of homologous chromosomes by Mendelian inheritance, is enhanced in female placental mammals to also promote conservation of

NIMA through vertically transferred maternal cells that establish microchimerism and enforce cross-generational reproductive fitness.

HIGHLIGHTS

- Enhanced accumulation of NIMA-specific regulatory T cells in female offspring

- Microchimeric maternal cells drive postnatal persistence of NIMA-specific tolerance

- Fetal antigen stimulation during pregnancy accentuates NIMA-specific tolerance

- Cross-generational protection against fetal wastage enforced by NIMA tolerance

INTRODUCTION

Reproductive health and pregnancy outcomes have traditionally been characterized from the viewpoint of maternal tolerance to immunologically foreign paternal antigens expressed by the fetus 1,2. However, compulsory fetal exposure to an equally diverse array of discordant non-inherited maternal antigens (NIMA) also occurs during in utero and early postnatal maturation. Maternal antigen stimulation in these developmental contexts imprints remarkably persistent tolerance to NIMA in offspring 3-5. Pioneering examples of tolerance to NIMA include blunted sensitization to erythrocyte Rh antigen among Rh-negative women born to Rh-positive mothers 6, and selective anergy to

NIMA-specific HLA haplotypes among transfusion dependent individuals broadly exposed to foreign HLA 7. More recently, prolonged survival of NIMA-matched human allografts after solid organ transplantation 8, and reduced graft versus host disease among NIMA-matched stem cell transplants highlight clinical benefits of NIMA-specific tolerance that persists in individuals through adulthood 9-11.

In human development, tolerance to mother begins in utero with suppressed activation of maturing immune cells with NIMA specificity for infants with a full numerical complement of adaptive immune components at the time of birth 3,12. In this scenario, postnatal persistence of NIMA-specific tolerance represents an expendable developmental remnant of immune suppressive mechanisms essential for in utero survival. However, this reasoning does not explain why tolerance imprinted by exposure to foreign antigens in utero is widely conserved across mammalian species (e.g. non- human primates, ruminants, rodents) regardless of fetal adaptive immune cell maturation relative to parturition 8,13-16. For example, prolonged survival of NIMA-

matched allografts in humans is consistently reproduced in mice despite the absence of peripheral T cells at the time of birth in this species 3,17,18. These results illustrating highly engrained phylogenetic roots of NIMA tolerance in mammalian reproduction strongly suggest the existence of universal biological benefits driving conserved tolerance to NIMA that persists through adulthood.

Given the necessity for sustained maternal tolerance to foreign fetal antigens in successful pregnancies across all eutherian placental mammals 19, postnatal NIMA- specific tolerance may be evolutionarily preserved to promote reproductive fitness by reinforcing fetal tolerance in future generation pregnancies. To address this hypothesis, immunological tools that allow precise identification of T cells with NIMA-specificity were uniquely combined with mouse models of allogeneic pregnancy, and pregnancy complications stemming from disruptions in fetal tolerance 20-22. Our data show obligatory developmental exposure to foreign maternal tissue primes expanded accumulation of NIMA-specific immune suppressive regulatory CD4+ T cells (Tregs) that reinforce fetal tolerance during next-generation pregnancies sired by males with overlapping MHC haplotype specificity. Expanded NIMA-specific Treg accumulation requires ongoing postnatal cognate antigen stimulation by maternal cells that establish microchimerism in offspring. In the broader context, cross-generational reproductive benefits conferred by tolerance to NIMA indicates genetic fitness is not restricted only to transmitting homologous chromosomes by Mendelian inheritance, but is enhanced through vertically transferred tolerogenic cells that establish microchimerism in offspring favoring preservation of non-inherited maternal alleles within a population.

RESULTS

Developmental exposure to maternal tissue drives expanded NIMA-specific regulatory T cell accumulation

To investigate the fundamental biology driving conserved persistence of NIMA tolerance across mammalian species, an instructive allogeneic mating strategy that transforms defined model antigens into surrogate NIMA was developed to precisely track T cells with NIMA specificity. Female mice heterozygous for a transgene encoding constitutive expression of a transmembrane recombinant protein containing ovalbumin (OVA) and

23,24 the 2W1S55-68 variant of I-Eα in all cells (behind the β-actin promoter) were mated with non-transgenic males – thereby transforming 2W1S55-68 and OVA into surrogate

NIMA in half the offspring (Figures S1A and S1B). This approach taking advantage of

MHC tetramer staining and enrichment techniques for identifying endogenous CD4+ T

b 25 cells with I-A :2W1S55-68 specificity , combined with tools for manipulating OVA- expressing cells allows NIMA-responsive and NIMA-expressing cells to be

b + simultaneously evaluated. To ensure shifts in I-A :2W1S55-68 specific CD4 T cells reflect developmental exposure to maternal tissue as opposed to 2W1S-OVA+ concepti within the same litter, offspring from reciprocal mating between males heterozygous for the 2W1S-OVA expression transgene and non-transgenic females that transforms

2W1S55-68 peptide and OVA into surrogate non-inherited paternal antigens (NIPA) were used as controls along with genetically identical naive mice without developmental

2W1S exposure (Figures S1C and S1D).

Sharply increased proportions of immune suppressive regulatory T cells (Tregs)

identified by expression of the lineage defining FOXP3 transcriptional regulator 26,27

+ b were found among CD4 T cells with I-A :2W1S55-68 specificity in adult NIMA-2W1S mice compared with age matched naive mice, and additional control mice exposed to the identical 2W1S-OVA recombinant protein as a surrogate NIPA or ubiquitous ‘self’ antigen (Figures 1A and S1). Interestingly, while the percentage and number of FOXP3+

b Tregs with I-A :2W1S55-68 specificity were significantly increased in the spleen and

b peripheral lymph nodes for NIMA-2W1S offspring, the total number of I-A :2W1S55-68- specific CD4+ T cells remained similar regardless of developmental 2W1S stimulation

(Figure 1B). Along with sharply reduced expression of Helios that marks thymus derived

b Tregs among cells with I-A :2W1S55-68 specificity in NIMA-2W1S compared with each group of control mice 28 (Figure 1C), these results strongly suggest developmental exposure to immunologically discordant maternal tissue primes induced FOXP3 expression among NIMA-specific CD4+ T cells. Importantly, these shifts were restricted to CD4+ T cells with NIMA-specificity since expanded Tregs and diminished Helios expression were eliminated among bulk CD4+ T cells in each group of mice regardless of developmental 2W1S stimulation (Figures 1A and 1C).

Considering exposure to maternal tissue begins in utero when fetal immune components are undergoing maturation, related experiments addressed whether expanded NIMA-specific Tregs require antigen presentation by maternal cells. Here, the

b I-A restricted nature of 2W1S55-68 peptide presentation was exploited to compare

NIMA-specific Tregs in genetically identical NIMA-2W1S offspring born to I-Ab/b or I-Ad/d mothers 24 (Figure 1D). Interestingly, expanded proportions of FOXP3+ Tregs and diminished cell-intrinsic Helios expression were each significantly reduced for I-

b + d/d A :2W1S55-68 specific CD4 T cells in NIMA-2W1S offspring born to I-A mothers to levels comparable to naive control mice (Figure 1D). The results highlighting tolerogenic properties of maternal cells in mouse offspring parallel the presence of maternal hematopoietic cells in human fetal lymph nodes during in utero development 12, and suggest compulsory developmental exposure to immunologically foreign maternal tissue actively primes expansion of immune suppressive Tregs with NIMA specificity.

Expanded NIMA-specific Treg accumulation may reflect in utero and/or postnatal exposure to immunologically foreign maternal antigen (e.g. soluble maternal HLA alloantigens, intact maternal cells in breast milk) 29,30. To dissociate the contributions of maternal antigen stimulation during each developmental context, the individual impacts of in utero and early postnatal NIMA exposure through breastfeeding were evaluated by cross-fostering offspring after birth with naive or 2W1S-OVA+ nursing mothers. In agreement with improved survival of NIMA-matched allografts in transplant recipients exposed to maternal antigen both in utero and through breastfeeding 17,31, maximal

NIMA-specific Treg expansion required in utero plus postnatal maternal antigen stimulation (Figure 1E). Comparatively, Helios expression remained at diminished levels with maternal tissue exposure in utero or through breastfeeding suggesting maternal antigen stimulation in either developmental context primes enriched proportions of

NIMA-specific CD4+ T cells poised for induced FOXP3 expression (Figure 1E). Taken together, these findings indicate immunologically foreign maternal antigen stimulation in utero and through breastfeeding work synergistically to promote expanded peripheral accumulation of NIMA-specific Tregs.

Microchimeric maternal cells maintain expanded NIMA-specific Tregs

Postnatal persistence of NIMA-specific tolerance coincides with long-term retention of microchimeric maternal cells in adult human and rodent offspring 4,12,32,33. Nonetheless, the immunological cause and effect relationship between these two interrelated phenomena engrained in mammalian reproduction remain undefined. One possibility is that postnatal maintenance of NIMA-specific tolerance actively prevents rejection of antigenically discordant maternal cells fostering their long-term survival in offspring.

Alternatively, retained microchimeric maternal cells may provide an essential postnatal source of cognate antigen required for maintaining expanded Tregs with NIMA- specificity. Having established early developmental exposure to 2W1S-OVA+ maternal tissues imprints persistent accumulation of NIMA-2W1S-specific Tregs, analysis of cells expressing these model antigens was extended to investigate the necessity for postnatal stimulation by microchimeric 2W1S-OVA+ maternal cells. We found OVA- encoding DNA, reflective of 2W1S-OVA+ maternal cells in systemic organs (e.g. heart, liver), at levels ranging from 1 in 105 to 106 cells in NIMA offspring consistent with quantities of microchimeric maternal cells identified using PCR for MHC haplotype alleles 4,12,34 (Figures 2A and S2A).

To definitively address the cause and effect relationship between NIMA-specific tolerance and microchimeric maternal cells, anti-OVA antibody that uniformly binds

2W1S-OVA+ cells was used to deplete microchimeric 2W1S-OVA+ maternal cells

(Figure 2B). In line with the efficiency whereby anti-OVA antibody depletes congenically marked 2W1S-OVA+ cells after adoptive transfer into non-transgenic recipients (Figure

S2B), OVA encoding DNA representative of microchimeric 2W1S-OVA+ cells in each

organ of NIMA offspring declined sharply within 12 days following in vivo anti-OVA antibody administration (Figure 2A). Remarkably, NIMA-2W1S-specific Treg accumulation and cell-intrinsic Helios down-regulation both returned to background levels found in naive control mice after elimination of microchimeric 2W1S-OVA+ cells within this time frame (Figure 2C). Thus, microchimeric maternal cells provide an essential source of cognate maternal antigen required for sustaining postnatal NIMA- specific tolerance.

Selectively enriched NIMA-specific Treg expansion in female offspring accentuated during pregnancy with NIMA-matched fetal antigen stimulation

Given the necessity for expanded maternal tolerance that encompasses immunologically foreign paternal-fetal antigens in successful pregnancy shared by all eutherian placental mammals 1,2,19, we reasoned reinforced fetal tolerance that promotes reproductive fitness may represent a more universal evolutionary driver for conserved NIMA-specific tolerance. This notion is supported by highly enriched microchimeric 2W1S-OVA+ maternal cells in female reproductive tissue (uterus) of

NIMA offspring, and their conspicuous absence in analogous male reproductive tissue

(prostate) (Figure 3A). In turn, NIMA-specific Treg expansion and reduced Helios expression were markedly more pronounced in female compared with male NIMA-

2W1S littermate offspring (Figure 3B). Thus, gender-specific differences favoring more robust NIMA-specific Treg expansion in females parallel the selective accumulation of microchimeric maternal cell in female reproductive tissue.

To further investigate the reproductive significance for gender-specific differences in postnatal persistence of NIMA-specific tolerance, shifts in NIMA-specific CD4+ T cells were evaluated in females during pregnancy after cognate fetal antigen stimulation.

During allogeneic pregnancies sired by 2W1S-OVA+ transgenic males, sharply accelerated expansion tempo occurred among 2W1S-specific Tregs in NIMA-2W1S female mice compared with naive control mice (7.2-fold compared with 3.4-fold expansion by midgestation in NIMA and naive mice, respectively [P = 0.004]) (Figure 4).

Accelerated NIMA-specific Treg expansion tempo during pregnancy represents a targeted response to cognate 2W1S stimulation by shared fetal-expressed antigen because NIMA-2W1S-specific Tregs did not expand during pregnancies sired by non- transgenic male mice (Figure 4). Thus, mammalian females contain an enriched pool of

NIMA-specific Tregs poised for accelerated re-expansion upon encounter with paternal- fetal antigen of overlapping specificity during pregnancy.

Microchimeric maternal cells enforce cross-generational protection against fetal wastage

Since the immunological identity of individuals is primarily defined by unique expression of MHC haplotype alleles 35, MHC haplotype alleles (e.g. H-2d, H-2k –along with 2W1S and OVA antigens), were transformed into surrogate NIMA to investigate functional properties of tolerance in the setting of broader NIMA overlap (Figure S3). In turn, the protective properties of NIMA-specific tolerance were probed by infection with the prenatal bacterial pathogen, Listeria monocytogenes, which disrupts fetal tolerance with

ensuing fetal wastage 21,36,37. Remarkably, fetal resorption and in utero L. monocytogenes invasion after prenatal infection in naive mice bearing allogeneic pregnancy were eliminated by overlap between NIMA and paternal-fetal MHC haplotype antigens (NIMA H-2d females mated with H-2d Balb/c male mice) (Figures 5A and S3).

Protection against fetal wastage occurred in an antigen-specific fashion requiring commonality between NIMA and paternal-fetal antigens since fetal resorption and in utero bacterial invasion each rebounded when third-party males bearing irrelevant MHC haplotype alleles (e.g. H-2k CBA/J mice) were used to sire allogeneic pregnancy in

NIMA H-2d female mice (Figures 5A and S3). Thus, protection against prenatal infection conferred by non-inherited antigenic overlap between maternal grandmother and the developing fetus and highlight profound cross-generational benefits of persistent NIMA- specific tolerance.

Given the requirement for postnatal maternal microchimerism to maintain expanded

NIMA-specific tolerance (Figure 2), we next addressed the necessity for microchimeric maternal cells to protect against fetal wastage after prenatal L. monocytogenes infection.

Here, cross-reactivity between anti-OVA antibody used to deplete H-2d-2W1S-OVA+ microchimeric maternal cells and fetal-expressed OVA antigen was avoided by exclusively using non-transgenic H-2d Balb/c male mice to sire allogeneic pregnancy in

H-2d-2W1S-OVA NIMA female mice (Figure S3). This analysis showed depletion of microchimeric maternal cells prior to mating efficiently overturns protection against fetal resorption and in utero L. monocytogenes invasion in NIMA female mice (Figure 5A).

Importantly, these reproductive benefits were not restricted to NIMA H-2d haplotype alleles since fetal resorption and in utero L. monocytogenes invasion were each

similarly averted among NIMA H-2k female mice during allogeneic pregnancy sired by

H-2k CBA/J male mice (Figure 5B). Conversely, protection against fetal wastage was lost in NIMA H-2k female mice if NIMA mismatched H-2d males were used to sire allogeneic pregnancy, or if H-2k-2W1S-OVA+ microchimeric maternal cells were depleted using anti-OVA antibody prior to mating with non-transgenic H-2k CBA/J male mice (Figure 5B). Thus, persistent postnatal tolerance to NIMA protects against fetal wastage triggered by prenatal infection.

To further extend this analysis to non-infectious disruptions in fetal tolerance stemming from blunted expansion of maternal Tregs (e.g. spontaneous abortion, preeclampsia) 38-

40, the protective benefits of NIMA-specific tolerance on fetal wastage triggered by partial depletion of bulk maternal FOXP3+ Tregs during allogeneic pregnancy were investigated. Diphtheria toxin administration to female mice heterozygous for co- expression of the high-affinity human diphtheria toxin receptor (DTR) with FOXP3 during pregnancy causes partial transient depletion of bulk maternal Tregs to levels comparable to virgin control mice with disruptions in fetal tolerance and ensuing fetal wastage 20,22,41. Therefore, our breeding strategy was modified to transform MHC haplotype alleles (e.g. H-2d, H-2k) along with 2W1S-OVA into surrogate NIMA in genetically identical H-2b FOXP3DTR/WT female mice (Figure S4). Remarkably, fetal resorption triggered by partial depletion of bulk maternal Tregs during allogeneic pregnancy in naive control female mice was reversed to near completion by overlap between NIMA and fetal expressed MHC haplotype antigens (NIMA H-2d females mated with H-2d Balb/c male mice, or NIMA H-2k females mated with H-2k CBA/J male mice) (Figures 6A and 6B). Protection against fetal wastage induced by partial maternal

Treg depletion was paternal-antigen specific and required ongoing stimulation by microchimeric maternal cells since fetal resorption rebounded during pregnancies sired by third-party males that express irrelevant MHC haplotype alleles or if 2W1S-OVA+ microchimeric maternal cells were depleted prior to mating (Figures 6A and 6B). Taken together, these findings demonstrate resiliency against fetal wastage conferred by persistent postnatal NIMA-specific tolerance encompasses both infectious and non- infectious perturbations in fetal tolerance during next generation pregnancies.

DISCUSSION

Reproductive success in female placental mammals requires sustained maternal tolerance to immunologically foreign paternal antigens expressed by the developing fetus 1,2,19. Reciprocally, disruptions in fetal tolerance are increasingly recognized in human cases of spontaneous abortion, preeclampsia and prematurity 38-40,42. In humans, over 10% of pregnancies are afflicted by these complications linked with disrupted fetal tolerance 43,44. Given this sustained backdrop of refining selection that likely occurs across all outbred mammalian species, conservation of phenotypic traits that improve reproductive fitness is a biological imperative.

Herein, this reasoning was applied to investigate the ontological conservation of NIMA- specific tolerance and maternal cell microchimerism across placental mammalian species 16,17,34,45. Using mice with defined MHC haplotype alleles in multi-generational breeding that transforms MHC haplotype alleles into surrogate NIMA, we show sharply increased resiliency against fetal wastage in the presence of overlap between NIMA and paternal-fetal antigen encountered during next generation pregnancies. Cross- generational reproductive benefits conferred by NIMA-specific tolerance shown here for mice are consistent with pioneering observations of reduced erythrocyte Rh antigen sensitization among Rh-negative women born to Rh-positive mothers 6. However, while developmental exposure to this single minor alloantigen does not prevent hemolytic disease of the newborn 6,46, we find broader non-inherited antigenic overlap between maternal grandmother and offspring that encompasses MHC haplotype alleles efficiently protects against fetal wastage (Figure 7). By establishing clear reproductive benefits for NIMA-specific tolerance applicable to all placental mammalian species,

100

these results highlight broad evolutionary advantages for persistent postnatal NIMA- specific tolerance beyond averting anti-maternal immunity for human and other species with comparatively more developed fetal adaptive immune components at the time of birth 3,12.

Dissecting the mechanistic relationship between NIMA-specific tolerance and microchimeric maternal cells that both persist in offspring through adulthood requires strategies for precisely identifying NIMA-specific immune components along with manipulation of microchimeric maternal cells. Prior limitations restricting these analyses were simultaneously bypassed by transforming defined model antigens into surrogate

NIMA using female mice heterozygous for a transgene encoding constitutive expression of model antigens for breeding with non-transgenic males (Figure S1). Using antigen- specific tools, endogenous CD4+ T cells with surrogate NIMA specificity were shown to be highly enriched for expression of the Treg lineage defining transcriptional regulator,

FOXP3 26,27. By establishing NIMA specificity for this essential immune regulatory CD4+

T cell subset, these results extend previously described reversal of NIMA-specific tolerance by depleting of bulk CD4+ T cells or Tregs 11,12,18,47. In turn, protection against fetal wastage conferred by expanded Tregs with shared NIMA plus fetal specificity also reinforce beneficial properties of expanded maternal Tregs with pre-exiting fetal specificity retained after prior pregnancy in partner specific protection against complications in subsequent pregnancy 22,48,49.

More importantly, the concurrent ability to deplete maternal cells retained in offspring allowed us to definitively establish the causative relationship between microchimeric maternal cells and postnatal persistence of NIMA-specific tolerance. Similar to the

101

necessity of low-level exposure to cognate antigen in numerical maintenance of

‘memory’ effector CD4+ T cells with foreign microbial specificity 50-52, sustained expansion of NIMA-specific FOXP3+ CD4+ T cells also requires postnatal exposure to cognate maternal antigen expressed by microchimeric maternal cells. Reciprocally, in vivo depletion of microchimeric maternal cells efficiently overturned both expanded

NIMA-specific Treg accumulation and protection against fetal wastage. Thus, vertically transferred maternal cells that establish microchimerism in offspring promote cross- generational reproductive fitness by preserving tolerance to NIMA along with noninherited genetic alleles within a population (Figure 7).

In the broader context, these results indicate genetic fitness, canonically thought to be restricted to transmitting only half of homologous chromosomes through Mendelian inheritance, is enhanced in female placental mammals to also promote conservation of non-inherited antigens by vertical transmission of tolerogenic maternal cells that establish microchimerism in offspring. However in nature, this engrained drive for genetic fitness in each individual is likely counterbalanced by pathogen-mediated selection for MHC diversity across the entire population 53. Nonetheless, our findings suggest more extended cross-generational analysis will illuminate the ongoing controversy regarding how MHC haplotype similarity impacts mate selection and pregnancy outcomes 54-56. Finally, reproductive advantages actively maintained by tolerogenic microchimeric maternal cells underscore the need for renewed consideration of immune tolerance from the intriguing perspective of constitutive chimerism beyond engrained pillars of “self” versus “non-self” antigen distinction defined using genetically homogenous inbred mice that artificially eliminates cross-generational

102

tolerance encountered among outbred populations57-59.

103

EXPERIMENTAL PROCEDURES

Mice. C57BL/6 (H-2b; CD45.2+), Balb/c (H-2d), CBA/J (H-2k), B6.C-H2d/bByJ (H-2d),

B6.Ak-H2k/J (H-2k), and B6.SJL-PtprcaPepcb/BoyJ (H-2b; CD45.1+) mice were purchased from The Jackson Laboratory. 2W1S-OVA+ transgenic mice that constitutively express recombinant 2W1S55-68-OVA protein behind the β-actin promoter, and FOXP3DTR/DTR mice where FOXP3+ cells are susceptible to diphtheria toxin induced ablation have each been described 23,41. 2W1S-OVA+ mice were maintained on either the C57BL/6 or Balb/c strain backgrounds after backcrossing for >10 generations. For cross-fostering, pregnant mice were checked twice daily for birth timing, and newborn offspring introduced to lactating foster mothers within 12 hours after birth, with weaning

21 days thereafter and analysis at 8 weeks of age. For partial transient maternal Treg depletion, FOXP3DTR/WT pregnant females were administered purified diphtheria toxin daily (Sigma-Aldrich, USA) (0.5 µg first dose, followed by 0.1 µg/dose) beginning midgestation (E11.5) for 5 consecutive days, and the frequency of fetal resorption evaluated E16.5. All experiments were performed using sex and aged matched controls under Cincinnati Children’s Hospital IACUC approved protocols.

Tetramer enrichment and flow cytometry. Cell surface staining with phycoerythrin

b (PE)-conjugated MHC class II I-A :2W1S55-68 tetramer followed by enrichment using anti-PE-conjugated magnetic beads (Miltenyi Biotec) have been described 22,25,60. To identify CD4+ T cells with I-Ab:2W1S specificity, cells in secondary lymphoid tissue

(spleen plus axillary, brachial, cervical, inguinal, mesenteric, pancreatic, para- aortic/uterine lymph nodes) of each mouse were combined, enriched with PE

104

b conjugated I-A :2W1S55-68 tetramer, and stained for cell-surface CD4 (GK1.5), CD8α

(53-7.3), CD25 (PC61), CD44 (IM7), CD11b (M1/70), CD11c (N418), B220 (RA3-B62),

F4/80 (BM8), along with intranuclear FOXP3 (FJK-16s) or Helios (22F6) expression using commercially available antibodies and cell permeabilization reagents (BD

PharMingen or eBioscience). For cell surface ovalbumin expression, cells were stained initially with polyclonal rabbit α-OVA (EMP Millipore) or IgG isotype antibodies followed by secondary staining with PE conjugated anti-rabbit IgG (eBioscience) antibody. Cells stained with fluorochrome-conjugated tetramer and/or antibody were acquired using a

FACSCanto cytometer (Becton Dickinson), and analyzed using FlowJo (TreeStar) software.

Bacteria. For infection, Listeria monocytogenes (wildtype strain 10403s) was grown to early log phase (OD600 0.1) in brain heart infusion media at 37ºC, washed and diluted with sterile saline, and inoculated intravenously via the lateral tail vein (104 CFUs) at midgestation (E11.5) as described 20,21. The inoculum for each experiment was confirmed by spreading diluted aliquots onto agar plates. Five days thereafter, fetal resorption and in utero bacteria invasion was evaluated by sterilely dissecting each concepti, homogenization in sterile saline containing 0.05% Triton X-100 to release intracellular bacteria, plating serial dilutions of each concepti homogenate onto agar plates, and enumeration after incubation at 37ºC for 24 hours.

DNA extraction and quantitative PCR. The heart, liver, uterus or prostate was sterilely dissected, and DNA extracted from each tissue using the QIAamp DNA extraction kit

(Qiagen). Thereafter, PCR for enumerating 2W1S-OVA+ DNA was performed in 20 separate wells per tissue each containing 333 ng genomic DNA (~3.33 x 105 cells) in 20

105

µl total volume supplemented with 10 µl Taqman Gene Expression Master Mix and 1 µl ovalbumin Taqman assay (Applied Biosystems), for a detection limit of ~1 in 6.66 x 106 cells per tissue. Amplification was performed using the 7500 Fast Real-Time PCR

System (Life Technologies) under the following program: 95°C for 10 minutes, followed by forty cycles of 95°C for 15 seconds and 60°C for 1 minute. For generating standard curve for 2W1S-OVA+ DNA, DNA from 2W1S-OVA+ splenocytes or C57BL/6 control mice were isolated, and combined with six serial 10-fold dilutions (10-1 to 10-6) of 2W1S-

OVA+ DNA into C57BL/6 control DNA so that the DNA concentration remained identical in each well (333 ng total DNA in 20 µl). The resulting linear regression equation y = -

1.137ln(x) + 38.443 (R2 = 0.986) was used to calculate the amount of 2W1S-OVA+ DNA in each tissue sample.

Depletion of microchimeric 2W1S-OVA+ maternal cells. To deplete 2W1S-OVA+ cells, 2W1S-NIMA mice were administered 650 µg purified rabbit α-OVA antibody (EMP

Millipore) or IgG isotype antibody (Sigma-Aldrich) by intraperitoneal injection, followed

10 days later by a second treatment with 325 µg of the same antibody. Two days after the second antibody inoculation, the level of 2W1S-OVA+ cells in each tissue was analyzed by quantitative real-time PCR, antigen-specific CD4+ T cells investigated using

b d k I-A :2W1S55-68 tetramer staining, or used for mating with H-2 Balb/c or H-2 CBA/J males to investigate pregnancy outcomes.

Statistical analysis. Where applicable, NIMA mice in each group were randomized for either administration of anti-OVA or isotype antibody, or for breeding with either NIMA- matched or NIMA-discordant MHC haplotype males. Considering data sets did not consistently show a normal distribution, differences between groups were analyzed

106

using the Mann-Whitney non-parametric test (Prism, GraphPad); and P < 0.05 was taken as statistical significance.

AUTHOR CONTRIBUTIONS

J.K., T.J., J.E., L.X., and B.S. performed the experiments. All authors participated in the experimental design and data analysis. J.K. and S.S.W. wrote the manuscript with editorial input from all the authors.

ACKNOWLEDGEMENTS

We thank Dr. Marc Jenkins for providing 2W1S-OVA transgenic mice; and Drs. Joseph

Qualls, James Moon, Louis Muglia, Harinder Singh, Anne Stevens, Kevin Urdahl for helpful discussions. This work as supported by the NIH-NIAID through awards R01-

AI100934 and R21-AI112186 (to SSW), and NIH-NHLBI through award R01-HL103745

(to AFS). SSW holds an Investigator in the Pathogenesis of Infectious Disease award from the Burroughs Wellcome Fund.

107

REFERENCES

1 Erlebacher, A. Mechanisms of T cell tolerance towards the allogeneic fetus. Nature reviews. Immunology 13, 23-33, doi:10.1038/nri3361 (2013).

2 Munoz-Suano, A., Hamilton, A. B. & Betz, A. G. Gimme shelter: the immune system during pregnancy. Immunol Rev 241, 20-38, doi:10.1111/j.1600- 065X.2011.01002.x (2011).

3 Mold, J. E. & McCune, J. M. Immunological tolerance during fetal development: from mouse to man. Adv. Immunol. 115, 73-111 (2012).

4 Dutta, P. et al. Microchimerism is strongly correlated with tolerance to noninherited maternal antigens in mice. Blood 114, 3578-3587 (2009).

5 Hirayama, M., Azuma, E. & Komada, Y. Tolerogenic effect of non-inherited maternal antigens in hematopoietic stem cell transplantation. Front Immunol 3, 135, doi:10.3389/fimmu.2012.00135 (2012).

6 Owen, R. D., Wood, H. R., Foord, A. G., Sturgeon, P. & Baldwin, L. G. Evidence for actively acquired tolerance to Rh antigens. Proc. Natl. Acad. Sci. U.S.A. 40, 420-424 (1954).

7 Claas, F. H., Gijbels, Y., van der Velden-de Munck, J. & van Rood, J. J. Induction of B cell unresponsiveness to noninherited maternal HLA antigens during fetal life. Science 241, 1815-1817 (1988).

8 Burlingham, W. J. et al. The effect of tolerance to noninherited maternal HLA antigens on the survival of renal transplants from sibling donors. N. Engl. J. Med. 339, 1657-1664 (1998).

9 Ichinohe, T. et al. Feasibility of HLA-haploidentical hematopoietic stem cell transplantation between noninherited maternal antigen (NIMA)-mismatched family members linked with long-term fetomaternal microchimerism. Blood 104, 3821-3828 (2004).

10 van Rood, J. J. et al. Effect of tolerance to noninherited maternal antigens on the occurrence of graft-versus-host disease after bone marrow transplantation from a parent or an HLA-haploidentical sibling. Blood 99, 1572-1577 (2002).

11 Matsuoka, K. et al. Fetal tolerance to maternal antigens improves the outcome of allogeneic bone marrow transplantation by a CD4+ CD25+ T-cell-dependent mechanism. Blood 107, 404-409, doi:10.1182/blood-2005-07-3045 (2006).

12 Mold, J. E. et al. Maternal alloantigens promote the development of tolerogenic fetal regulatory T cells in utero. Science 322, 1562-1565 (2008).

108

13 Billingham, R. E., Brent, L. & Medawar, P. B. Actively acquired tolerance of foreign cells. Nature 172, 603-606 (1953).

14 Owen, R. D. Immunogenetic Consequences of Vascular Anastomoses between Bovine Twins. Science 102, 400-401 (1945).

15 Picus, J., Aldrich, W. R. & Letvin, N. L. A naturally occurring bone-marrow- chimeric primate. I. Integrity of its immune system. Transplantation 39, 297-303 (1985).

16 Dutta, P. & Burlingham, W. J. Microchimerism: tolerance vs. sensitization. Current opinion in organ transplantation 16, 359-365, doi:10.1097/MOT.0b013e3283484b57 (2011).

17 Andrassy, J. et al. Tolerance to noninherited maternal MHC antigens in mice. Journal of immunology 171, 5554-5561 (2003).

18 Akiyama, Y. et al. Transplantation tolerance to a single noninherited MHC class I maternal alloantigen studied in a TCR-transgenic mouse model. Journal of immunology 186, 1442-1449, doi:10.4049/jimmunol.1003023 (2011).

19 Samstein, R. M., Josefowicz, S. Z., Arvey, A., Treuting, P. M. & Rudensky, A. Y. Extrathymic generation of regulatory T cells in placental mammals mitigates maternal-fetal conflict. Cell 150, 29-38, doi:10.1016/j.cell.2012.05.031 (2012).

20 Rowe, J. H., Ertelt, J. M., Aguilera, M. N., Farrar, M. A. & Way, S. S. Foxp3(+) regulatory T cell expansion required for sustaining pregnancy compromises host defense against prenatal bacterial pathogens. Cell Host Microbe 10, 54-64 (2011).

21 Chaturvedi, V. et al. CXCR3 blockade protects against Listeria monocytogenes infection induced fetal wastage. Journal of Clinical Investigation [In Press] (2015).

22 Rowe, J. H., Ertelt, J. M., Xin, L. & Way, S. S. Pregnancy imprints regulatory memory that sustains anergy to fetal antigen. Nature 490, 102-106 (2012).

23 Moon, J. J. et al. Quantitative impact of thymic selection on Foxp3+ and Foxp3- subsets of self-peptide/MHC class II-specific CD4+ T cells. Proc. Natl. Acad. Sci. U.S.A. 108, 14602-14607 (2011).

24 Rees, W. et al. An inverse relationship between T cell receptor affinity and antigen dose during CD4(+) T cell responses in vivo and in vitro. Proc. Natl. Acad. Sci. U.S.A. 96, 9781-9786 (1999).

25 Moon, J. et al. Naive CD4(+) T cell frequency varies for different epitopes and predicts repertoire diversity and response magnitude. Immunity 27, 203-213 (2007).

109

26 Fontenot, J. D., Gavin, M. A. & Rudensky, A. Y. Foxp3 programs the development and function of CD4+CD25+ regulatory T cells. Nat Immunol 4, 330-336, doi:10.1038/ni904 (2003).

27 Hori, S., Nomura, T. & Sakaguchi, S. Control of regulatory T cell development by the transcription factor FoxP3. Science 299, 1057-1061 (2003).

28 Thornton, A. M. et al. Expression of Helios, an Ikaros transcription factor family member, differentiates thymic-derived from peripherally induced Foxp3+ T regulatory cells. Journal of immunology 184, 3433-3441 (2010).

29 Molitor, M. L., Haynes, L. D., Jankowska-Gan, E., Mulder, A. & Burlingham, W. J. HLA class I noninherited maternal antigens in cord blood and breast milk. Hum. Immunol. 65, 231-239 (2004).

30 Zhou, L. et al. Two independent pathways of maternal cell transmission to offspring: through placenta during pregnancy and by breast-feeding after birth. Immunology 101, 570-580 (2000).

31 Campbell, D. A., Jr. et al. Breast feeding and maternal-donor renal allografts. Possibly the original donor-specific transfusion. Transplantation 37, 340-344 (1984).

32 Maloney, S. et al. Microchimerism of maternal origin persists into adult life. The Journal of clinical investigation 104, 41-47, doi:10.1172/JCI6611 (1999).

33 Loubiere, L. S. et al. Maternal microchimerism in healthy adults in lymphocytes, monocyte/macrophages and NK cells. Lab Invest 86, 1185-1192, doi:10.1038/labinvest.3700471 (2006).

34 Bakkour, S. et al. Analysis of maternal microchimerism in rhesus monkeys (Macaca mulatta) using real-time quantitative PCR amplification of MHC polymorphisms. Chimerism 5, 6-15 (2014).

35 Zinkernagel, R. M. & Doherty, P. C. MHC-restricted cytotoxic T cells: studies on the biological role of polymorphic major transplantation antigens determining T- cell restriction-specificity, function, and responsiveness. Advances in immunology 27, 51-177 (1979).

36 Mylonakis, E., Paliou, M., Hohmann, E. L., Calderwood, S. B. & Wing, E. J. Listeriosis during pregnancy: a case series and review of 222 cases. Medicine 81, 260-269 (2002).

37 Rowe, J. H., Ertelt, J. M., Xin, L. & Way, S. S. Listeria monocytogenes cytoplasmic entry induces fetal wastage by disrupting maternal FoxP3+ regulatory cell-sustained fetal tolerance. PLoS Pathogens 8 (2012).

110

38 Jiang, T. T. et al. Regulatory T cells: new keys for further unlocking the enigma of fetal tolerance and pregnancy complications. Journal of immunology 192, 4949- 4956, doi:10.4049/jimmunol.1400498 (2014).

39 Santner-Nanan, B. et al. Systemic increase in the ratio between Foxp3+ and IL- 17-producing CD4+ T cells in healthy pregnancy but not in preeclampsia. Journal of immunology 183, 7023-7030 (2009).

40 Sasaki, Y. et al. Decidual and peripheral blood CD4+CD25+ regulatory T cells in early pregnancy subjects and spontaneous abortion cases. Mol Hu Reprod 10, 347-353 (2004).

41 Kim, J. M., Rasmussen, J. P. & Rudensky, A. Y. Regulatory T cells prevent catastrophic autoimmunity throughout the lifespan of mice. Nat Immunol 8, 191- 197 (2007).

42 Sasaki, Y. et al. Proportion of peripheral blood and decidual CD4+ CD25(bright) regulatory T cells in pre-eclampsia. Clinical and experimental immunology 149, 139-145 (2007).

43 Blencowe, H. et al. National, regional, and worldwide estimates of preterm birth rates in the year 2010 with time trends since 1990 for selected countries: a systematic analysis and implications. Lancet 379, 2162-2172, doi:10.1016/S0140-6736(12)60820-4 (2012).

44 Duley, L. The global impact of pre-eclampsia and eclampsia. Seminars in perinatology 33, 130-137, doi:10.1053/j.semperi.2009.02.010 (2009).

45 Gammill, H. S., Stephenson, M. D., Aydelotte, T. M. & Nelson, J. L. Microchimerism in women with recurrent miscarriage. Chimerism, 0, doi:10.1080/19381956.2015.1017241 (2015).

46 Booth, P. B., Dunsford, I., Grant, J. & Murray, S. Haemolytic disease in first-born infants. British medical journal 2, 41-42 (1953).

47 Molitor-Dart, M. L. et al. Developmental exposure to noninherited maternal antigens induces CD4+ T regulatory cells: relevance to mechanism of heart allograft tolerance. Journal of immunology 179, 6749-6761 (2007).

48 Campbell, D., MacGillivray, I. & Carr-Hill, R. Pre-eclampsia in second pregnancy. Br J Obstet Gynaecol 92, 131-140 (1985).

49 Trupin, L., Simon, L. & Eskenazi, B. Change in paternity: a risk factor for preeclampsia in multiparas. Epidemiology 7, 240-244 (1996).

50 Belkaid, Y., Piccirillo, C. A., Mendez, S., Shevach, E. M. & Sacks, D. L. CD4+CD25+ regulatory T cells control Leishmania major persistence and immunity. Nature 420, 502-507, doi:10.1038/nature01152 (2002).

111

51 Nelson, R. W., McLachlan, J. B., Kurtz, J. R. & Jenkins, M. K. CD4+ T cell persistence and function after infection are maintained by low-level peptide:MHC class II presentation. Journal of immunology 190, 2828-2834 (2013).

52 Uzonna, J. E., Wei, G., Yurkowski, D. & Bretscher, P. Immune elimination of Leishmania major in mice: implications for immune memory, vaccination, and reactivation disease. Journal of immunology 167, 6967-6974 (2001).

53 Spurgin, L. G. & Richardson, D. S. How pathogens drive genetic diversity: MHC, mechanisms and misunderstandings. Proc. Biol. Sci. 277, 979-988 (2010).

54 Israeli, M., Kristt, D., Nardi, Y. & Klein, T. Genetic Considerations in Human Sex- Mate Selection: Partners Share Human Leukocyte Antigen but not Short- Tandem-Repeat Identity Markers. Am. J. Reprod. Immunol. 71, 467-471 (2014).

55 Chaix, R., Cao, C. & Donnelly, P. Is mate choice in humans MHC-dependent? PLoS genetics 4, e1000184, doi:10.1371/journal.pgen.1000184 (2008).

56 Ober, C. et al. HLA and mate choice in humans. American journal of human genetics 61, 497-504, doi:10.1086/515511 (1997).

57 Jenkins, M. K., Chu, H. H., McLachlan, J. B. & Moon, J. J. On the composition of the preimmune repertoire of T cells specific for Peptide-major histocompatibility complex ligands. Annual review of immunology 28, 275-294, doi:10.1146/annurev-immunol-030409-101253 (2010).

58 Nelson, J. L. The otherness of self: microchimerism in health and disease. Trends Immunol. 33, 421-427 (2012).

59 Rudensky, A., Rath, S., Preston-Hurlburt, P., Murphy, D. B. & Janeway, C. A., Jr. On the complexity of self. Nature 353, 660-662, doi:10.1038/353660a0 (1991).

60 Moon, J. J. et al. Tracking epitope-specific T cells. Nat. Protoc. 4, 565-581 (2009).

112

A Naive NIMA NIPA Self B C Naive NIMA NIPA Self 200 77 38 55 70 *** ***

CD44 20 I-Ab:2W1S Helios :2W1S Tregs Tregs :2W1S permouse 13 46 22 20 b I-A 2 100 *** ** 75 FOXP3 FOXP3 cells among +

CD25 hi 200 50

80 *** *** +

Helios 25 60 FOXP3 % among + 20 0 T cells T + 40 I-Ab:2W1S + − + − + − + − :2W1S CD4:2W1S b

CD4 20 Naiv NIM NIPA Self I-A T cells per mouse T e A %FOXP3 2 0 I-Ab:2W1S + − + − + − + − Naiv NIM NIPA Self e A

E : : b b

D A b/b A 80 80 Maternal I-A *** 60 60 T cells T T cells T + 2W1S + +/- among I- d/d OVA b/b among I- 40 40 X + I-A I-A + (H-2d/d) (H-2b/b) 20 20 I-Ab:2W1S presentation by 2W1SCD4 maternal and fetal cells 2W1SCD4 0 0 %FOXP3 %FOXP3 100 100 Maternal I-Ad/d *** 75 75 among among hi

2W1S hi +/- 50 50 I-Ab/b X OVA I-Ad/d b/b d/d :2W1S Tregs Tregs :2W1S (H-2 ) (H-2 ) Tregs :2W1S b b

Helios 25 Helios 25 I-A I-A

b % I-A :2W1S presentation % exclusively by fetal cells 0 0 Maternal MHC I-Ab I-Ad I-Ab I-Ad In utero + + − − haplotype Breastfeeding + − + − NIMA Naive

Figure 1. Developmental exposure to maternal tissue primes expanded NIMA- specific FOXP3+ Tregs. (A) Representative plots showing the gating strategy used to identify I-Ab:2W1S specific among CD4+ T cells (top), FOXP3+ Tregs among I-Ab:2W1S specific CD4+ T cells (middle), and composite data (bottom) for percent FOXP3+ among CD4+ T cells with I- Ab:2W1S specificity (filled) compared with bulk CD4+ T cells (open) in the spleen plus peripheral lymph nodes of naive (blue), NIMA-2W1S (red), NIPA-2W1S (green), or 2W1S-self (gray) 8 week old adult mice. (B) Total number of I-Ab:2W1S specific FOXP3+ Tregs (top) and CD4+ T cells (bottom) for each group of mice described in panel A. (C) Percent Helioshi among I-Ab:2W1S specific (red line) or bulk (gray shaded) FOXP3+ CD4+ T cells for each group of mice described in panel A. (D) Mating strategy for generating genetically identical NIMA-2W1S offspring born to b/b b either MHC class II I-A (I-A :2W1S55-68 peptide presented by cells of both maternal d/d b and fetal origin) or I-A (I-A :2W1S55-68 peptide only presented by cells of fetal origin)

113

haplotype mothers, and composite data for percent FOXP3+ among I-Ab:2W1S specific CD4+ T cells and Helioshi among I-Ab:2W1S specific FOXP3+ cells for each group of NIMA-2W1S (red) compared with naive (blue) mice. (E) Percent FOXP3+ among I-Ab:2W1S specific CD4+ T cells, and Helioshi among I- Ab:2W1S specific FOXP3+ cells for each group of cross-fostered offspring exposed to 2W1S-OVA in utero and/or postnatally through breastfeeding by 2W1S-OVA+ mothers. Each point represents the result from an individual female mouse, and these data are representative of at least three separate experiments each with similar results. Bars, mean ± 95% confidence interval. ** P < 0.01, *** P < 0.001. See also Figure S1.

114

A B 2 102 102 Heart 102 10 Liver 2W1S-OVA+ Naive 1 1 1 1 cells 10 10 * cells 10 10 ** 5 5

/10 /10 0 100 100 100 10

Geq Geq -1 10-1 10-1 10-1 10

Est. Est. L.O.D. Est. L.O.D. α-OVA -2 -2 -2 10-2 10 10 10 antibody α-OVA − − + α-OVA − − + Naive NIMA Naive NIMA

C NIMA Naive NIMA α-OVA :

19 60 20 b

A 80 100 *** *** 60 75 T cells T + among hi FOXP3 FOXP3 among I-

+ 40 50 CD25 :2W1S Tregs Tregs :2W1S b 68 48 70 20 Helios 25 I-A % 2W1SCD4

%FOXP3 0 0 α-OVA − − + α-OVA − − + Naive NIMA Naive NIMA Helios

Figure 2. NIMA-specific Treg expansion requires persistent postnatal exposure to microchimeric maternal cells. (A) Maternal 2W1S-OVA+ microchimeric cell encoding DNA levels in each tissue of naive (blue filled), NIMA-2W1S (red filled), or NIMA-2W1S mice treated with anti-OVA depleting antibody (red open). (B) Cell surface OVA expression levels among splenocytes from 2W1S-OVA+ compared with naive control mice after staining with anti-OVA (red line) or rabbit IgG isotype (gray shaded) antibodies. (C) Representative plots and composite data for FOXP3+ Tregs among I-Ab:2W1S specific CD4+ T cells, and Helios expression among I-Ab:2W1S specific FOXP3+ cells for each group of mice described in panel A. Each point represents the result from an individual female mouse at 8 weeks of age, and these data are representative of at least three separate experiments each with similar results. Bars, mean ± 95% confidence interval. * P < 0.05, ** P < 0.01, *** P < 0.001. See also Figure S2.

115

A Uterus/ Heart Liver Prostate NIMA Female 102 102 102 102 NIMANIMA FemaleMale 102 1 * 101 1 1 NIMANaiveNaive MaleFemale cells 10 10 10 10 NIMA Female 5 2 101 NaiveNIMANIMA MaleFemale

/10 10 100 100 100 0 1 10 NaiveNaive MaleFemale 100 Geq -1 10-1 -1 -1 Naive Male 10 10 0 10 10 10-1-1 Est. Est. 10-2 L.O.D.L.O.D. -2 -2 10-2 10-1 10 10 10-2-2 L.O.D. 10-3 10-2 L.O.D. 10-3 : 10-3 b B NIMA NIMA Naive Naive A 80 ** 10-3 60 T cells T + among I-

+ 40 CD44 20 I-Ab:2W1S 2W1SCD4

51 30 14 18 %FOXP3 0 100 ** 75 FOXP3 FOXP3 among CD25 hi 50 :2W1S Tregs Tregs :2W1S b 26 58 58 61 Helios 25 I-A % 0

Helios NIMA Naive

Figure 3. Expanded NIMA-specific Treg accumulation in female offspring parallels discordant maternal cell microchimerism in gender-specific reproductive tissue. (A) 2W1S-OVA+ encoding DNA levels in each tissue among NIMA-2W1S female (red circle), littermate 2W1S-NIMA male (red triangle), naive female (blue circle) and naive male (blue triangle) mice. (B) Representative plots and composite data showing I-Ab:2W1S specific CD4+ T cells (top), FOXP3+ Tregs among I-Ab:2W1S specific CD4+ T cells (middle), and Helios expression among Tregs with I-Ab:2W1S specificity (red line) or bulk specificity (gray shaded) among NIMA-2W1S female compared with NIMA-2W1S littermate male mice. Each point represents the result from an individual mouse at 8 weeks of age, and these data are representative of at least three separate experiments each with similar results. Bars, mean ± 95% confidence interval. * P < 0.05, ** P < 0.01

116

NIMA-2W1S − − + + + Pregnant − + − + + 2W1S+ + + − CD44 I-Ab:2W1S 14 19 56 80 21 FOXP3 FOXP3 CD25

: *** b

A 80

T cells T 60 + among I- + 40

20 2W1SCD4

%FOXP3 0 103 7.2x 3.4x 102

1 :2W1S Tregs Tregs :2W1S

permouse 10 b I-A 100 NIMA-2W1S − − + + + Pregnant − + − + + 2W1S+ + + −

Figure 4. NIMA-specific Treg expansion accelerated during pregnancy with NIMA- matched fetal antigen stimulation. Representative plots and composite data showing I-Ab:2W1S specific CD4+ T cells (top), FOXP3+ Tregs among I-Ab:2W1S specific CD4+ T cells (middle), and composite data for percent and number of FOXP3+ CD4+ T cells with I-Ab:2W1S specificity in virgin and midgestation (E11.5) naive female (blue) compared with NIMA-2W1S (red) female mice after mating with 2W1S-OVA+ transgenic male mice or non-transgenic controls. Each point represents the result from an individual mouse, these data are representative of at least three separate experiments each with similar results. Bars, mean ± 95% confidence interval. *** P < 0.001

117

A *** B *** *** *** 100 100

75 75

50 50

25 25 %fetalresorption %fetalresorption 0 0 *** *** 7 7 ) )

10 6 6 *** 10 *** 5 5 (log 4 (log 4 3 3 concepti concepti LmCFUsamong 2 2 L.O.D. LmCFUsamong L.O.D. 1 1 NIMA H-2d NIMA H-2k (2W1S-OVA) − + + + (2W1S-OVA) − + + + H2d H2d H2k H2d H2k H2k H2d H2k α-OVA − − − + α-OVA − − − + antibody antibody

Figure 5. Overlap between NIMA and paternal-fetal alloantigen protects against fetal resorption and in utero bacterial invasion following prenatal infection. (A) Percent fetal resorption (top) and average recoverable bacterial CFUs from each concepti per litter (bottom) five days following L. monocytogenes intravenous maternal infection initiated midgestation (E11.5) for naive female (blue) compared with NIMA-H- 2d-(2W1S-OVA) (red) female mice during allogeneic pregnancy sired by H-2d or third party H-2k males, or depletion of microchimeric 2W1S-OVA+ maternal cells with anti- OVA antibody prior to mating. (B) Percent fetal resorption (top) and average recoverable bacterial CFUs from each concepti per litter (bottom) five days following L. monocytogenes intravenous maternal infection initiated midgestation (E11.5) for naive female (blue) compared with NIMA-H- 2k-(2W1S-OVA) (red) female mice during allogeneic pregnancy sired by H-2k or third party H-2d males, or depletion of microchimeric 2W1S-OVA+ maternal cells with anti- OVA antibody prior to mating. Each point represents the result from an individual mouse, and these data are representative of at least three separate experiments each with similar results. Bars, mean ± 95% confidence interval. *** P < 0.001. See also Figure S3

118

A B * ** 100 *** 100 * *** 75 *** 75

50 50

25 25 %fetalresorption %fetalresorption 0 0 FOXP3DTR/WT − + + + + FOXP3DTR/WT − + + + + NIMA H-2d NIMA H-2k

(2W1S-OVA) − − + + + (2W1S-OVA) − − + + + H2d H2d H2d H2k H2d H2k H2k H2k H2d H2k α-OVA α-OVA − − − − + − − − − + antibody antibody

Figure 6. Overlap between NIMA and paternal-fetal antigen protects against fetal wastage triggered by partial depletion of maternal FOXP3+ regulatory T cells. (A) Percent fetal resorption for naive (blue) FOXP3WT/WT and FOXP3DTR/WT female mice compared with each group of NIMA-H-2d FOXP3DTR/WT (red) female mice five days after initiating diphtheria toxin during allogeneic pregnancy sired by H-2d or third party H-2k males, or depletion of microchimeric 2W1S-OVA+ maternal cells with anti-OVA antibody prior to mating. (B) Percent fetal resorption for naive (blue) FOXP3WT/WT and FOXP3DTR/WT female mice compared with each group of NIMA-H-2k FOXP3DTR/WT (red) female mice five days after initiating diphtheria toxin during allogeneic pregnancy sired by H-2k or third party H-2d males, or depletion of microchimeric 2W1S-OVA+ maternal cells with anti-OVA antibody prior to mating. Each point represents the result from an individual mouse, these data are representative of at least three separate experiments each with similar results. Bars, mean ± 95% confidence interval. * P < 0.05, ** P < 0.01, *** P < 0.001. See also Figure S4.

119

Traditional Mendelian genetics

Two unique MHC haplotypes alleles Mother (red, green), transmitted separately to individual offspring

Common susceptibility to Female fetal wastage offspring regardless of paternal-fetal MHC haplotype

Cross-generational reproductive fitness (enforced by tolerance to non-inherited maternal antigen)

Mother Increased resiliency against fetal wastage in pregnancies sired by males with shared NIMA specificity Female offspring Background susceptibility to fetal wastage in pregnancies sired by males without Microchimeric shared NIMA specificity maternal cells

Figure 7. Cross-generational reproductive fitness enforced by vertically transferred microchimeric maternal cells in eutherian placental mammals. In traditional Mendelian genetics (top), pregnancies among female offspring are equally susceptible to fetal wastage or other complications stemming from disruptions in fetal tolerance regardless of paternal MHC haplotype specificity. Comparatively, persistent postnatal maintenance of tolerogenic microchimeric maternal cells in female offspring promotes cross-generational reproductive fitness (bottom) by selectively protecting against fetal wastage during next generation pregnancies sired by males with shared overlapping NIMA specificity.

120

2W1S55-68 A peptide

rabbit β-globin polyA

cytomegalovirus chicken β- ovalbumin Db trans- immediate early actin Kb signal membrane enhancer promoter sequence

B C

2W1S 2W1S X +/- OVA OVA+/- X H-2d/d H-2b/b H-2d/d H-2b/b I-Ad/d I-Ab/b I-Ad/d I-Ab/b

50% 50% 50% 50%

2W1S 2W1S 2W1S 2W1S OVA+/- OVA+/- OVA+/- OVA+/-

NIMA-2W1S Self-2W1S Self-2W1S NIPA-2W1S

X H-2d/d H-2b/b I-Ad/d I-Ab/b

100%

Naive-2W1S

Figure S1. Mating strategy that transform 2W1S and OVA into surrogate NIMA, NIPA, or self antigens, related to Figure 1. (A) Schematic of OVA-2W1S transgenic expression construct. The coding sequence for

2W1S55-68 peptide was embedded with a membrane-bound form of chicken ovalbumin. Adapted from (Moon et al., 2011). (B) Female mice on the C57BL/6 (H-2b/b I-Ab/b) background heterozygous for the transgene encoding constitutive cell surface expression of a recombinant protein +/- containing ovalbumin (OVA) and 2W1S55-68 peptide (2W1S-OVA ) were mated with d/d d/d non-transgenic Balb/c (H-2 I-A ) male mice, thereby transforming OVA and 2W1S55-

68 into non-inherited maternal antigens (NIMA) in half of the offspring. (C) Non-transgenic female C57BL/6 (H-2b/b I-Ab/b) mice were mated with male mice on the Balb/c (H-2d/d I-Ad/d) background heterozygous for the same transgene that encodes constitutive cell surface expression of a recombinant protein containing ovalbumin

121

+/- (OVA) and 2W1S55-68 peptide (2W1S-OVA ), thereby transforming OVA and 2W1S55-68 into non-inherited paternal antigens (NIPA) in half of the offspring. (D) F1 offspring from mating between non-transgenic female C57BL/6 (H-2b/b I-Ab/b) and male Balb/c (H-2d/d I-Ad/d) mice were used as naive controls.

122

A 5

4 2W1S-OVA+ 4040# y = -1.137ln(x) + 38.443 R² = 0.98576 10-1 3535# 3 10-2 10-3 3030# 2 10-4 10-5

Cyclethreshold 2525# 1 (Baselinesubtracted) 20

Relativefluorescence units 20#

-6 - 100 10 1 0.1 0.01 0.001 0 10 and 2W1S-OVA % 2W1S-OVA+ cells

0 10 20 30 40 50 Cycle Threshold

B 1.81 2W1S +/- OVA + Rabbit IgG

CD45.1 CD45.2 CD45.2 CD45.1

0.01 2W1S OVA+/- + α-OVA antibody

CD45.1 CD45.2 CD45.2 CD45.1 Figure S2. Enumeration and depletion of 2W1S-OVA+ cells, related to Figure 2. (A) Representative real-time PCR (left) and standard curve (right) demonstrating the linear relationship between undiluted OVA+ DNA (from 2W1S-OVA+ mice), and serial 10-fold dilutions of OVA+ DNA added to OVA- control DNA (from naive C57BL/6 control mice). (B) Splenocytes from 2W1S-OVA+ CD45.1+ mice were adoptively transferred into isogenic CD45.2+ C57BL/6 recipients treated with depleting anti-OVA or rabbit IgG isotype control antibody (650 µg each per mouse). Three days after cell transfer and antibody administration, 2W1S-OVA+ CD45.1+ donor cells among splenocytes remaining in each group were evaluated by flow cytometry.

123

2W1S H-2b/b OVA+/- X H-2d/d Generation 1 breeding 50% 50%

2W1S 2W1S All females +/- +/- OVA OVA H-2b/d

2W1S +/- H-2b/b X OVA H-2b/d Generation 2 breeding

25% 25% 25% 25%

2W1S 2W1S 2W1S 2W1S OVA+/- OVA+/- OVA+/- OVA+/-

H-2b/d H-2b/d H-2b/b H-2b/b H-2b/b H-2b/b H-2b/d H-2b/d NIMA-H-2d 2W1S-OVA

Generation 3 experimental breeding Fetal antigen with H-2d/d shared NIMA specificity

No overlap between H-2k/k NIMA and fetal antigen

Figure S3. Breeding strategy for generating mice to investigate pregnancy outcomes when MHC haplotype alleles and 2W1S-OVA are simultaneously transformed into surrogate NIMA, related to Figure 5. Female H-2d/d haplotype mice on the C57BL/6 background (B6.C-H2d/bByJ) were mated with H-2b/b haplotype 2W1S- OVA+ C57BL/6 male mice (Generation 1 breeding). All F1 offspring are H-2b/d, and females heterozygous for the 2W1S-OVA expression transgene were used for breeding with C57BL/6 (H-2b/b) male mice (Generation 2 breeding). Thereafter, 25% of offspring do not inherit genes encoding H-2d haplotype alleles or 2W1S-OVA (H-2d and 2W1S- OVA simultaneously transformed into surrogate NIMA). Pregnancy outcomes among the latter group of H-2b/b female mice were evaluated during allogeneic pregnancy sired by Balb/c (H-2d/d; fetal antigen with shared NIMA specificity) or CBA (H-2k/k, no overlap between NIMA and fetal antigen) (Generation 3 experimental breeding). For transforming H-2k haplotype antigens into NIMA in H-2b/b mice, female H-2k/k haplotype mice (B6.Ak-H2k/J) were substituted for H-2d/d B6.C-H2d/bByJ mice beginning in generation 1 breeding using this mating scheme.

124

2W1S H-2b/b OVA+/- X H-2d/d 25% Generation 1 breeding 50% 50%

2W1S 2W1S All females +/- +/- OVA OVA H-2b/d

2W1S +/- X OVA H-2b/d Generation 2 breeding H-2b/b FOXP3DTR

25% 25% 25%

2W1S 2W1S 2W1S 2W1S All females OVA+/- OVA+/- OVA+/- OVA+/- FOXP3DTR/WT

H-2b/d H-2b/d H-2b/b H-2b/b H-2b/b H-2b/b H-2b/d H-2b/d NIMA-H-2d 2W1S-OVA

Generation 3 experimental breeding Fetal antigen with H-2d/d shared NIMA specificity

No overlap between H-2k/k NIMA and fetal antigen

Figure S4. Strategy for generating FOXP3DTR/WT mice to investigate pregnancy outcomes when MHC haplotype alleles and 2W1S-OVA are simultaneously transformed into surrogate NIMA after partial depletion of maternal FOXP3+ Tregs, related to Figure 6. Female H-2d/d haplotype mice on the C57BL/6 background (B6.C- H2d/bByJ) were mated with H-2b/b haplotype 2W1S-OVA+ C57BL/6 male mice (Generation 1 breeding). All F1 offspring are H-2b/d, and females heterozygous for the 2W1S-OVA expression transgene were used for breeding with FOXP3DTR/o male mice on the C57BL/6 (H-2b/b) background (Generation 2 breeding). Thereafter, all female progeny are FOXP3DTR/WT, and 25% of offspring do not inherit genes encoding H-2d haplotype alleles or 2W1S-OVA (H-2d and 2W1S-OVA simultaneously transformed into surrogate NIMA). Pregnancy outcomes among the latter group of H-2b/b female mice were evaluated after partial depletion of bulk maternal FOXP3+ Tregs with diphtheria toxin administration beginning midgestation during allogeneic pregnancy sired by Balb/c (H-2d/d; fetal antigen encompasses NIMA specificity) or CBA (H-2k/k, no overlap between NIMA and fetal antigen) (Generation 3 experimental breeding). For transforming H-2k haplotype antigens into NIMA in H-2b/b mice, female H-2k/k haplotype mice (B6.Ak-H2k/J)

125

were substituted for H-2d/d B6.C-H2d/bByJ mice beginning in generation 1 breeding using this mating scheme.

126 Chapter 6: Sustained protection against fetal wastage conferred by prior pregnancy despite numerical loss of maternal regulatory CD4+ T cell memory

* This manuscript is in preparation

127

128 Expanded immune tolerance required for accommodating foreign paternal antigens expressed by the developing fetus parallels expanded accumulation of immune-suppressive maternal regulatory CD4+ T cells (Tregs)1-4. Reciprocally, blunted accumulation of maternal Tregs occurs in many human pregnancy complications linked with fractured fetal tolerance including spontaneous abortion and preeclampsia1,5-8. A similar necessity for sustained expansion of maternal Tregs occurs in animals where even partial depletion of maternal Tregs from expanded pregnancy levels disrupts fetal tolerance and triggers fetal wastage9,10. We have previously shown pregnancy primes selective expansion of maternal Tregs with fetal specificity, and that these cells persists at expanded levels long-term after partuirtion9. With fetal antigen re-stimulation during secondary pregnancy, sharply accelerated tempo of maternal memory Treg re- accumulation protect against pregnancy complications by enforce fetal tolerance9. These protective memory features for maternal Tregs with fetal specificity in mice recapitulate human partner-specific protective benefits of prior successful pregnancy against complications in subsequent pregnancies11-14.

Herein, we investigate whether memory regulatory CD4+ T cells share with memory effector CD4+ T cells the need for antigenic reminders and the potential sources of genetically foreign fetal antigen that persist in mothers postpartum.

We show genetically foreign fetal cells retained in mothers after pregnancy sustain quantitative retention of maternal Tregs with fetal-specificity postpartum.

Interestingly however, despite the loss of expanded memory maternal Tregs after depletion of fetal microchimeric cells, the protective features of maternal memory

129 Tregs remain intact and parallel enriched accumulation of fetal-specific CD4+ T cells poised to re-acquire FOXP3 expression following secondary fetal-antigen stimulation. Thus, maternal Tregs adapt to loss of fetal-antigen stimulation by transiently silencing FOXP3 expression while functionally retaining the ability to re-express FOXP3 upon secondary fetal antigen stimulation.

Antigen-specific CD4+ and CD8+ T cells fundamentally differ in their requirements for cognate antigen stimulation for long-term persistence as memory in immune cells. The requirements for maintenance of protective memory features for effector T cells indicate

CD4+ T cells are more reliant on persistent antigen stimulation to facilitate their quantitative retention, while CD8+ memory T cells form durable memory that persists despite complete lack of cognate antigen stimulation. For example, in response to acute infection with lymphocytic choriomeningitis virus (LCMV) [Armstrong clone 53b], viral- specific CD8+ T cells form stable life-long memory while CD4+ T cells with viral- specificity within the same host continuously decline over time15. In turn, for pathogens that establish persistent infection such as Leishmania major and Salmonella enterica

[serovar Tymphimurium], pathogen-specific CD4+ T cells and protection from secondary protection are retained16-18. However, retention of quantitative or functional memory following secondary challenge relies on sustained low-level antigen stimulation, as both are lost following either natural clearance or anti-microbial pathogen elimination.

Together these and other studies have characterized distinct requirements for CD8+ compared with CD4+ effector T cell lineages to form “memory” following infection.

We and others have demonstrated similar memory features also extend to immune- suppressive regulatory CD4+ T cells (Tregs) characterized by FOXP3 expression. Initial

130 demonstration of Treg memory utilized tools allowing tracking of CD4+ T cells with defined specificity to ovalbumin (OVA) that represents a surrogate self-antigen, inducibly expressed in the skin. Initial exposure to self-OVA antigen in the skin drove systemic expansion of activated Tregs with OVA-specificity19. Moreover, these Tregs are retained within the skin following antigen-removal and attenuate auto-inflammatory disease upon secondary antigen expression 30 days later. Expanding on these findings for Tregs with specificity to self-antigens, we have shown maternal Tregs with specificity

b to the model antigen I-A :2W1S55-68 expressed as a non-self surrogate fetal antigen accumulate throughout allogeneic pregnancy and are required for maintaining maternal- fetal tolerance9. Following delivery of the fetus and other gross products of conception, maternal Tregs with fetal-specificity persist at enriched levels up to 100 days postpartum and subsequently re-expand with accelerated tempo during secondary pregnancy. Given the importance of maternal Tregs with fetal-specificity in enforcing fetal tolerance, their accelerated re-expansion in subsequent pregnancies likely mediates partner-specific protection from idiopathic pregnancy complications associated with fractured fetal tolerance9,11-14.

In addition to previous reports of diminished pregnancy complications from non- infectious triggers, we reasoned protective maternal Treg memory associated with secondary pregnancy may also provide increased resiliency from fetal wastage that stems from prenatal infection. Here, infection with Listeria monocytogenes (Lm), where disruptions in fetal tolerance drive fetal wastage and congenital invasion20, was used to probe whether expanded maternal Tregs with fetal specificity protect against prenatal infection. Remarkably, secondary pregnancies sired by MHC haplotype-identical fathers

131 (H-2d) were significantly more refractory to Lm-induced fetal wastage and ensuing in utero pathogen invasion compared to primary pregnancies (Fig 1). Importantly, these protective benefits were partner-specific since fetal wastage rebounded when third party males that express irrelevant MHC haplotype alleles (e.g. H-2k) were used to sire secondary pregnancy (Fig 1). Thus, the protective benefits of sustained maternal Treg memory extend to fractured fetal tolerance stemming from prenatal infection.

Given the distinct requirements for antigen persistence to sustain memory for effector T cell subsets we wondered if sustained Treg memory requires similar low-level antigenic stimulation. Previous studies have demonstrated bi-directional trafficking of maternal and fetal cells across the placenta during pregnancy21-24. Extended postpartum analysis revealed genetically foreign fetal cells engraft in multiple maternal tissues and establish long-term microchimerism25-27. We hypothesized fetal microchimeric cells transferred to mothers during gestation that persist within multiple maternal tissues may be provide a necessary source of low-level antigenic reminders required to sustain maternal Treg memory similar to their FOXP3- CD4+ effector T cell counterparts.

In support of this hypothesis, we have previously demonstrated vertically transferred maternal microchimeric cells are required to sustain expanded accumulation of Tregs with specificity to non-inherited maternal antigens (NIMA) in offspring28. To directly investigate the requirement for persistent low-level stimulation from fetal microchimeric cells in postpartum retention of maternal Treg memory to fetal antigen, we utilized an innovative mating strategy that transforms model antigens into surrogate fetal antigens.

Male mice heterozygous for a transgene that encodes ubiquitous cell surface expression of a recombinant fusion protein containing ovalbumin (OVA) and the

132 2W1S55-68 variant of MHC class II I-Eα were used to establish allogeneic pregnancy with non-transgenic females, thereby transforming OVA and 2W1S55-68 peptide into fetal antigend in the mother9,29 (Extended Data Fig 1a,b). We reasoned this approach that

+ b 30 exploits the high frequency of endogenous CD4 T cells with I-A :2W1S55-68 specificity , combined with tools allowing quantification and selective manipulation of OVA- expressing fetal cells in vivo28 would permit more definitive analysis of the necessary cross-talk between fetal-antigen expressing and fetal-antigen responsive cells.

Previous studies that quantified microchimeric fetal cells based on discordant fetal Y- chromosome DNA, recovered fetal cells at a frequency of 1 fetal cell in 103 to 105 maternal cells in biopsy samples from women with previous male offspring27. Consistent with these prior studies, 2W1S-OVA+ fetal cells were found in the uterus, heart and a wide range of other tissues of female mice at 30 days postpartum at levels ranging from

1 fetal cell in 103 to 106 maternal cells (Fig. 2a and Extended Data Fig 1c). In turn, accumulation of maternal Tregs with fetal-2W1S specificity that parallels postpartum retention of fetal microchimeric cells was lost following depletion of microchimeric fetal cells with in vivo administration of anti-OVA antibodies (Fig 2b,c). These results establish a definitive cause and effect relationship for persistent microchimeric fetal cells that prime and sustain fetal-specific Treg expansion, highlighting a shared necessity for low-level antigen persistence in retention of antigen-experienced Tregs and effector

CD4+ T cells17,18.

Given the necessity for maternal Treg memory in reinforcing fetal tolerance during secondary pregnancy9, we considered whether a decline in fetal-specific Tregs following fetal microchimeric cell depletion may also diminish protection against fetal wastage in

133 secondary pregnancy. Here male mice with homozygous expression of the 2W1S-OVA transgene were used to sire allogeneic pregnancies in non-transgenic females

(Extended Data Figure 2a). This allowed complete depletion of fetal microchimeric cells with in vivo administration of anti-OVA antibody that would simultaneously reduce quantitative retention of maternal memory Tregs with specificity for fetal 2W1S/OVA, but also MHC-haplotype antigens (H-2d) (Extended Data Figure 2a). Remarkably, robust fetal wastage and in utero invasion triggered by prenatal Lm infection during primary pregnancy was dramatically reduced during secondary pregnancy even when fetal microchimeric cells were depleted with anti-OVA antibody prior to pregnancy (Fig 3a).

Thus functional Treg memory in mothers depleted of fetal microchimeric cells is retained despite a decline in maternal memory Tregs postpartum. Together, these results demonstrate differential requirements exist regarding the need for fetal antigenic stimulation to sustain quantitative compared to functional Treg memory.

Since protection from pregnancy complications during secondary pregnancy is associated with more accelerated accumulation of fetal-specific Tregs9, we investigated the expansion kinetics of maternal Tregs during secondary pregnancy following fetal microchimeric cell depletion. In mothers treated with anti-OVA antibody prior to secondary pregnancy, maternal Tregs with fetal-2W1S specificity expanded to a set point at mid-gestation [E11.5] similar to untreated secondary pregnant females (Fig 3b).

Together, these results suggests while quantitative retention of maternal memory Tregs with fetal-specificity requires postpartum persistence of fetal microchimeric cells their functional capacity is preserved even in the absence of fetal antigen stimulation.

134 Previous reports using lineage fate-tracking mice have demonstrated a subset of CD4+

FOXP3- cells known as “exFoxp3 cells” that transiently silence FOXP3 expression but remain epigenetically poised to re-express FOXP3 upon re-stimulation31. While the biological mechanism underlying the generation and biological imperative for exFoxp3 cells remains controversial32, other recent studies suggest certain subsets of exFoxp3 cells may arise during states of inflammation or lymphopenia and in some cases their instability could lead to conversion into auto-reactive T cells that precipitate autoimmunity33. However, exFoxp3 cells have demonstrated remarkable lineage stability through this epigenetic maintenance that poises these cells for rapid re- expression of FOXP3 re-expression upon TCR re-stimulation31. Given that the total number of CD4+ T cells with fetal specificity does not change following microchimeric cell depletion (Fig 2), we reasoned a quantitative decline in fetal-specific Tregs may not be to loss of these cells, but may instead reflect similar silencing of FOXP3 expression.

In turn, generation of maternal exFoxp3 cells with fetal-specificity that retain the ability to re-express FOXP3 after secondary fetal-antigen stimulation may explain retention of fetal-specific maternal Treg memory despite fetal microchimeric cell depletion. To test this we generated female mice, expressing green fluorescent protein (GFP) and Cre recombinase protein (CRE) driven by the FOXP3 promoter on a bacterial artificial chromosome (BAC-FOXP3GFP-CRE) together with tdTomato fluorescent protein expression controlled by the Rosa26 promoter after excision of a loxP-flanked (R26- tdTomato mice), that allowed tracing of FOXP3 expression among CD4+ cells. This allowed tracking of fetal-antigen specific FOXP3+ and exFoxp3 generation postpartum and following fetal microchimeric cell depletion. This analysis of maternal CD4+ T cells

135 with fetal specificity revealed the postpartum decline of maternal Tregs following depletion of fetal microchimeric cells reflects significant accumulation of tdTomato+

FOXP3-GFP- exFoxp3 cells (Fig 4a). Additionally, loss of FOXP3 in these cells was also associated with decreased expression of Treg-associated molecules including the high affinity IL-2Rα, CD25 and glucocorticoid-induced TNFR-related protein (GITR) consistent with previous analysis of exFoxp3 cells (Extended Data Figure 3a). To determine whether exFoxp3 cells with fetal-specificity were the source of rapidly accumulated fetal-specific Tregs during secondary pregnancy we tracked the expansion of Tregs with fetal-2W1S specificity following secondary fetal-antigen stimulation among congenically marked splenocytes from primary postpartum FOXP3WT/WT and

FOXP3DTR/DTR female donors treated with isotype or anti-OVA antibodies after delivery and co-transferred into naive recipients at postpartum day 30 (Extended Data Fig 3b).

This strategy allows depletion of FOXP3+ cells that co-express the high-affinity human diphtheria toxin receptor while avoiding depletion of exFoxp3 (FOXP3-) cells from transferred FOXP3DTR/DTR donor cells with low-dose DT treatment (Extended Data Fig

3c). In mice depleted of fetal microchimeric cells accumulation of exFoxp3 cells with fetal-specificity renders these cells resistant to DT and allows direct comparison of the expansion capacity of exFoxp3 cells to the full complement of maternal Tregs with fetal- specificity in FOXP3WT/WT donor cells within the same recipient. Remarkably, while DT- mediated FOXP3+ cell ablation diminished the expansion of fetal-specific Tregs during secondary pregnancy in isotype treated donor cells, maternal Tregs with fetal-specificity in both FOXP3WT/WT and FOXP3DTR/DTR donor cells from anti-OVA treated females rapidly accumulated with similar expansion kinetics (Fig 4b).

136 Together these findings suggest maternal Tregs with fetal-specificity depend on persistence of fetal microchimeric cells for quantitative retention. However, in the event fetal antigen stimulation is loss, maternal memory Tregs adapt by turning off FOXP3 expression but retaining the ability to rapidly re-express FOXP3 upon fetal-antigen stimulation during subsequent pregnancies and mediate their protective features. While accumulation of exFoxp3 cells with fetal-specificity relies on artificial elimination of fetal microchimeric cells through antibody-mediated depletion, in nature this phenomenon may result from natural decline of fetal cell stemming from extended durations between matings or even displacement of fetal microchimeric cells from a prior pregnancy by fetal microchimeric cells from a third-party male mating. In this case loss of FOXP3 expression may allow a state of quiescence associated with decreased proliferation allowing diminished cell turnover and increased lifespan for memory Tregs in the event of fetal-antigen re-stimulation in subsequent future pregnancies. In the broader context, generation of exFoxp3 cells may also allow retention of tolerogenic memory to other foreign, harmless antigens including commensal microbes and food proteins whose expression may also be transient and spontaneous over time. Furthermore, studies investigating the requirement to sustain memory in CD4+ T cells have primarily focused on quantitative retention and not functional protective properties associated with memory. Therefore our findings suggest revisiting the dependence of protective effector

CD4+ T cell memory on persistent antigen stimulation following primary infection or vaccination.

137 Materials and Methods

Mice. C57BL/6 (H-2b/b; CD45.2+), Balb/c (H-2d/d), CBA/J (H-2k/k), B6.Cg-

Gt(ROSA)26Sortm14(CAG-tdTomato)Hze/J, B6129S-Tg(FOXP3-EGFP/cre)1aJbs/J, and

B6.SJL-PtprcaPepcb/BoyJ (H-2b/b; CD45.1+) mice were purchased from The Jackson

Laboratory. 2W1S-OVA+ transgenic mice that constitutively express recombinant

DTR/DTR 2W1S55-68-OVA protein under the β-actin promoter, and FOXP3 mice where

FOXP3+ cells are susceptible to diphtheria toxin induced ablation have each been described29,34. 2W1S-OVA+ mice were maintained on the Balb/c background after backcrossing for >10 generations9. 2W1S-OVA+/- males and females were mated together to generate 2W1S-OVA+/+ males. Males who sired two consecutive litters of all

2W1S-OVA+/- pups were considered 2W1S-OVA+/+ and subsequently used for experimental matings. For partial transient maternal Treg ablation, FOXP3DTR/WT pregnant females were administered purified diphtheria toxin daily (Sigma-Aldrich, USA)

(0.5 µg first dose, followed by 0.1 µg/dose thereafter) beginning midgestation (E11.5) for 5 consecutive days, and frequency of fetal resorption evaluated E16.5 aged matched controls under Cincinnati Children’s Hospital IACUC approved protocols.

Tetramer enrichment and flow cytometry. Cells were stained with phycoerythrin

b (PE)-conjugated MHC class II I-A :2W1S55-68 tetramer followed by enrichment with anti-

PE-conjugated magnetic beads (Miltenyi Biotec) as previously described9,30,35. To identify CD4+ T cells with defined I-Ab:2W1S specificity, cells isolated from pooled secondary lymphoid tissue (spleen plus axillary, brachial, cervical, inguinal, mesenteric, pancreatic, para-aortic/uterine lymph nodes) of each mouse, were enriched with PE

b conjugated I-A :2W1S55-68 tetramer, and stained for cell-surface CD3e (145-2C11), CD4

138 (GK1.5), CD8α (53-7.3), CD25 (PC61), CD44 (IM7), CD11b (M1/70), CD11c (N418),

B220 (RA3-B62), F4/80 (BM8), and intranuclear FOXP3 (FJK-16s) expression using commercially available antibodies and cell permeabilization reagents (BD PharMingen or eBioscience). Cells stained with fluorochrome-conjugated tetramer and/or antibody were acquired using a FACSCanto cytometer (Becton Dickinson), and analyzed using

FlowJo (TreeStar) software. Lineage tracking of fetal-2W1S specific exFoxp3 cells,

CD4+ I-Ab:2W1S+ were analyzed for dTomato fluorescence in FOXP3-CRE+ ROSA26- dTomato+ female mice at postpartum day 30 following pregnancies sired with Balb/c-

2W1S. All dTomato+ CD4 cells with fetal-2W1S specificity were further analyzed for

FOXP3-eGFP expression to identify exFoxpe3 cells (dTomato+ GFP-).

Bacteria. For infection, virulent Listeria monocytogenes (wildtype strain 10403s) was grown to early log phase (OD600 0.1) in brain heart infusion media at 37ºC, washed and diluted with sterile saline, and inoculated intravenously via the lateral tail vein (104

CFUs) at midgestation (E11.5)20. The inoculum for each experiment was confirmed by spreading a diluted aliquot onto agar plates. Five days thereafter, fetal resorption and in utero bacteria invasion was evaluated by sterilely dissecting each concepti, homogenization in sterile saline containing 0.05% Triton X-100 to release intracellular bacteria, plating serial dilutions of each concepti homogenate onto brain heart infusion agar plates, and enumerating average recoverable Lm CFUs for concepti within each litter after incubation at 37ºC for 24 hours.

DNA extraction and quantitative PCR. All tissues were sterilely dissected, and DNA extracted from each tissue using the QIAamp DNA extraction kit (Qiagen). Following

DNA isolation, PCR for enumerating OVA+ DNA was performed in 12 separate wells per

139 tissue each containing 333 ng genomic DNA (GEq ~3.33 x 105 cells) in 20 µl total volume together with 10 µl Taqman Gene Expression Master Mix and 1 µl ovalbumin- specific Taqman assay (Applied Biosystems), for a detection limit of ~1 in 4 x 106 cells per tissue28. Amplification was performed using the 7500 Fast Real-Time PCR System

(Life Technologies) under the following program: 95°C for 10 minutes, followed by forty cycles of 95°C for 15 seconds and 60°C for 1 minute. A standard curve for OVA+ DNA was generated and utilized as previously described28.

Depletion of microchimeric 2W1S-OVA+ maternal cells. To deplete microchimeric

2W1S-OVA+ cells, following pregnancies sired by 2W1S-OVA+ males, female mice were administered 1mg purified rabbit α-OVA (Acris R1101) or IgG isotype antibody (Sigma-

Aldrich) by intraperitoneal injection, beginning at postpartum day 2 and again 14 days later with a second treatment of 500 µg of the same antibody. On postpartum day 30 following primary pregnancy, the level of maternal 2W1S-OVA+ cell microchimerism in each tissue was analyzed by quantitative real-time PCR and corresponding antigen-

+ b specific CD4 T cells with fetal-2W1S specificity were investigated using I-A :2W1S55-68 tetramer staining.

Statistical analysis. In each experiment, allogeneic pregnant female mice were randomized for either administration of anti-OVA or isotype antibody, or for secondary matings to investigate protection from prenatal infection. Where applicable, data sets with a normal distribution, were analyzed using parametric statistical test (t-test or

ANOVA) (Prism, GraphPad); and P < 0.05 was taken as statistical significance.

140 100 Uninfected (H-2d father) 75 1' (H-2d father) 2' (H-2d father) 50 2' (H-2k father) 25 %fetalresorption 0 7

) 6 10 5 (log 4 3 concepti

LmCFUsamong 2 1 L.O.D.

Figure 1. Partner-specific protection from prenatal infection and fetal wastage triggered by disruptions in fetal tolerance. Percent fetal resorption (top) and mean recoverable bacterial CFUs (bottom) from each litter five days after Lm infection initiated at E11.5 during primary allogeneic pregnancy (H2d father, black squares), secondary pregnancy (H2d father, blue closed squares), secondary pregnancy with third-party male (H2k father, blue open squares), or uninfected primary allogeneic pregnant female controls (grey squares). Each data point represents an individual mouse, pooled from at least three separate experiments each with similar results. Bars, mean ± SEM. L.O.D., limit of detection.

Figure 1

141 A Uterus Heart 103 * * 101 Virgin cells 2 cells 5 10 5 Postpartum 100 * * /10 /10 Postpartum α-OVA 101 Geq 100 Geq 10-1 -1 Est. Est. 10 Est. L.O.D. L.O.D.

Postpartum Virgin Postpartum B α-OVA : b 60 *** *** A

T cells T 40 + CD3 among I- I-Ab:2W1S + 20 12 40 18 2W1SCD4

%FOXP3 0

FOXP3

C 2000 2000

+ ** *** *** 200 200 :2W1S CD4:2W1S :2W1S Tregs Tregs :2W1S

20 b 20 permouse b I-A I-A T cells per mouse T 2 2

Figure 2. Microchimeric fetal cells sustain retention of fetal-specific Treg in mothers postpartum. A. Level of fetal 2W1S-OVA+ cell DNA in the indicated tissue among day 30 post-partum females impregnated by Balb/c-2W1S+ fathers and treated with isotype (blue filled) or anti-OVA antibody (red filled), or virgin controls (black filled). B. Representative flow cytometry plots (left) and composite graph (right) showing Figure 2 percentage FOXP3+ among I-Ab:2W1S-specific (black line) or bulk CD4+ T cells (grey shaded) for each group of mice described in panel A. C. Number of FOXP3+ or total CD4+ T cells with I-Ab:2W1S specificity for each group of mice described in panel A. Each data point represents an individual female mouse 7-8 weeks of age at initial mating, pooled from at least three separate experiments each with similar results. Bars, mean ± SEM. L.O.D., limit of detection. *p < 0.05, **p < 0.01, ***p < 0.001

142 A B Isotype α-OVA 34 12 100 pp30 75 65 63 50 E11.5 (2°) 25 %fetalresorption FOXP3 FOXP3 0 CD25

: 4.2x 8.1x 7 b A 75 ) 10

5 cells T (log + 50 among I- + 3 25 concepti LmCFUsamong

L.O.D 2W1SCD4 1 %FOXP3 0 α-OVA − − − + Pregnant − + − + 1° 2° Isotype α-OVA

Figure 3. Durable functional maternal Treg memory despite depletion of fetal microchimeric cells. A. Percent fetal resorption and mean recoverable bacterial CFUs from each litter in uninfected females (grey) or five days following prenatal Lm infection initiated at E11.5 during primary allogeneic pregnancy (2W1S+/+H2d father, black), secondary pregnancy with an MHC-haplotype identical (H2d) father and treated with isotype (blue) or anti-OVA (red) antibody prior to pregnancy. B. Representative plots (top) and composite graph (bottom) illustrating frequency of FOXP3+ CD4+ cells with fetal I-Ab:2W1S specificity at postpartum day 30 following primary pregnancy (circles) or at mid-gestation (E11.5) during secondary pregnancy (squares) in allogeneic pregnant + Figurefemales 3 mated with Balb/c-2W1S fathers and treated with isotype (blue) or anti-OVA (red) antibody prior to pregnancy. Each data point represents an individual mouse, pooled from at least three separate experiments each with similar results. Bars, mean ± SEM. L.O.D., limit of detection.

143

Figure 4. Accumulation of exFoxp3 cells with fetal-specificity allows rapid re- expansion following secondary fetal antigen re-stimulation. A. Representative flow cytometry plots (left) illustrating loss of FOXP3-eGFP expression in dTomato+ fetal I- Ab:2W1S specific (black line) compared with bulk CD4+ T cells (grey shaded), and composite data (right) enumerating shifts in dTomato+ FOXP3+ (green) and FOXP3- “exFoxp3” CD4 T cells with fetal-specificity following post-partum anti-OVA antibody treatment. B. Donor CD4 T cells from post-partum FOXP3WT/WT FOXP3DTR/DTR females mated with 2W1S+/+H2d males and treated with anti-OVA antibody or isotype control were adoptively transferred into congenically discordant virgin females, and administered diphtheria toxin the next day. Representative FOXP3 expression among fetal-specific CD4 T cells from each donor (left) was assessed post-partum after mating recipients with 2W1S+/+H2d males. Composite data (right) of the ratio of FOXP3+ fetal- specific CD4 T cells recovered from FOXP3DTR/DTR compared with FOXP3WT/WT donors within primary postpartum recipients treated with isotype (blue) or anti-OVA (red) antibody following delivery.

144 2W1S55-68 peptide A rabbit β-globin polyA cytomegalovirus chicken β- ovalbumin Db trans- immediate early actin Kb signal membrane enhancer promoter sequence

2W1S OVA+/- X H-2d/d H-2b/b I-Ad/d I-Ab/b

C Virgin 102 Postpartum 1

cells 10 5

/10 100

Geq 10-1 Est. Est. 10-2 LN Lung Liver Spleen Brain Blood Kidney Thymus Intestines D

among 40 + T cells T + 20 CD4 %FOXP3 0

Extended Data Figure 1. Mating strategy that allows detection of OVA+ fetal cells andExtended parallel trackingData Figure of CD4 1 cells with surrogate-fetal 2W1S specificity. A. Schematic of the Act-2W1S transgene. The coding sequence for the immune-dominant b peptide 2W1S55-68 was placed between the full-length chicken ovalbumin protein and D transmembrane domain. B. Schematic illustrating transformation of ovalbumin and

2W1S55-68 into surrogate fetal antigens by mating non-transgenic C57BL/6 females with Act-2W1S Balb/c males. C. Levels of fetal OVA+ DNA recovered from systemic tissues of day 30 post-partum females following allogeneic mating described in panel B (blue) compared with virgin female controls (black). D. Composite data showing frequency of FOXP3 expression among all CD4+ cells in virgin (black) or females 30 days postpartum

145 from matings described in panel B and treated with isotype (blue) or anti-OVA (red) antibodies following delivery.

146 A 2W1S OVA+/ X H-2d/d H-2b/b I-Ad/d I-Ab/b 100%

2W1S 2W1S 2W1S 2W1S H-2b/d OVA+/- OVA+/- OVA+/- OVA+/- I-Ab/d

Extended Data Figure 2. Mating strategy allowing total depletion of OVA+ fetal cells in mothers. Balb/c male mice expressing 2W1S and OVA transgene were screened for homozygosity and used to sire allogeneic pregnancies in non-transgenic C57BL/6 females allowing depletion of all fetal cells with anti-OVA antibody.

Extended Data Figure 2

147 A

GITR CD25 B H-2b/b Postpartum day 30 (Balb/c-2W1S+ father) I-Ab/b Isotype or anti-OVA treated

+DT following 2’ fetal-antigen adoptive transfer re-stimulation

FOXP3DTR/DTR CD45.1 2W1S +/+ CD90.2 X OVA H-2d/d I-Ad/d CD45.2 CD90.1

FOXP3WT/WT CD45.2 CD90.2

Extended Data Figure 3. Strategy for characterization of exFoxp3 CD4+ cells with fetal-specificity. A. Histogram plots showing cell-surface GITR and CD25 (IL-2Ra) expression on FOXP3+ (green) compared with exFoxp3 (orange) CD4+ dTomato+ with fetal I-Ab:2W1S specificity. B. Schematic demonstrating adoptive transfer of congenically marked splenocytes from FOXP3WT/WT and FOXP3DTR/DTR females 30 days following primary allogeneic pregnancy sired by Balb/c-2W1S+ fathers and treated with isotype or anti-OVA antibody following delivery. Following adoptive transfer of splenocytes into naive recipients, donor Tregs from Foxp3DTR/DTR were ablated with DT Extended Data Figure 3 treatment. Accumulation of donor fetal-2W1S specific CD4+ Tregs was analyzed following secondary fetal-antigen re-stimulation by mating recipient with Balb/c-2W1S+ father.

148 References 1 Jiang, T. T. et al. Regulatory T cells: new keys for further unlocking the enigma of fetal tolerance and pregnancy complications. Journal of immunology 192, 4949- 4956, doi:10.4049/jimmunol.1400498 (2014). 2 Somerset, D., Zheng, Y., Kilby, M., Sansom, D. & Drayson, M. Normal human pregnancy is associated with an elevation in the immune suppressive CD25+ CD4+ regulatory T-cell subset. Immunology 112, 38-43 (2004). 3 Heikkinen, J., Mottonen, M., Alanen, A. & Lassila, O. Phenotypic characterization of regulatory T cells in the human decidua. Clinical and experimental immunology 136, 373-378, doi:10.1111/j.1365-2249.2004.02441.x (2004). 4 Mahmoud, F. et al. Pregnancy-associated changes in peripheral blood lymphocyte subpopulations in normal Kuwaiti women. Gynecol Obstet Invest 52, 232-236 (2001). 5 Sasaki, Y. et al. Proportion of peripheral blood and decidual CD4+ CD25(bright) regulatory T cells in pre-eclampsia. Clinical and experimental immunology 149, 139- 145 (2007). 6 Sasaki, Y. et al. Decidual and peripheral blood CD4+CD25+ regulatory T cells in early pregnancy subjects and spontaneous abortion cases. Mol Hu Reprod 10, 347-353 (2004). 7 Toldi, G. et al. Decreased number of FoxP3+ regulatory T cells in preeclampsia. Acta obstetricia et gynecologica Scandinavica 87, 1229-1233 (2008). 8 Yang, H. et al. Proportional change of CD4+CD25+ regulatory T cells in decidua and peripheral blood in unexplained recurrent spontaneous abortion patients. Fertil Steril 89, 656-661 (2008). 9 Rowe, J. H., Ertelt, J. M., Xin, L. & Way, S. S. Pregnancy imprints regulatory memory that sustains anergy to fetal antigen. Nature 490, 102-106 (2012). 10 Rowe, J. H., Ertelt, J. M., Aguilera, M. N., Farrar, M. A. & Way, S. S. Foxp3(+) regulatory T cell expansion required for sustaining pregnancy compromises host defense against prenatal bacterial pathogens. Cell Host Microbe 10, 54-64 (2011). 11 Campbell, D., MacGillivray, I. & Carr-Hill, R. Pre-eclampsia in second pregnancy. Br J Obstet Gynaecol 92, 131-140 (1985). 12 Li, D. K. & Wi, S. Changing paternity and the risk of preeclampsia/eclampsia in the subsequent pregnancy. American journal of epidemiology 151, 57-62 (2000). 13 Saftlas, A. F. et al. Abortion, changed paternity, and risk of preeclampsia in nulliparous women. American journal of epidemiology 157, 1108-1114 (2003). 14 Trupin, L., Simon, L. & Eskenazi, B. Change in paternity: a risk factor for preeclampsia in multiparas. Epidemiology 7, 240-244 (1996). 15 Homann, D., Teyton, L. & Oldstone, M. Differential regulation of antiviral T-cells immunity results stable CD8+ by declining CD4+ T-cell memory. Nat Med 7, 913-919 (2001). 16 Belkaid, Y., Piccirillo, C. A., Mendez, S., Shevach, E. M. & Sacks, D. L. CD4+CD25+ regulatory T cells control Leishmania major persistence and immunity. Nature 420, 502-507, doi:10.1038/nature01152 (2002).

149 17 Nelson, R. W., McLachlan, J. B., Kurtz, J. R. & Jenkins, M. K. CD4+ T cell persistence and function after infection are maintained by low-level peptide:MHC class II presentation. Journal of immunology 190, 2828-2834 (2013). 18 Uzonna, J. E., Wei, G., Yurkowski, D. & Bretscher, P. Immune elimination of Leishmania major in mice: implications for immune memory, vaccination, and reactivation disease. Journal of immunology 167, 6967-6974 (2001). 19 Rosenblum, M. et al. Response to self antigen imprints regulatory memory in tissues. Nature 480, 538-542 (2011). 20 Rowe, J. H., Ertelt, J. M., Xin, L. & Way, S. S. Listeria monocytogenes cytoplasmic entry induces fetal wastage by disrupting maternal FoxP3+ regulatory cell-sustained fetal tolerance. PLoS Pathogens 8 (2012). 21 Gammill, H. & Nelson, J. Naturally acquired microchimerism. Int J Dev Biol 54, 531- 543 (2010). 22 Schroder, J. & De la Chapelle, A. Fetal lymphocytes in the maternal blood. Blood 39, 153-162 (1972). 23 Walknowska, J., Conte, F. A. & Grumbach, M. M. Practical and theoretical implications of fetal-maternal lymphocyte transfer. Lancet 1, 1119-1122 (1969). 24 Hall, J. M. et al. Detection of maternal cells in human umbilical cord blood using fluorescence in situ hybridization. Blood 86, 2829-2832 (1995). 25 Bianchi, D. W., Zickwolf, G., Weil, G., Sylvester, S. & DeMaria, M. Male fetal progenitor cells persist in maternal blood for as long as 27 years postpartum. PNAS 93, 705- 708 (1996). 26 Evans, P. C. et al. Long-term fetal microchimerism in peripheral blood mononuclear cell subsets in healthy women and women with scleroderma. Blood 93, 2033-2037 (1999). 27 Khosrotehrani, K., Johnson, K. L., Cha, D. H., Salomon, R. N. & Bianchi, D. W. Transfer of fetal cells with multilineage potential to maternal tissue. JAMA : the journal of the American Medical Association 292, 75-80, doi:10.1001/jama.292.1.75 (2004). 28 Kinder, J. M. et al. Cross-generational reproductive fitness enforced by microchimeric maternal cells. Cell 162, 505-515, doi:10.1016/j.cell.2015.07.006 (2015). 29 Moon, J. J. et al. Quantitative impact of thymic selection on Foxp3+ and Foxp3- subsets of self-peptide/MHC class II-specific CD4+ T cells. Proc. Natl. Acad. Sci. U.S.A. 108, 14602-14607 (2011). 30 Moon, J. et al. Naive CD4(+) T cell frequency varies for different epitopes and predicts repertoire diversity and response magnitude. Immunity 27, 203-213 (2007). 31 Miyao, T. et al. Plasticity of Foxp3(+) T cells reflects promiscuous Foxp3 expression in conventional T cells but not reprogramming of regulatory T cells. Immunity 36, 262-275, doi:10.1016/j.immuni.2011.12.012 (2012). 32 Rubtsov, Y. P. et al. Stability of the regulatory T cell lineage in vivo. Science 329, 1667-1671, doi:10.1126/science.1191996 (2010). 33 Bailey-Bucktrout, S. L. et al. Self-antigen-driven activation induces instability of regulatory T cells during an inflammatory autoimmune response. Immunity 39, 949- 962, doi:10.1016/j.immuni.2013.10.016 (2013).

150 34 Kim, J. M., Rasmussen, J. P. & Rudensky, A. Y. Regulatory T cells prevent catastrophic autoimmunity throughout the lifespan of mice. Nat Immunol 8, 191-197 (2007). 35 Moon, J. J. et al. Tracking epitope-specific T cells. Nat. Protoc. 4, 565-581 (2009).

151 Chapter 7: Summary and Discussion

152 Overall Summary

Bi-directional transfer of cells between mother and offspring during pregnancy, and the long-term persistence of these genetically foreign cells in both individuals after parturition has been well described1,2. Despite many studies that have aimed to characterize the tissue distribution, cellular phenotype, and biological consequences of these genetically foreign cells, two overarching questions remained unanswered:

1. Is the transfer of these cells deliberate or accidental? If intentional, then what are the biological benefits provided by maternal and fetal cells that favor their evolutionary conservation across mammalian species?

2. How do these genetically foreign cells transferred during reproduction avert rejection and promote their long-term survival?

So, far these gaps in knowledge reflect technical limitations in the ability to identify immune cells with defined specificity to the developmentally relevant antigens expressed by these cells combined with the lack of tools allowing selective manipulation of these cells in vivo. Broadly speaking, the goal of these studies was to apply innovate transgenic mouse matings together with cutting-edge cellular immunological tools to allow tracking of immune components with defined-specificity to the foreign antigens expressed by microchimeric cells together with selective manipulation of these genetically discordant cells in vivo.

Since each individual has a mother, we reasoned applying this approach to first tackle the questions above in the context of maternal microchimerism would provide the most broadly relevant findings and may also elucidate similar mechanisms that are employed by fetal microchimeric cells. The findings outlined in Chapter 5 of this dissertation

153 reveal novel reproductive benefits provided by maternal microchimeric cells that promote cross-generational conservation of non-inherited maternal traits. Our major findings are outlined below:

Chapter 5: Cross-generational reproductive fitness enforced by microchimeric maternal cells

1. Postnatal retention of maternal microchimeric cells drives sustained accumulation of immune suppressive regulatory T cells (Tregs) with specificity to the non-inherited maternal antigens (NIMA) they express (Fig. 1-2).

2. Selective seeding of female compared with male reproductive tissues parallels increased expansion of NIMA-specific Tregs in female offspring (Fig 3).

3. During allogeneic pregnancies in female offspring, expanded Tregs with NIMA- specificity mediate resiliency against infectious and non-infections pregnancy complications associated with fractured fetal tolerance only when the father expresses MHC-haplotypes that overlap to NIMA-MHC expressed by the maternal microchimeric cells (Fig. 4-6).

4. In offspring, the sustained accumulation of Tregs with NIMA-specificity and reproductive benefits associated with their expansion requires persistence of maternal microchimeric cells, as each is lost when maternal cells are depleted (Fig. 2, 5-6).

Based on our findings in Chapter 5, we re-directed these transgenic mouse and cellular immunological tools to investigate if similar reproductive benefits and tolerance mechanisms were responsible for the long-term retention of fetal microchimeric cells in mothers. Previous studies in our lab uncovered maternal Treg memory to paternal antigens expressed by the fetus is retained in mothers following primary pregnancy and

154 parallels increased resiliency against fetal wastage stemming from fractured fetal tolerance during secondary pregnancy3. Therefore, we set out to determine if cross-talk between fetal microchimeric cells that express these foreign paternal antigens is responsible for the postpartum persistence of maternal Treg memory. Our studies described in Chapter 6 of this dissertation uncover maternal Treg memory to paternal- fetal antigen is durable and is maintained even despite depletion of fetal microchimeric cells postpartum. These findings are described below:

Chapter 6: Sustained protection against fetal wastage conferred by prior pregnancy despite numerical loss of maternal regulatory CD4+ T cell memory.

1. Fetal microchimeric cells are retained postpartum and parallel systemic accumulation of maternal Tregs with corresponding fetal antigen specificity (Fig 2).

2. Enriched maternal memory Tregs decline following depletion of fetal microchimeric cells postpartum (Fig 2).

3. Protective capacity and secondary expansion of maternal memory Tregs is retained despite fetal microchimeric cell depletion (Fig. 1 and Fig. 3).

4. Retained functional memory following depletion of fetal microchimeric cells is associated with accumulation of fetal-specific CD4+ T cells with transiently silenced FOXP3 expression that remain poised for FOXP3 re-expression upon secondary stimulation (Fig. 4).

Together these findings highlight important differences in the requirement for microchimeric cells to maintain immune tolerance to NIMA compared with fetal-paternal

155 antigens. While this could result from differences in immunological maturity between mother and fetus at the time of foreign antigen exposure, it also advocates careful dissection of the cellular and molecular phenotypes of microchimeric fetal and maternal cells responsible for driving these tolerogenic effects.

156 Discussion

Successful reproduction is the cornerstone for the survival of any species. In turn, selective pressures favor the constant refinement of methods of procreation to promote enhanced reproductive fitness. The evolution of physiological adaptations such as the placenta allowed the emergence of viviparity, a new method of reproduction that increased parental investment in fetal development in return for enhanced offspring survival4. In addition to these physiological adaptations the success of viviparity necessitated immunological shifts in the mother that allowed survival and longer-term growth of the genetically foreign fetus5-8. Furthermore, in humans the with more mature development and peripheral seeding of fetal adaptive immune components prior to birth required additional compensatory mechanisms to avert fetal rejection of the mother9,10.

Thus, the biological imperative of reproductive success has forced the development of multiple non-overlapping mechanisms that act both locally and systemically during gestation to mediate bi-directional maternal-fetal tolerance.

Remarkably, while these immune shifts that favor tolerance in both mother and baby develop to promote pregnancy success, immune tolerance to these developmentally relevant foreign antigens persists long after delivery in both mothers and offspring. For fetal antigens, maternal tolerance is sustained following initial pregnancy and explains why prior successful pregnancy affords partner-specific protection in women against complications associated with fractured fetal tolerance (e.g. preeclampsia) in future pregnancies3,11,12. On the other hand, numerous examples of persistent postnatal tolerance to non-inherited maternal antigens also exist including increased long-term

157 survival of solid organ grafts from siblings matched from non-inherited maternal human leukocyte and reduced severity of graft versus host disease (GVHD) after stem cell transplantation among recipients of non-inherited maternal antigen-compatible donor stem cells13,14. Together, these remarkable human examples demonstrate immunological tolerance imprinted by developmentally exposed foreign antigens is retained in both mother and her offspring following delivery.

Along with these remarkable examples of expanded immune tolerance is the bi- directional trafficking of cells between mother and child that ubiquitously occurs during pregnancy. Interestingly, these genetically foreign cells transferred during pregnancy persist in mother and child long after pregnancy as exceptionally rare microchimeric cells. For example, Y-chromosome expressing cells have been identified in women who have previously given birth to a son, even decades after pregnancy15-19. Similarly, maternal cells that express discordant DNA have also been recovered across multiple tissues in offspring postnatally20-22. Despite the parallel existence of tolerance to developmentally exposed antigens and the long-term persistence of microchimeric cells that express the same antigens, the relationship between these two phenomena was unknown. While persistent immune tolerance may prevent the rejection of these genetically foreign cells, another potential possibility is that microchimeric cells intentionally drive immune tolerance to actively avoid rejection. This cause and effect relationship was previously unknown and stemmed from technical limitations precluding the manipulation of microchimeric cells and immune cells with specificity to the foreign antigens they expressed.

158 In Chapter 5 we addressed this critical gap in knowledge for maternal microchimeric cells and persistent tolerance to NIMA. Here, we employed an innovative mating strategy utilizing female mice with heterozygous expression of a transgene encoding

b ubiquitous cell-surface expression of the model antigen I-A :2W1S55-68 together with the ovalbumin (OVA) protein23. In turn, mating these females with non-transgenic males transformed these model antigens into surrogate NIMAs and allowed us to track in offspring how developmental exposure to NIMA altered their immune responsiveness.

Reciprocally, cell-surface expression of OVA allowed selective targeted and depletion of maternal OVA+ cells in vivo. We found, postnatal retention of maternal microchimeric cells paralleled the sustained expansion of immune suppressive Tregs with NIMA-2W1S specificity. Furthermore, the persistence of maternal microchimeric cells was absolutely required as depletion of maternal cells with α-OVA antibodies triggered a reciprocal decline in Tregs with NIMA-specificity. Therefore, our results demonstrate persistence of genetically foreign maternal microchimeric cells is not an accident or developmental remnant, but instead microchimeric maternal cells actively force immune tolerance to the foreign NIMAs they express to avoid rejection.

While these findings definitively addressed the chicken and egg relationship between maternal microchimeric cells and NIMA tolerance, the biological benefits that promote retention of these genetically foreign cells remained an unanswered question. Some evidence suggested that tolerance to NIMA in humans with a full complement of immune cells at birth was required to prevent maternal rejection by the developing fetus9,10. However, this would only necessitate generation of NIMA tolerance in utero and suggests its postnatal persistence is instead just a developmental remnant. Our

159 findings, demonstrating maternal microchimeric cells are intentionally retained together with studies indicating retention of microchimeric cells and NIMA tolerance is conserved across mammalian species suggests other teleological benefits have promoted its evolutionary conservation1,24-27.

To this end, we were struck by the enhanced systemic accumulation of Tregs with

NIMA-specificity in female offspring that paralleled the selective retention of maternal microchimeric cells in female compared with male reproductive tissue (uterus vs prostate). These findings, together with previous studies demonstrating the importance of maternal Tregs with specificity to fetal antigen in enforcing fetal tolerance3,28,29, suggested to us that retained maternal microchimeric cells in female offspring, may participate in enhancing the success of her future pregnancies. Remarkably, Tregs with specificity to NIMA in female offspring demonstrated enhanced accumulation in comparison to naive females following stimulation with paternal-fetal antigen with overlapping NIMA-specificity. This expansion required commonality between fetal- paternal and NIMA antigen as NIMA-specific Tregs did not expand in pregnancies sired by males that did not express NIMA. Thus, this retained pool of Tregs with specificity to

NIMA in female offspring expands further following stimulation with their cognate antigen.

Although this was the first pregnancy for this female, overlap between paternal MHC- haplotype and the non-inherited MHC-haplotype of her mother resulted in a functional response analogous to protective memory demonstrated for maternal Tregs during secondary pregnancies3. Specifically, commonality between paternal-fetal antigen and

NIMA during pregnancy, resulted in significantly reduced fetal wastage associated with

160 fractured fetal tolerance stemming from both infectious (prenatal Lm) and non-infectious

(partial maternal Treg ablation) insults when compared to naive pregnant females or

NIMA-exposed females mated by males expressing third-party MHC haplotypes. In addition to the partner-specificity the persistence of maternal microchimeric cells that sustains expansion of Tregs with NIMA-specificity was also required. In turn fetal wastage stemming from fractured fetal tolerance rebounded following depletion of maternal microchimeric cells with α-OVA antibody treatment despite commonality between fetal-paternal and NIMA antigens.

Together, these studies Mother demonstrate maternal

Enforced fetal microchimeric cells vertically tolerance during pregnancy sired by male with shared transferred during gestation are Female NIMA specificity offspring Background intentionally retained and promote susceptibility to complications in cross-generational reproductive pregnancies sired by males fitness in female offspring (Fig.1). Microchimeric without shared maternal cells NIMA specificity enforce tolerance Thus, in nature females that to non-inherted maternal antigens (NIMA) encounter males with commonality to NIMA will experience reinforced Figure 1. Cross-generational reproductive fitness enforced by microchimeric maternal cells fetal tolerance during pregnancy

(Fig. 1). This increased resiliency enhances the chances of fetal survival and promotes the retention of non-inherited traits in future generations. In turn, this genetic benefit provided to the individual mother is most likely counterbalanced by selective pressure

161 favoring the expansion of MHC-haplotypes necessary for combatting spread of pathogen across the population30.

While we have established reproductive benefits for persistent maternal microchimeric cells, our model allowing postnatal depletion of maternal microchimeric cells uniquely poises us to address multiple other questions regarding the costs and benefits of being constitutively chimeric. For example, the transfer of maternal microchimeric cells parallels the development of the offspring’s immune system and suggests these cells may act to somehow stimulate offspring immune maturation31. We are now poised to test this idea through assessing neonatal susceptibility to infection following depletion of maternal microchimeric cells in the early postnatal phase. If ongoing stimulation from maternal microchimeric cells is necessary to promote immunity to foreign pathogens, we expect depletion of maternal cells post-natally would increase the neonate’s susceptibility to infection in the early neonatal period.

In Chapter 6 we re-directed our approach for elucidating biological benefits afforded by maternal microchimeric cells to investigate the requirement for fetal microchimeric cells to provide partner-specific benefits related to maternal Treg memory during secondary pregnancy. We reasoned since the requirement for antigenic stimulation for retained memory has been established for effector CD4+ T cells their Foxp3+ CD4+ Treg counterparts may also require similar low-level stimulatory reminders32-35. In turn, fetal microchimeric cells that persist across multiple tissues in mothers following delivery, could provide a potential source of low-level antigen stimulation required to sustain maternal Treg memory.

162 Here, we employed a previously described mating strategy from our lab whereby male mice with heterozygous expression of the 2W1S-OVA encoding transgene described above were mated to non-transgenic females, transforming these model antigens into surrogate fetal antigens3,23. In this case, the same tools could be applied to track maternal immune cells with defined specificity to these surrogate fetal antigens in their corresponding response to in vivo depletion of OVA+ expressing fetal cells we observed in mothers following delivery36. Similar to our results tracking Tregs with specificity to

NIMA, here maternal Tregs with fetal-2W1S specificity also declined to background levels following depletion of fetal microchimeric cells. Thus maternal Tregs with fetal specificity require persistent fetal microchimeric cells for retained quantitative memory.

In stark contrast however to the reproductive benefits provided by sustained Tregs with

NIMA specificity36, exposure to fetal antigens during initial pregnancy primed durable functional maternal Treg memory during secondary pregnancy. Specifically, while the quantitative retention of maternal Tregs with fetal specificity was lost following postpartum depletion of fetal microchimeric cells, protection from fetal wastage stemming from fractured fetal tolerance triggered by prenatal Lm infection was retained despite the loss of fetal cells. In turn, the retained functional properties of maternal

Tregs following fetal microchimeric cell depletion was associated with the ability of fetal- specific Tregs to re-accumulate similarly following secondary fetal antigen stimulation in

α-OVA treated or untreated postpartum females. These findings highlight distinct differences in the requirement for persistence of maternal compared with fetal microchimeric cells for retained functional memory. While all reproductive benefits

163 associated with retained maternal microchimeric cell are lost following their depletion, functional Treg memory to fetal antigen seems to be more durable36.

We are currently designing and optimizing new strategies to more comprehensively investigate the differences between antigenic stimulation in these two Figure 2. Magnetic bead enrichment combined with multi-parameter flow developmentally cytometry adapted to identify rare microchimeric maternal cells. A. Allogeneic mating between H-2k/k males and H-2d/b GFP+/- females on the B6 background that simultaneously transforms H-2d and GFP into surrogate NIMA in ¼ of the offspring. important contexts. B. Enrichment of H-2d microchimeric maternal cells using fluorochrome (PE) conjugated antibody and anti-PE conjugated magnetic beads. C. Application to One strategy identify intact microchimeric maternal cells among single cell suspensions from the uterus of naive control compared with NIMA-H-2d, NIMA-GFP virgin 8-week old female mice after enrichment with PE conjugated anti-H-2Kd (clone SF1-1) antibody. involves re-focusing (D) Verification of H-2Kd GFP double positive cells (red line) compared with endogenous uterine cells (gray shaded) after staining with a non-overlapping clone of anti-H-2d antibody (anti-H-2Kd/Dd, clone 34-1-2S) and anti-GFP, along with their tools in our lab for relative expression of molecules required for T cell antigen-presentation (MHC class II [I-Ad]) and/or co-stimulation/co-inhibition (CD40, CD80, PDL-1). tracking rare endogenous T cells with defined specificity to instead focus on recovering intact microchimeric cells to allow more broad characterization of the cellular and molecular phenotype employed by these cells to promote tolerance (Fig. 2). For example, to allow enrichment of maternal microchimeric cells in offspring, we have employed a mating scheme that transforms MHC haplotype alleles (e.g. H-2d) into surrogate NIMA (Fig.

164 2A). Additional specificity that allows more precise identification of these rare cells is added by using mice that also contain ubiquitous expression of GFP that simultaneously transforms GFP into a surrogate NIMA. We have begun applying this unbiased enrichment approach in proof of principle studies that have identified 191 – 693 H-2d-

GFP+ maternal cells among ~20 million uterine cells within individual female mice (Fig.

2B,C). Interestingly, in our initial studies we have observed near uniform expression of

MHC class II along with other T cell co-stimulatory (CD40, CD80) and co-inhibitory

(PDL-1) molecules (Fig. 2D). This suggests microchimeric maternal cells have the capacity to directly present antigen to T cells themselves. In additional studies we plan to further characterize the cellular identify of these cells through analysis of cell-surface molecules that define specific cell subsets including the broad leukocyte marker CD45 but also individual leukocyte subsets (CD11c, CD11b, CD19/B220, F4/80, CD3, etc.).

We reason this approach that shifts away from identification of microchimeric cells based on DNA (PCR or FISH) that can only establish the existence of these cells will instead allow analysis of the molecular basis of how these cells work. However, while flow-cytometry analysis of post-enriched maternal microchimeric cells allows initial characterization of their cell-surface phenotype, it remains inadequate for addressing how microchimeric cells work. For example, given that many fluorescent channels are needed to accurately identify these rare cells, we are limited to analyzing on a few molecules at a time. Furthermore, our analysis is restricted by the availability of antibodies to individual molecules and therefore biases our analysis to these pathways.

To overcome these limitations, we plan to combine our microchimeric cell enrichment approach together with single-cell gene expression analysis (RNAseq) that will allow

165 more broad unbiased analysis of individual maternal and fetal microchimeric cells across a wide range of tissues.

Using lineage-fate tracking mice that allowed identification of cells with prior Foxp3 expression, we observed loss of maternal Tregs with fetal specificity following fetal microchimeric cell depletion reflected a reciprocal accumulation of CD4+ cells that had silenced Foxp3 expression. In mothers depleted of fetal microchimeric cells, these exFoxp3 cells with fetal specificity quickly regained Foxp3 expression following secondary fetal-antigen re-stimulation accumulating to levels observed in untreated females. Thus, the generation of exFoxp3 cells with fetal specificity allows retention of maternal memory to fetal antigen despite depletion of fetal microchimeric cells.

However, one important question that remains is why silencing Foxp3 expression and transition to an exFoxp3 cell is favored over remaining Foxp3+.

We hypothesize this cell fate choice to silence Foxp3 expression may facilitate a lower rate of proliferative turnover and extend the life of these memory cells. In turn, these exFoxp3 cells would consume fewer resources, but more importantly this quiescent state may favor long-term stability and survival of maternal tolerance in the absence of fetal antigen stimulation. To test this, we plan to measure the proliferation (KI-67 expression) of Foxp3+ versus exFoxp3 cells in addition to assessing cell death (annexin

V expression levels). If the transition to exFoxp3 cells favors increased long-term survival, we would expect decreased proliferation and death of exFoxp3 compared with

Foxp3+ cells.

166 Furthermore, while our studies rely on artificial depletion of fetal microchimeric cells through use of anti-OVA antibodies we wondered if parallel mechanisms might exist in nature. While natural reductions in fetal microchimeric cells is expected over time evidence in humans suggests fetal cells can be found in mothers decades after delivery.

However, we wondered whether microchimeric cells acquired from different sources

(fetal vs fetal or maternal vs fetal) could persist together or instead the acquisition of new microchimeric cells may facilitate the natural displacement of previous residents. In nature, mating with a different partner or even the same partner would no doubt yield a new set of fetal microchimeric cells. In this case, if these new cells displaced the previous microchimeric cells, silencing of Foxp3 expression for maternal Tregs with specificity to fetal antigen encountered during the first pregnancy would allow retention of regulatory memory if those antigens were encountered again in the future. To test this, we have set up secondary allogeneic matings where males expressing third-party

MHC haplotypes to fetal-paternal MHC encountered during primary pregnancy are used to displace resident fetal microchimeric cells. Following this second mating, we will quantify the level of fetal microchimerism from the first and secondary pregnancies since fetal cells from either pregnancy express discordant DNA to the mother. In addition, we will also examine the accumulation of exFoxp3 cells with fetal antigen specificity following displacement that could allow retention of maternal Treg memory.

In broad context, if accumulation of exFoxp3 cells with fetal specificity following loss of fetal microchimerism allows retained functional Treg memory in pregnancy, it is also possible these mechanisms has been co-opted to provide sustained tolerance in other contexts. For instance foreign, harmless antigens from food or commensal microbes

167 represent another set of non-self antigens that our immune systems have adapted to tolerate. However, exposure to food proteins and commensal antigens is rapidly evolving over time providing mostly transient stimulation of Tregs with specificity for these antigens. Therefore while transition to an exFoxp3 cell may allow retained memory following pregnancy this tolerance mechanism may also extend more broadly to encompass other beneficial non-self antigens.

Broadly speaking, we have shown maternal and fetal cells are transferred during pregnancy and their long-term persistence following delivery affords reproductive benefits during subsequent pregnancies. However, while maternal microchimeric cells are absolutely required to sustain expanded accumulation of Tregs with NIMA- specificity in offspring and tolerance to NIMA the same is not true for fetal microchimeric cells. In turn, while quantitative retention of maternal Treg with specificity fetal antigens requires fetal microchimeric cells, the protective properties of maternal Treg memory are durable even following loss of fetal microchimerism. This highlights potentially interesting discordances in the tolerogenic properties of maternal versus fetal microchimeric cells that will require more definitive and in-depth analysis of live, intact microchimeric cells to distinguish how these cells differ in their ability to promote immune tolerance to reproductive antigens.

168 References

1 Dutta, P. & Burlingham, W. J. Microchimerism: tolerance vs. sensitization. Current opinion in organ transplantation 16, 359-365, doi:10.1097/MOT.0b013e3283484b57 (2011).

2 Gammill, H. & Nelson, J. Naturally acquired microchimerism. Int J Dev Biol 54, 531-543 (2010).

3 Rowe, J. H., Ertelt, J. M., Xin, L. & Way, S. S. Pregnancy imprints regulatory memory that sustains anergy to fetal antigen. Nature 490, 102-106 (2012).

4 Roberts, R. M., Green, J. A. & Schulz, L. C. The evolution of the placenta. Reproduction 152, R179-189, doi:10.1530/REP-16-0325 (2016).

5 Erlebacher, A. Mechanisms of T cell tolerance towards the allogeneic fetus. Nature reviews. Immunology 13, 23-33, doi:10.1038/nri3361 (2013).

6 Erlebacher, A. Immunology of the maternal-fetal interface. Annual review of immunology 31, 387-411, doi:10.1146/annurev-immunol-032712-100003 (2013).

7 Jiang, T. T. et al. Regulatory T cells: new keys for further unlocking the enigma of fetal tolerance and pregnancy complications. Journal of immunology 192, 4949- 4956, doi:10.4049/jimmunol.1400498 (2014).

8 Munoz-Suano, A., Hamilton, A. B. & Betz, A. G. Gimme shelter: the immune system during pregnancy. Immunol Rev 241, 20-38, doi:10.1111/j.1600- 065X.2011.01002.x (2011).

9 Mold, J. E. & McCune, J. M. Immunological tolerance during fetal development: from mouse to man. Adv. Immunol. 115, 73-111 (2012).

10 Mold, J. E. et al. Maternal alloantigens promote the development of tolerogenic fetal regulatory T cells in utero. Science 322, 1562-1565 (2008).

11 Campbell, D., MacGillivray, I. & Carr-Hill, R. Pre-eclampsia in second pregnancy. Br J Obstet Gynaecol 92, 131-140 (1985).

12 Trupin, L., Simon, L. & Eskenazi, B. Change in paternity: a risk factor for preeclampsia in multiparas. Epidemiology 7, 240-244 (1996).

13 Burlingham, W. J. et al. The effect of tolerance to noninherited maternal HLA antigens on the survival of renal transplants from sibling donors. N. Engl. J. Med. 339, 1657-1664 (1998).

14 Matsuoka, K. et al. Fetal tolerance to maternal antigens improves the outcome of allogeneic bone marrow transplantation by a CD4+ CD25+ T-cell-dependent mechanism. Blood 107, 404-409, doi:10.1182/blood-2005-07-3045 (2006).

169 15 Bianchi, D. W., Zickwolf, G., Weil, G., Sylvester, S. & DeMaria, M. Male fetal progenitor cells persist in maternal blood for as long as 27 years postpartum. PNAS 93, 705-708 (1996).

16 Evans, P. C. et al. Long-term fetal microchimerism in peripheral blood mononuclear cell subsets in healthy women and women with scleroderma. Blood 93, 2033-2037 (1999).

17 Bayes-Genis, A. et al. Identification of male cardiomyocytes of extracardiac origin in the hearts of women with male progeny: male fetal cell microchimerism of the heart. J Heart Lung Transplant 24, 2179-2183, doi:10.1016/j.healun.2005.06.003 (2005).

18 O'Donoghue, K. et al. Microchimerism in female bone marrow and bone decades after fetal mesenchymal stem-cell trafficking in pregnancy. Lancet 364, 179-182, doi:10.1016/S0140-6736(04)16631-2 (2004).

19 Stevens, A. M. et al. Liver biopsies from human females contain male hepatocytes in the absence of transplantation. Lab Invest 84, 1603-1609, doi:10.1038/labinvest.3700193 (2004).

20 Hall, J. M. et al. Detection of maternal cells in human umbilical cord blood using fluorescence in situ hybridization. Blood 86, 2829-2832 (1995).

21 Jonsson, A. M., Uzunel, M., Gotherstrom, C., Papadogiannakis, N. & Westgren, M. Maternal microchimerism in human fetal tissues. American journal of obstetrics and gynecology 198, 325 e321-326, doi:10.1016/j.ajog.2007.09.047 (2008).

22 Loubiere, L. S. et al. Maternal microchimerism in healthy adults in lymphocytes, monocyte/macrophages and NK cells. Lab Invest 86, 1185-1192, doi:10.1038/labinvest.3700471 (2006).

23 Moon, J. J. et al. Quantitative impact of thymic selection on Foxp3+ and Foxp3- subsets of self-peptide/MHC class II-specific CD4+ T cells. Proc. Natl. Acad. Sci. U.S.A. 108, 14602-14607 (2011).

24 Andrassy, J. et al. Tolerance to noninherited maternal MHC antigens in mice. Journal of immunology 171, 5554-5561 (2003).

25 Billingham, R. E., Brent, L. & Medawar, P. B. Actively acquired tolerance of foreign cells. Nature 172, 603-606 (1953).

26 Owen, R. D. Immunogenetic Consequences of Vascular Anastomoses between Bovine Twins. Science 102, 400-401 (1945).

170 27 Picus, J., Aldrich, W. R. & Letvin, N. L. A naturally occurring bone-marrow- chimeric primate. I. Integrity of its immune system. Transplantation 39, 297-303 (1985).

28 Aluvihare, V., Kallikourdis, M. & Betz, A. Regulatory T cells mediate maternal tolerance to the fetus. Nat Immunol 5, 266-271 (2004).

29 Rowe, J. H., Ertelt, J. M., Aguilera, M. N., Farrar, M. A. & Way, S. S. Foxp3(+) regulatory T cell expansion required for sustaining pregnancy compromises host defense against prenatal bacterial pathogens. Cell Host Microbe 10, 54-64 (2011).

30 Spurgin, L. G. & Richardson, D. S. How pathogens drive genetic diversity: MHC, mechanisms and misunderstandings. Proc. Biol. Sci. 277, 979-988 (2010).

31 Stelzer, I. A., Thiele, K. & Solano, M. E. Maternal microchimerism: lessons learned from murine models. J Reprod Immunol 108, 12-25, doi:10.1016/j.jri.2014.12.007 (2015).

32 Belkaid, Y., Piccirillo, C. A., Mendez, S., Shevach, E. M. & Sacks, D. L. CD4+CD25+ regulatory T cells control Leishmania major persistence and immunity. Nature 420, 502-507, doi:10.1038/nature01152 (2002).

33 Homann, D., Teyton, L. & Oldstone, M. Differential regulation of antiviral T-cells immunity results stable CD8+ by declining CD4+ T-cell memory. Nat Med 7, 913- 919 (2001).

34 Nelson, R. W., McLachlan, J. B., Kurtz, J. R. & Jenkins, M. K. CD4+ T cell persistence and function after infection are maintained by low-level peptide:MHC class II presentation. Journal of immunology 190, 2828-2834 (2013).

35 Uzonna, J. E., Wei, G., Yurkowski, D. & Bretscher, P. Immune elimination of Leishmania major in mice: implications for immune memory, vaccination, and reactivation disease. Journal of immunology 167, 6967-6974 (2001).

36 Kinder, J. M. et al. Cross-generational reproductive fitness enforced by microchimeric maternal cells. Cell 162, 505-515, doi:10.1016/j.cell.2015.07.006 (2015).

171 Appendices

172 RightsLink Printable License https://s100.copyright.com/App/PrintableLicenseFrame.jsp?publish...

ELSEVIER LICENSE TERMS AND CONDITIONS Oct 17, 2016

This Agreement between Jeremy M Kinder ("You") and Elsevier ("Elsevier") consists of your license details and the terms and conditions provided by Elsevier and Copyright Clearance Center.

License Number 3971400415880

License date Oct 17, 2016

Licensed Content Publisher Elsevier

Licensed Content Publication Immunity

Licensed Content Title Offspring’s Tolerance of Mother Goes Viral

Licensed Content Author Jeremy M. Kinder,Tony T. Jiang,Sing Sing Way

Licensed Content Date 17 May 2016

Licensed Content Volume 44 Number

Licensed Content Issue 5 Number

Licensed Content Pages 3

Start Page 1085

End Page 1087

Type of Use reuse in a thesis/dissertation

Portion full article

Format both print and electronic

Are you the author of this Yes Elsevier article?

Will you be translating? No

Order reference number

Title of your Tolerogenic Properties of Maternal and Fetal Microchimeric Cells thesis/dissertation

Expected completion date Oct 2016

Estimated size (number of 250 pages)

Elsevier VAT number GB 494 6272 12

Requestor Location Jeremy M Kinder 2639 Observatory Ave #2

CINCINNATI, OH 45208 United States

1 of 6 10/17/16, 10:55 AM RightsLink Printable License https://s100.copyright.com/App/PrintableLicenseFrame.jsp?publish...

Attn: Jeremy M Kinder

Total 0.00 USD

Terms and Conditions INTRODUCTION 1. The publisher for this copyrighted material is Elsevier. By clicking "accept" in connection with completing this licensing transaction, you agree that the following terms and conditions apply to this transaction (along with the Billing and Payment terms and conditions established by Copyright Clearance Center, Inc. ("CCC"), at the time that you opened your Rightslink account and that are available at any time at http://myaccount.copyright.com). GENERAL TERMS 2. Elsevier hereby grants you permission to reproduce the aforementioned material subject to the terms and conditions indicated. 3. Acknowledgement: If any part of the material to be used (for example, figures) has appeared in our publication with credit or acknowledgement to another source, permission must also be sought from that source. If such permission is not obtained then that material may not be included in your publication/copies. Suitable acknowledgement to the source must be made, either as a footnote or in a reference list at the end of your publication, as follows: "Reprinted from Publication title, Vol /edition number, Author(s), Title of article / title of chapter, Pages No., Copyright (Year), with permission from Elsevier [OR APPLICABLE SOCIETY COPYRIGHT OWNER]." Also Lancet special credit - "Reprinted from The Lancet, Vol. number, Author(s), Title of article, Pages No., Copyright (Year), with permission from Elsevier." 4. Reproduction of this material is confined to the purpose and/or media for which permission is hereby given. 5. Altering/Modifying Material: Not Permitted. However figures and illustrations may be altered/adapted minimally to serve your work. Any other abbreviations, additions, deletions and/or any other alterations shall be made only with prior written authorization of Elsevier Ltd. (Please contact Elsevier at [email protected]) 6. If the permission fee for the requested use of our material is waived in this instance, please be advised that your future requests for Elsevier materials may attract a fee. 7. Reservation of Rights: Publisher reserves all rights not specifically granted in the combination of (i) the license details provided by you and accepted in the course of this licensing transaction, (ii) these terms and conditions and (iii) CCC's Billing and Payment terms and conditions. 8. License Contingent Upon Payment: While you may exercise the rights licensed immediately upon issuance of the license at the end of the licensing process for the transaction, provided that you have disclosed complete and accurate details of your proposed use, no license is finally effective unless and until full payment is received from you (either by publisher or by CCC) as provided in CCC's Billing and Payment terms and conditions. If full payment is not received on a timely basis, then any license preliminarily granted shall be deemed automatically revoked and shall be void as if never granted. Further, in the event that you breach any of these terms and conditions or any of CCC's Billing and Payment terms and conditions, the license is automatically revoked and shall be void as if never granted. Use of materials as described in a revoked license, as well as any use of the materials beyond the scope of an unrevoked license, may constitute copyright infringement

2 of 6 10/17/16, 10:55 AM RightsLink Printable License https://s100.copyright.com/App/PrintableLicenseFrame.jsp?publish...

and publisher reserves the right to take any and all action to protect its copyright in the materials. 9. Warranties: Publisher makes no representations or warranties with respect to the licensed material. 10. Indemnity: You hereby indemnify and agree to hold harmless publisher and CCC, and their respective officers, directors, employees and agents, from and against any and all claims arising out of your use of the licensed material other than as specifically authorized pursuant to this license. 11. No Transfer of License: This license is personal to you and may not be sublicensed, assigned, or transferred by you to any other person without publisher's written permission. 12. No Amendment Except in Writing: This license may not be amended except in a writing signed by both parties (or, in the case of publisher, by CCC on publisher's behalf). 13. Objection to Contrary Terms: Publisher hereby objects to any terms contained in any purchase order, acknowledgment, check endorsement or other writing prepared by you, which terms are inconsistent with these terms and conditions or CCC's Billing and Payment terms and conditions. These terms and conditions, together with CCC's Billing and Payment terms and conditions (which are incorporated herein), comprise the entire agreement between you and publisher (and CCC) concerning this licensing transaction. In the event of any conflict between your obligations established by these terms and conditions and those established by CCC's Billing and Payment terms and conditions, these terms and conditions shall control. 14. Revocation: Elsevier or Copyright Clearance Center may deny the permissions described in this License at their sole discretion, for any reason or no reason, with a full refund payable to you. Notice of such denial will be made using the contact information provided by you. Failure to receive such notice will not alter or invalidate the denial. In no event will Elsevier or Copyright Clearance Center be responsible or liable for any costs, expenses or damage incurred by you as a result of a denial of your permission request, other than a refund of the amount(s) paid by you to Elsevier and/or Copyright Clearance Center for denied permissions. LIMITED LICENSE The following terms and conditions apply only to specific license types: 15. Translation: This permission is granted for non-exclusive world English rights only unless your license was granted for translation rights. If you licensed translation rights you may only translate this content into the languages you requested. A professional translator must perform all translations and reproduce the content word for word preserving the integrity of the article. 16. Posting licensed content on any Website: The following terms and conditions apply as follows: Licensing material from an Elsevier journal: All content posted to the web site must maintain the copyright information line on the bottom of each image; A hyper-text must be included to the Homepage of the journal from which you are licensing at http://www.sciencedirect.com/science/journal/xxxxx or the Elsevier homepage for books at http://www.elsevier.com; Central Storage: This license does not include permission for a scanned version of the material to be stored in a central repository such as that provided by Heron/XanEdu. Licensing material from an Elsevier book: A hyper-text link must be included to the Elsevier homepage at http://www.elsevier.com . All content posted to the web site must maintain the copyright information line on the bottom of each image.

3 of 6 10/17/16, 10:55 AM RightsLink Printable License https://s100.copyright.com/App/PrintableLicenseFrame.jsp?publish...

Posting licensed content on Electronic reserve: In addition to the above the following clauses are applicable: The web site must be password-protected and made available only to bona fide students registered on a relevant course. This permission is granted for 1 year only. You may obtain a new license for future website posting. 17. For journal authors: the following clauses are applicable in addition to the above: Preprints: A preprint is an author's own write-up of research results and analysis, it has not been peer-reviewed, nor has it had any other value added to it by a publisher (such as formatting, copyright, technical enhancement etc.). Authors can share their preprints anywhere at any time. Preprints should not be added to or enhanced in any way in order to appear more like, or to substitute for, the final versions of articles however authors can update their preprints on arXiv or RePEc with their Accepted Author Manuscript (see below). If accepted for publication, we encourage authors to link from the preprint to their formal publication via its DOI. Millions of researchers have access to the formal publications on ScienceDirect, and so links will help users to find, access, cite and use the best available version. Please note that Cell Press, The Lancet and some society-owned have different preprint policies. Information on these policies is available on the journal homepage. Accepted Author Manuscripts: An accepted author manuscript is the manuscript of an article that has been accepted for publication and which typically includes author- incorporated changes suggested during submission, peer review and editor-author communications. Authors can share their accepted author manuscript:

- immediately via their non-commercial person homepage or blog by updating a preprint in arXiv or RePEc with the accepted manuscript via their research institute or institutional repository for internal institutional uses or as part of an invitation-only research collaboration work-group directly by providing copies to their students or to research collaborators for their personal use for private scholarly sharing as part of an invitation-only work group on commercial sites with which Elsevier has an agreement - after the embargo period via non-commercial hosting platforms such as their institutional repository via commercial sites with which Elsevier has an agreement

In all cases accepted manuscripts should:

- link to the formal publication via its DOI - bear a CC-BY-NC-ND license - this is easy to do - if aggregated with other manuscripts, for example in a repository or other site, be shared in alignment with our hosting policy not be added to or enhanced in any way to appear more like, or to substitute for, the published journal article.

Published journal article (JPA): A published journal article (PJA) is the deﬁnitive ﬁnal record of published research that appears or will appear in the journal and embodies all

4 of 6 10/17/16, 10:55 AM RightsLink Printable License https://s100.copyright.com/App/PrintableLicenseFrame.jsp?publish...

value-adding publishing activities including peer review co-ordination, copy-editing, formatting, (if relevant) pagination and online enrichment. Policies for sharing publishing journal articles differ for subscription and gold open access articles: Subscription Articles: If you are an author, please share a link to your article rather than the full-text. Millions of researchers have access to the formal publications on ScienceDirect, and so links will help your users to ﬁnd, access, cite, and use the best available version. Theses and dissertations which contain embedded PJAs as part of the formal submission can be posted publicly by the awarding institution with DOI links back to the formal publications on ScienceDirect. If you are afﬁliated with a library that subscribes to ScienceDirect you have additional private sharing rights for others' research accessed under that agreement. This includes use for classroom teaching and internal training at the institution (including use in course packs and courseware programs), and inclusion of the article for grant funding purposes. Gold Open Access Articles: May be shared according to the author-selected end-user license and should contain a CrossMark logo, the end user license, and a DOI link to the formal publication on ScienceDirect. Please refer to Elsevier's posting policy for further information. 18. For book authors the following clauses are applicable in addition to the above: Authors are permitted to place a brief summary of their work online only. You are not allowed to download and post the published electronic version of your chapter, nor may you scan the printed edition to create an electronic version. Posting to a repository: Authors are permitted to post a summary of their chapter only in their institution's repository. 19. Thesis/Dissertation: If your license is for use in a thesis/dissertation your thesis may be submitted to your institution in either print or electronic form. Should your thesis be published commercially, please reapply for permission. These requirements include permission for the Library and Archives of Canada to supply single copies, on demand, of the complete thesis and include permission for Proquest/UMI to supply single copies, on demand, of the complete thesis. Should your thesis be published commercially, please reapply for permission. Theses and dissertations which contain embedded PJAs as part of the formal submission can be posted publicly by the awarding institution with DOI links back to the formal publications on ScienceDirect.

Elsevier Open Access Terms and Conditions You can publish open access with Elsevier in hundreds of open access journals or in nearly 2000 established subscription journals that support open access publishing. Permitted third party re-use of these open access articles is defined by the author's choice of Creative Commons user license. See our open access license policy for more information. Terms & Conditions applicable to all Open Access articles published with Elsevier: Any reuse of the article must not represent the author as endorsing the adaptation of the article nor should the article be modified in such a way as to damage the author's honour or reputation. If any changes have been made, such changes must be clearly indicated. The author(s) must be appropriately credited and we ask that you include the end user license and a DOI link to the formal publication on ScienceDirect. If any part of the material to be used (for example, figures) has appeared in our publication with credit or acknowledgement to another source it is the responsibility of the user to ensure their reuse complies with the terms and conditions determined by the rights holder.

5 of 6 10/17/16, 10:55 AM RightsLink Printable License https://s100.copyright.com/App/PrintableLicenseFrame.jsp?publish...

Additional Terms & Conditions applicable to each Creative Commons user license: CC BY: The CC-BY license allows users to copy, to create extracts, abstracts and new works from the Article, to alter and revise the Article and to make commercial use of the Article (including reuse and/or resale of the Article by commercial entities), provided the user gives appropriate credit (with a link to the formal publication through the relevant DOI), provides a link to the license, indicates if changes were made and the licensor is not represented as endorsing the use made of the work. The full details of the license are available at http://creativecommons.org/licenses/by/4.0. CC BY NC SA: The CC BY-NC-SA license allows users to copy, to create extracts, abstracts and new works from the Article, to alter and revise the Article, provided this is not done for commercial purposes, and that the user gives appropriate credit (with a link to the formal publication through the relevant DOI), provides a link to the license, indicates if changes were made and the licensor is not represented as endorsing the use made of the work. Further, any new works must be made available on the same conditions. The full details of the license are available at http://creativecommons.org/licenses/by-nc-sa/4.0. CC BY NC ND: The CC BY-NC-ND license allows users to copy and distribute the Article, provided this is not done for commercial purposes and further does not permit distribution of the Article if it is changed or edited in any way, and provided the user gives appropriate credit (with a link to the formal publication through the relevant DOI), provides a link to the license, and that the licensor is not represented as endorsing the use made of the work. The full details of the license are available at http://creativecommons.org/licenses/by-nc-nd/4.0. Any commercial reuse of Open Access articles published with a CC BY NC SA or CC BY NC ND license requires permission from Elsevier and will be subject to a fee. Commercial reuse includes:

- Associating advertising with the full text of the Article - Charging fees for document delivery or access - Article aggregation - Systematic distribution via e-mail lists or share buttons

Posting or linking by commercial companies for use by customers of those companies.

20. Other Conditions:

v1.8

Questions? [email protected] or +1-855-239-3415 (toll free in the US) or +1-978-646-2777.

6 of 6 10/17/16, 10:55 AM RightsLink Printable License https://s100.copyright.com/App/PrintableLicenseFrame.jsp?publish...

ELSEVIER LICENSE TERMS AND CONDITIONS Sep 30, 2016

This Agreement between Jeremy M Kinder ("You") and Elsevier ("Elsevier") consists of your license details and the terms and conditions provided by Elsevier and Copyright Clearance Center.

License Number 3958920296725

License date Sep 30, 2016

Licensed Content Publisher Elsevier

Licensed Content Publication Cell

Licensed Content Title Cross-Generational Reproductive Fitness Enforced by Microchimeric Maternal Cells

Licensed Content Author Jeremy M. Kinder,Tony T. Jiang,James M. Ertelt,Lijun Xin,Beverly S. Strong,Aimen F. Shaaban,Sing Sing Way

Licensed Content Date 30 July 2015

Licensed Content Volume 162 Number

Licensed Content Issue 3 Number

Licensed Content Pages 11

Start Page 505

End Page 515

Type of Use reuse in a thesis/dissertation

Portion full article

Format both print and electronic

Are you the author of this Yes Elsevier article?

Will you be translating? No

Order reference number

Title of your Tolerogenic Properties of Maternal and Fetal Microchimeric Cells thesis/dissertation

Expected completion date Oct 2016

Estimated size (number of 250 pages)

Elsevier VAT number GB 494 6272 12

Requestor Location Jeremy M Kinder 2639 Observatory Ave #2

1 of 6 9/30/16, 2:44 PM RightsLink Printable License https://s100.copyright.com/App/PrintableLicenseFrame.jsp?publish...

CINCINNATI, OH 45208 United States Attn: Jeremy M Kinder

Total 0.00 USD

2 of 6 9/30/16, 2:44 PM RightsLink Printable License https://s100.copyright.com/App/PrintableLicenseFrame.jsp?publish...

materials beyond the scope of an unrevoked license, may constitute copyright infringement and publisher reserves the right to take any and all action to protect its copyright in the materials. 9. Warranties: Publisher makes no representations or warranties with respect to the licensed material. 10. Indemnity: You hereby indemnify and agree to hold harmless publisher and CCC, and their respective officers, directors, employees and agents, from and against any and all claims arising out of your use of the licensed material other than as specifically authorized pursuant to this license. 11. No Transfer of License: This license is personal to you and may not be sublicensed, assigned, or transferred by you to any other person without publisher's written permission. 12. No Amendment Except in Writing: This license may not be amended except in a writing signed by both parties (or, in the case of publisher, by CCC on publisher's behalf). 13. Objection to Contrary Terms: Publisher hereby objects to any terms contained in any purchase order, acknowledgment, check endorsement or other writing prepared by you, which terms are inconsistent with these terms and conditions or CCC's Billing and Payment terms and conditions. These terms and conditions, together with CCC's Billing and Payment terms and conditions (which are incorporated herein), comprise the entire agreement between you and publisher (and CCC) concerning this licensing transaction. In the event of any conflict between your obligations established by these terms and conditions and those established by CCC's Billing and Payment terms and conditions, these terms and conditions shall control. 14. Revocation: Elsevier or Copyright Clearance Center may deny the permissions described in this License at their sole discretion, for any reason or no reason, with a full refund payable to you. Notice of such denial will be made using the contact information provided by you. Failure to receive such notice will not alter or invalidate the denial. In no event will Elsevier or Copyright Clearance Center be responsible or liable for any costs, expenses or damage incurred by you as a result of a denial of your permission request, other than a refund of the amount(s) paid by you to Elsevier and/or Copyright Clearance Center for denied permissions. LIMITED LICENSE The following terms and conditions apply only to specific license types: 15. Translation: This permission is granted for non-exclusive world English rights only unless your license was granted for translation rights. If you licensed translation rights you may only translate this content into the languages you requested. A professional translator must perform all translations and reproduce the content word for word preserving the integrity of the article. 16. Posting licensed content on any Website: The following terms and conditions apply as follows: Licensing material from an Elsevier journal: All content posted to the web site must maintain the copyright information line on the bottom of each image; A hyper-text must be included to the Homepage of the journal from which you are licensing at http://www.sciencedirect.com/science/journal/xxxxx or the Elsevier homepage for books at http://www.elsevier.com; Central Storage: This license does not include permission for a scanned version of the material to be stored in a central repository such as that provided by Heron/XanEdu. Licensing material from an Elsevier book: A hyper-text link must be included to the Elsevier homepage at http://www.elsevier.com . All content posted to the web site must maintain the

3 of 6 9/30/16, 2:44 PM RightsLink Printable License https://s100.copyright.com/App/PrintableLicenseFrame.jsp?publish...

In all cases accepted manuscripts should:

Published journal article (JPA): A published journal article (PJA) is the deﬁnitive ﬁnal

4 of 6 9/30/16, 2:44 PM RightsLink Printable License https://s100.copyright.com/App/PrintableLicenseFrame.jsp?publish...

record of published research that appears or will appear in the journal and embodies all value-adding publishing activities including peer review co-ordination, copy-editing, formatting, (if relevant) pagination and online enrichment. Policies for sharing publishing journal articles differ for subscription and gold open access articles: Subscription Articles: If you are an author, please share a link to your article rather than the full-text. Millions of researchers have access to the formal publications on ScienceDirect, and so links will help your users to ﬁnd, access, cite, and use the best available version. Theses and dissertations which contain embedded PJAs as part of the formal submission can be posted publicly by the awarding institution with DOI links back to the formal publications on ScienceDirect. If you are afﬁliated with a library that subscribes to ScienceDirect you have additional private sharing rights for others' research accessed under that agreement. This includes use for classroom teaching and internal training at the institution (including use in course packs and courseware programs), and inclusion of the article for grant funding purposes. Gold Open Access Articles: May be shared according to the author-selected end-user license and should contain a CrossMark logo, the end user license, and a DOI link to the formal publication on ScienceDirect. Please refer to Elsevier's posting policy for further information. 18. For book authors the following clauses are applicable in addition to the above: Authors are permitted to place a brief summary of their work online only. You are not allowed to download and post the published electronic version of your chapter, nor may you scan the printed edition to create an electronic version. Posting to a repository: Authors are permitted to post a summary of their chapter only in their institution's repository. 19. Thesis/Dissertation: If your license is for use in a thesis/dissertation your thesis may be submitted to your institution in either print or electronic form. Should your thesis be published commercially, please reapply for permission. These requirements include permission for the Library and Archives of Canada to supply single copies, on demand, of the complete thesis and include permission for Proquest/UMI to supply single copies, on demand, of the complete thesis. Should your thesis be published commercially, please reapply for permission. Theses and dissertations which contain embedded PJAs as part of the formal submission can be posted publicly by the awarding institution with DOI links back to the formal publications on ScienceDirect.

5 of 6 9/30/16, 2:44 PM RightsLink Printable License https://s100.copyright.com/App/PrintableLicenseFrame.jsp?publish...

ensure their reuse complies with the terms and conditions determined by the rights holder. Additional Terms & Conditions applicable to each Creative Commons user license: CC BY: The CC-BY license allows users to copy, to create extracts, abstracts and new works from the Article, to alter and revise the Article and to make commercial use of the Article (including reuse and/or resale of the Article by commercial entities), provided the user gives appropriate credit (with a link to the formal publication through the relevant DOI), provides a link to the license, indicates if changes were made and the licensor is not represented as endorsing the use made of the work. The full details of the license are available at http://creativecommons.org/licenses/by/4.0. CC BY NC SA: The CC BY-NC-SA license allows users to copy, to create extracts, abstracts and new works from the Article, to alter and revise the Article, provided this is not done for commercial purposes, and that the user gives appropriate credit (with a link to the formal publication through the relevant DOI), provides a link to the license, indicates if changes were made and the licensor is not represented as endorsing the use made of the work. Further, any new works must be made available on the same conditions. The full details of the license are available at http://creativecommons.org/licenses/by-nc-sa/4.0. CC BY NC ND: The CC BY-NC-ND license allows users to copy and distribute the Article, provided this is not done for commercial purposes and further does not permit distribution of the Article if it is changed or edited in any way, and provided the user gives appropriate credit (with a link to the formal publication through the relevant DOI), provides a link to the license, and that the licensor is not represented as endorsing the use made of the work. The full details of the license are available at http://creativecommons.org/licenses/by-nc-nd/4.0. Any commercial reuse of Open Access articles published with a CC BY NC SA or CC BY NC ND license requires permission from Elsevier and will be subject to a fee. Commercial reuse includes:

- Associating advertising with the full text of the Article - Charging fees for document delivery or access - Article aggregation - Systematic distribution via e-mail lists or share buttons

Posting or linking by commercial companies for use by customers of those companies.

20. Other Conditions:

v1.8

Questions? [email protected] or +1-855-239-3415 (toll free in the US) or +1-978-646-2777.

Pregnancy-induced maternal regulatory T cells, bona fide memory or maintenance by antigenic reminder from fetal cell microchimerism?

Jeremy M Kinder, Tony T Jiang, Dayna R Clark, Vandana Chaturvedi, Lijun Xin, James M Ertelt, and Sing Sing Way* Division of Infectious Diseases; Cincinnati Children’s Hospital Medical Center; Cincinnati, OH USA

ong-term maintenance of immune for immunological memory has critical Lcomponents with defined speci- implications for developing more durable ficity, without antigen is the hallmark vaccines and other immune modulatory feature of immunological memory. therapies. However, there are fundamental differ- Emerging studies highlight an inter- ences in how memory CD8+ compared esting discordance in necessity for with CD4+ T cells are maintained. After antigen persistence in maintaining long- complete antigen elimination, CD8+ term retention of CD8+ compared with T cells can persist as a self-renewing CD4+ T cells with defined specificity.1-7 numerically stable cell population, and This is best illustrated by the dynamics therefore satisfy the most stringent of pathogen-specific CD8+ and CD4+ definition of “memory.” Comparatively, T cells after infection with viruses or CD4+ T cell maintenance is consider- other intracellular pathogens that do not ably less stable, often requiring low-level cause persistence. While both T cell sub- antigen persistence or antigenic remind- sets expand robustly during acute infec- ers. Recent studies show these basic tion, a numerically stable self-renewing memory features, classically ascribed to pool of pathogen-specific CD8+ T cells is effector CD8+ and CD4+ T cells, extend maintained indefinitely despite complete to immune suppressive Foxp3+ regula- antigen elimination. By contrast, CD4+ tory CD4+ T cells (Tregs). In particular, T cells responding to the same acute gestational expansion and postpartum infection undergo protracted, but stable retention of maternal Tregs with fetal contraction with an estimated half-life specificity may explain the protective of 15 to 40 d.8-10 This discordance may benefits of primary pregnancy on com- reflect the necessity for each T cell subset Keywords: T cells, immunological plications in subsequent pregnancy. in host defense. For acute infection with memory, regulatory T cells, pregnancy, Herein, the possibility of ongoing anti- viruses or other intra-cytoplasmic patho- microchimerism genic reminders from fetal cell micro- gens (e.g., influenza A, lymphocytic *Correspondence to: Sing Sing Way; chimerism in postpartum maintenance choriomeningitis virus, or Listeria mono- Email: [email protected] of maternal Tregs with fetal specificity cytogenes) where protection is conferred + Submitted: 01/17/2014 is considered. by CD8 T cells, these cells are chosen Revised: 01/27/2014 for selective retention. Comparatively for The mammalian immune system is pathogens that primarily cause persistent Accepted: 02/14/2014 endowed not only with efficient self, infection and reside within the phago- Published Online: 02/19/2014 non-self discrimination, but also the abil- cytic vacuole of infected cells thereby http://dx.doi.org/10.4161/chim.28241 ity to “remember” antigenic encounters. escaping detection or elimination by For immunologically foreign antigens, CD8+ T cells (e.g., Mycobacterium tuber- Addendum to: Rowe JH, Ertelt JM, Xin L, Way prior stimulation has the potential to culosis, Leishmania major, or Salmonella SS. Pregnancy imprints regulatory memory prime long-term retention of “memory” spp.), pathogen-specific CD4+ T cells that sustains anergy to fetal antigen. Nature immune cells with specificity to the play a more dominant protective role.11-13 2012; 490:102-6; PMID:23023128; http://dx.doi. inciting antigen. In turn, establishing Importantly however, while CD8+ T cell org/10.1038/nature11462 the molecular and cellular requirements mediated protection against secondary

www.landesbioscience.com Chimerism 1 infection is maintained well after anti- secondary pregnancy upon encounter to low-level antigen stimulation in the gen elimination, protection by retained with the same paternal-fetal antigen. later stages of persistent infection. Fetal memory CD4+ T cells requires low-level Considering the necessity for expanded cell microchimerism initiated during antigen persistence. Accordingly for maternal Tregs in maintaining fetal tol- pregnancy and sustained postpartum pathogens capable of establishing per- erance during pregnancy,19-24 these find- probably occurs ubiquitously, but this sistent infection, antigen elimination ings likely provide critical mechanistic phenomenon has become only recently that occurs naturally or with adjunctive insights for how primary pregnancy pro- widely appreciated with the use of molec- antimicrobials, accelerates contraction tects against complications stemming ular tools that allow these rare (~1 in 106) of pathogen-specific CD4+ T cells and from fractured fetal tolerance in subse- cells to be consistently identified.27-29 overrides the protective benefits of prior quent pregnancy.19,20,25,26 In turn, applied Accordingly, antigenic reminder from infection.14-16 Therefore, unlike CD8+ to the basic biology of CD4+ T cells, fetal cell microchimerism may be pivotal T cells, the long-term maintenance of these findings together establish Foxp3+ for sustaining memory among pregnancy- CD4+ T cell memory appear to require Tregs, like effector T cells, can persist as induced maternal Tregs. Moreover, if more frequent, if not constant, antigenic memory immune cells. maternal CD4+ Treg memory is sustained reminders. Given the discordance in necessity by fetal cell microchimerism, it would While the memory features of CD4+ for antigen persistence in sustaining be interesting to consider the necessity T cells has been best characterized for long-term retention of CD8+ compared for comparable antigenic reminders in IFN-γ producing Th1 cells, other CD4+ with CD4+ effector T cells with defined maintaining regulatory CD8+ T cells effector lineages (e.g., Th2 or Th17 cells) specificity, these findings also open up shown in other contexts.30-32 Along with appear to share a similar potential for long- exciting new questions regarding whether the long-term maintenance of maternal term retention.7,10 By redirecting tools for retained Tregs reflects bona fide memory cell microchimerism sustained by fetal tracking antigen-specific T cells, we and or maintenance in response to antigen Tregs in offspring,33 this emerging body others have recently shown these memory persistence. In the case of Tregs with of evidence highlight remarkably potent features classically described for effec- specificity for surrogate-self ovalbumin and long-lived immunological program- tor T cells also extend to immune sup- antigen within the skin, ongoing stimu- ming that occurs naturally with the bi- pressive regulatory CD4+ T cells (Tregs) lation is unlikely since naive ovalbumin directional transfer of cells and antigens identified by Foxp3 expression. Treg specific T cells failed to proliferate after between mother and fetus through in memory was first shown using transgenic adoptive transfer without induced anti- utero exposure. mice where the model antigen, ovalbu- gen expression.17 Similarly, Tregs retained Based on these findings, we propose min, could be inducibly expressed within after influenza A infection are unlikely important next steps are to more meticu- the skin.17 Primary dermal stimulation to reflect stimulation from residual viral lously dissect the physiological milieu of with this surrogate self-antigen primed antigen, since this pathogen is not known pregnancy and in utero development that expansion and retention of ovalbumin to cause persistent infection.18 However primes immunological tolerance and Treg specific Tregs that dampens the sever- in each of these models, the longer-term memory. Taking cues from effector CD4+ ity of localized autoimmune reactions durability of Tregs, with specificity to T cell memory,1-7,34 this will likely include when this antigen was re-expressed ~30 either self or pathogen, remain undefined interrelated contributions from naive cell d later. Likewise, a complementary study since the impacts of secondary antigen precursor frequency, primary expansion tracking Tregs after influenza A infec- challenge were reported at most ~35 d magnitude, antigen avidity, and response tion showed accelerated accumulation of after silencing primary antigen stimula- to cytokine growth factors, along with virus-specific Tregs after secondary, com- tion.17,18 In our studies tracking maternal increased frequency of antigenic remind- pared with primary infection, which may Tregs with surrogate fetal-2W1S speci- ers. Furthermore, given the potential for be important for limiting pathological ficity, enriched cells were maintained Treg conversion into inflammatory cyto- airway inflammation from over-exuber- through at least 100 d postpartum despite kine producing effector T cells with the ant effector CD8+ T cell activation.18 Our progressively diminishing cell numbers.19 same specificity,35,36 microchimeric fetal own studies investigating maternal Tregs In particular, the postpartum decay cells also have the dangerous potential with specificity to the immune-dominant kinetics of maternal Tregs with fetal spec- for sensitizing responses that may trigger b 37-39 I-A :2W1S55–68 peptide expressed as a ificity (estimated t1/2 of 25 d) show strik- autoimmunity. This is analogous to surrogate fetal antigen during allogeneic ing similarity with effector CD4+ T cells pathological responses to microchimeric pregnancy, showed Foxp3+ CD4+ T cells primed by acute infection. maternal cells in offspring with various with this specificity progressively expand On the other hand and in sharp con- diverse autoimmune disorders including throughout gestation.19 Interestingly trast to the tempo of antigen stimulation diabetes,40 biliary atresia,41 and derma- after delivery of the fetus and other gross that occurs after acute infection condi- tomyositis.42 Therefore, establishing the products of conception, maternal Tregs tions, retained maternal Tregs with fetal molecular signals that reinforce Treg dif- with fetal specificity were maintained at specificity are likely to have more frequent ferentiation stability are of equally high markedly enriched levels; and these cells antigenic encounters from fetal cells that importance and priority. re-expand with accelerated tempo during establish microchimerism, analogous

2 Chimerism Volume 5 Issue 1 8. Homann D, Teyton L, Oldstone MB. Differential 22. Somerset DA, Zheng Y, Kilby MD, Sansom DM, Nevertheless, applied to the devastat- regulation of antiviral T-cell immunity results in Drayson MT. Normal human pregnancy is associ- ing complications in human pregnancy stable CD8+ but declining CD4+ T-cell memory. ated with an elevation in the immune suppressive Nat Med 2001; 7:913-9; PMID:11479623; http:// CD25+ CD4+ regulatory T-cell subset. Immunology that stem from underlying defects in dx.doi.org/10.1038/90950 2004; 112:38-43; PMID:15096182; http://dx.doi. fetal tolerance (preeclampsia, prema- 9. Harrington LE, Janowski KM, Oliver JR, Zajac org/10.1111/j.1365-2567.2004.01869.x turity, miscarriage), basic investigation AJ, Weaver CT. Memory CD4 T cells emerge from 23. Sasaki Y, Darmochwal-Kolarz D, Suzuki D, Sakai M, Ito M, Shima T, Shiozaki A, Rolinski + effector T-cell progenitors. Nature 2008; 452:356- on the fundamental biology of CD4 60; PMID:18322463; http://dx.doi.org/10.1038/ J, Saito S. Proportion of peripheral blood and T cells and memory features for protec- nature06672 decidual CD4(+) CD25(bright) regulatory T cells tive regulatory subsets provides renewed 10. Pepper M, Linehan JL, Pagán AJ, Zell T, Dileepan T, in pre-eclampsia. Clin Exp Immunol 2007; Cleary PP, Jenkins MK. Different routes of bacterial 149:139-45; PMID:17459078; http://dx.doi. hope for new, more efficacious therapeu- infection induce long-lived TH1 memory cells and org/10.1111/j.1365-2249.2007.03397.x tic approaches. In turn, given the striking short-lived TH17 cells. Nat Immunol 2010; 11:83- 24. Toldi G, Svec P, Vásárhelyi B, Mészáros G, Rigó J, + 9; PMID:19935657; http://dx.doi.org/10.1038/ Tulassay T, Treszl A. Decreased number of FoxP3+ parallels between Treg and effector CD4 ni.1826 regulatory T cells in preeclampsia. Acta Obstet T cell memory, unraveling how maternal 11. Reiley WW, Shafiani S, Wittmer ST, Tucker-Heard Gynecol Scand 2008; 87:1229-33; PMID:19016357; Treg memory is sustained will likely also G, Moon JJ, Jenkins MK, Urdahl KB, Winslow http://dx.doi.org/10.1080/00016340802389470 GM, Woodland DL. Distinct functions of antigen- 25. Campbell DM, MacGillivray I, Carr-Hill R. provide critical insights for priming more specific CD4 T cells during murine Mycobacterium Pre-eclampsia in second pregnancy. Br J Obstet durable effector T cells with pathogen tuberculosis infection. Proc Natl Acad Sci U S A Gynaecol 1985; 92:131-40; PMID:3970893; http:// 2010; 107:19408-13; PMID:20962277; http:// dx.doi.org/10.1111/j.1471-0528.1985.tb01064.x specificity for augmenting host defense dx.doi.org/10.1073/pnas.1006298107 26. Trupin LS, Simon LP, Eskenazi B. Change against infection. 12. Winslow GM, Roberts AD, Blackman MA, in paternity: a risk factor for preeclamp- Woodland DL. Persistence and turnover of antigen- sia in multiparas. Epidemiology 1996; Disclosure of Potential Conflicts of Interest specific CD4 T cells during chronic tuberculosis 7:240-4; PMID:8728435; http://dx.doi. infection in the mouse. J Immunol 2003; 170:2046- org/10.1097/00001648-199605000-00004 No potential conflicts of interest were 52; PMID:12574375 27. Bianchi DW, Zickwolf GK, Weil GJ, Sylvester S, disclosed. 13. McSorley SJ, Cookson BT, Jenkins MK. DeMaria MA. Male fetal progenitor cells persist Characterization of CD4+ T cell responses during in maternal blood for as long as 27 years post- natural infection with Salmonella typhimurium. J partum. Proc Natl Acad Sci U S A 1996; 93:705- Acknowledgments Immunol 2000; 164:986-93; PMID:10623848 8; PMID:8570620; http://dx.doi.org/10.1073/ 14. Uzonna JE, Wei G, Yurkowski D, Bretscher P. pnas.93.2.705 This work was supported in part by the Immune elimination of Leishmania major in mice: 28. Gammill HS, Nelson JL. Naturally acquired NIAID through awards R01-AI087830, implications for immune memory, vaccination, and microchimerism. Int J Dev Biol 2010; 54:531- R01-AI100934, and R21-AI112186. reactivation disease. J Immunol 2001; 167:6967-74; 43; PMID:19924635; http://dx.doi.org/10.1387/ PMID:11739516 ijdb.082767hg S.S.W. holds an Investigator in the 15. Belkaid Y, Piccirillo CA, Mendez S, Shevach EM, 29. Khosrotehrani K, Johnson KL, Cha DH, Salomon Pathogenesis of Infectious Disease award Sacks DL. CD4+CD25+ regulatory T cells control RN, Bianchi DW. Transfer of fetal cells with Leishmania major persistence and immunity. Nature multilineage potential to maternal tissue. JAMA from the Burroughs Wellcome Fund. 2002; 420:502-7; PMID:12466842; http://dx.doi. 2004; 292:75-80; PMID:15238593; http://dx.doi. org/10.1038/nature01152 org/10.1001/jama.292.1.75 References 16. Nelson RW, McLachlan JB, Kurtz JR, Jenkins MK. 30. Robb RJ, Lineburg KE, Kuns RD, Wilson YA, 1. Seder RA, Ahmed R. Similarities and differ- CD4+ T cell persistence and function after infec- Raffelt NC, Olver SD, Varelias A, Alexander KA, ences in CD4+ and CD8+ effector and memory tion are maintained by low-level peptide:MHC Teal BE, Sparwasser T, et al. Identification and T cell generation. Nat Immunol 2003; 4:835-42; class II presentation. J Immunol 2013; 190:2828- expansion of highly suppressive CD8(+)FoxP3(+) PMID:12942084; http://dx.doi.org/10.1038/ni969 34; PMID:23382562; http://dx.doi.org/10.4049/ regulatory T cells after experimental allogeneic bone 2. Williams MA, Holmes BJ, Sun JC, Bevan jimmunol.1202183 marrow transplantation. Blood 2012; 119:5898- MJ. Developing and maintaining protective 17. Rosenblum MD, Gratz IK, Paw JS, Lee K, Marshak- 908; PMID:22538855; http://dx.doi.org/10.1182/ CD8+ memory T cells. Immunol Rev 2006; Rothstein A, Abbas AK. Response to self antigen blood-2011-12-396119 211:146-53; PMID:16824124; http://dx.doi. imprints regulatory memory in tissues. Nature 2011; 31. Fleissner D, Hansen W, Geffers R, Buer J, Westendorf org/10.1111/j.0105-2896.2006.00389.x 480:538-42; PMID:22121024 AM. Local induction of immunosuppressive CD8+ 3. Prlic M, Williams MA, Bevan MJ. Requirements 18. Brincks EL, Roberts AD, Cookenham T, Sell S, T cells in the gut-associated lymphoid tissues. PLoS for CD8 T-cell priming, memory generation and Kohlmeier JE, Blackman MA, Woodland DL. One 2010; 5:e15373; PMID:20975955; http:// maintenance. Curr Opin Immunol 2007; 19:315- Antigen-specific memory regulatory CD4+Foxp3+ dx.doi.org/10.1371/journal.pone.0015373 9; PMID:17433873; http://dx.doi.org/10.1016/j. T cells control memory responses to influenza 32. Uss E, Rowshani AT, Hooibrink B, Lardy NM, van coi.2007.04.010 virus infection. J Immunol 2013; 190:3438-46; Lier RA, ten Berge IJ. CD103 is a marker for alloan- 4. Harty JT, Badovinac VP. Shaping and reshap- PMID:23467933; http://dx.doi.org/10.4049/ tigen-induced regulatory CD8+ T cells. J Immunol ing CD8+ T-cell memory. Nat Rev Immunol jimmunol.1203140 2006; 177:2775-83; PMID:16920912 2008; 8:107-19; PMID:18219309; http://dx.doi. 19. Rowe JH, Ertelt JM, Xin L, Way SS. Pregnancy 33. Mold JE, Michaëlsson J, Burt TD, Muench MO, org/10.1038/nri2251 imprints regulatory memory that sustains Beckerman KP, Busch MP, Lee TH, Nixon DF, 5. Sprent J, Surh CD. Normal T cell homeosta- anergy to fetal antigen. Nature 2012; 490:102- McCune JM. Maternal alloantigens promote the sis: the conversion of naive cells into memory- 6; PMID:23023128; http://dx.doi.org/10.1038/ development of tolerogenic fetal regulatory T cells in phenotype cells. Nat Immunol 2011; 12:478-84; nature11462 utero. Science 2008; 322:1562-5; PMID:19056990; PMID:21739670; http://dx.doi.org/10.1038/ 20. Aluvihare VR, Kallikourdis M, Betz AG. Regulatory http://dx.doi.org/10.1126/science.1164511 ni.2018 T cells mediate maternal tolerance to the fetus. Nat 34. Gratz IK, Truong HA, Yang SH, Maurano MM, Lee 6. van Leeuwen EM, Sprent J, Surh CD. Generation Immunol 2004; 5:266-71; PMID:14758358; http:// K, Abbas AK, Rosenblum MD. Cutting Edge: mem- and maintenance of memory CD4(+) T Cells. Curr dx.doi.org/10.1038/ni1037 ory regulatory t cells require IL-7 and not IL-2 for Opin Immunol 2009; 21:167-72; PMID:19282163; 21. Rowe JH, Ertelt JM, Aguilera MN, Farrar MA, Way their maintenance in peripheral tissues. J Immunol http://dx.doi.org/10.1016/j.coi.2009.02.005 SS. Foxp3(+) regulatory T cell expansion required 2013; 190:4483-7; PMID:23543753; http://dx.doi. org/10.4049/jimmunol.1300212 7. Pepper M, Jenkins MK. Origins of CD4(+) effec- for sustaining pregnancy compromises host defense tor and central memory T cells. Nat Immunol against prenatal bacterial pathogens. Cell Host 35. Bailey-Bucktrout SL, Martinez-Llordella M, Zhou 2011; 12:467-71; PMID:21739668; http://dx.doi. Microbe 2011; 10:54-64; PMID:21767812; http:// X, Anthony B, Rosenthal W, Luche H, Fehling org/10.1038/ni.2038 dx.doi.org/10.1016/j.chom.2011.06.005 HJ, Bluestone JA. Self-antigen-driven activation induces instability of regulatory T cells during an inflammatory autoimmune response. Immunity 2013; 39:949-62; PMID:24238343; http://dx.doi. org/10.1016/j.immuni.2013.10.016

www.landesbioscience.com Chimerism 3 36. Komatsu N, Okamoto K, Sawa S, Nakashima T, 39. Mosca M, Curcio M, Lapi S, Valentini G, D’Angelo 41. Suskind DL, Rosenthal P, Heyman MB, Kong Oh-Hora M, Kodama T, Tanaka S, Bluestone JA, S, Rizzo G, Bombardieri S. Correlations of Y chro- D, Magrane G, Baxter-Lowe LA, Muench Takayanagi H. Pathogenic conversion of Foxp3(+) mosome microchimerism with disease activity in MO. Maternal microchimerism in the livers of T cells into TH17 cells in autoimmune arthritis. patients with SLE: analysis of preliminary data. patients with biliary atresia. BMC Gastroenterol Nat Med 2014; 20:62-8; PMID:24362934; http:// Ann Rheum Dis 2003; 62:651-4; PMID:12810428; 2004; 4:14; PMID:15285784; http://dx.doi. dx.doi.org/10.1038/nm.3432 http://dx.doi.org/10.1136/ard.62.7.651 org/10.1186/1471-230X-4-14 37. Adams KM, Nelson JL. Microchimerism: an inves- 40. Nelson JL, Gillespie KM, Lambert NC, Stevens 42. Reed AM, Picornell YJ, Harwood A, Kredich DW. tigative frontier in autoimmunity and transplanta- AM, Loubiere LS, Rutledge JC, Leisenring WM, Chimerism in children with juvenile dermatomyo- tion. JAMA 2004; 291:1127-31; PMID:14996783; Erickson TD, Yan Z, Mullarkey ME, et al. Maternal sitis. Lancet 2000; 356:2156-7; PMID:11191546; http://dx.doi.org/10.1001/jama.291.9.1127 microchimerism in peripheral blood in type 1 http://dx.doi.org/10.1016/S0140-6736(00)03500-5 38. Stevens AM. Microchimeric cells in systemic lupus diabetes and pancreatic islet beta cell microchi- erythematosus: targets or innocent bystanders? merism. Proc Natl Acad Sci U S A 2007; 104:1637- Lupus 2006; 15:820-6; PMID:17153857; http:// 42; PMID:17244711; http://dx.doi.org/10.1073/ dx.doi.org/10.1177/0961203306070068 pnas.0606169104

4 Chimerism Volume 5 Issue 1 Chimerism

ISSN: 1938-1956 (Print) 1938-1964 (Online) Journal homepage: http://www.tandfonline.com/loi/kchm20

Tolerance to noninherited maternal antigens, reproductive microchimerism and regulatory T cell memory: 60 years after ‘Evidence for actively acquired tolerance to Rh antigens’

Jeremy M. Kinder, Tony T. Jiang, James M. Ertelt, Lijun Xin, Beverly S. Strong, Aimen F. Shaaban & Sing Sing Way

To cite this article: Jeremy M. Kinder, Tony T. Jiang, James M. Ertelt, Lijun Xin, Beverly S. Strong, Aimen F. Shaaban & Sing Sing Way (2015) Tolerance to noninherited maternal antigens, reproductive microchimerism and regulatory T cell memory: 60 years after ‘Evidence for actively acquired tolerance to Rh antigens’, Chimerism, 6:1-2, 8-20, DOI: 10.1080/19381956.2015.1107253

To link to this article: http://dx.doi.org/10.1080/19381956.2015.1107253

© 2016 The Author(s). Published with Accepted author version posted online: 30 license by Taylor & Francis.© Jeremy M. Oct 2015. Kinder, Tony T. Jiang, James M. Ertelt, Lijun Published online: 30 Oct 2015. Xin, Beverly S. Strong, Aimen F. Shaaban, and Sing Sing Way. Submit your article to this journal Article views: 421

View related articles View Crossmark data

Citing articles: 1 View citing articles

Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=kchm20

Download by: [University of Cincinnati Libraries] Date: 14 October 2016, At: 05:26 CHIMERISM 2016, VOL. 6, NOS. 1–2, 8–20 http://dx.doi.org/10.1080/19381956.2015.1107253

REVIEW Tolerance to noninherited maternal antigens, reproductive microchimerism and regulatory T cell memory: 60 years after ‘Evidence for actively acquired tolerance to Rh antigens’

Jeremy M. Kindera, Tony T. Jianga, James M. Ertelta, Lijun Xina, Beverly S. Strongb, Aimen F. Shaabanb, and Sing Sing Waya aDivision of Infectious Diseases and Perinatal Institute, Cincinnati Children’s Hospital, Cincinnati, OH, USA; bCenter for Fetal Cellular and Molecular Therapy, Cincinnati Children’s Hospital, Cincinnati, OH, USA

ABSTRACT ARTICLE HISTORY Compulsory exposure to genetically foreign maternal tissue imprints in offspring sustained Received 6 August 2015 tolerance to noninherited maternal antigens (NIMA). Immunological tolerance to NIMA was first Revised 25 September 2015 described by Dr. Ray D. Owen for women genetically negative for erythrocyte rhesus (Rh) antigen Accepted 7 October 2015 with reduced sensitization from developmental Rh exposure by their mothers. Extending this analysis to HLA haplotypes has uncovered the exciting potential for therapeutically exploiting NIMA-specific tolerance naturally engrained in mammalian reproduction for improved clinical outcomes after allogeneic transplantation. Herein, we summarize emerging scientific concepts stemming from tolerance to NIMA that includes postnatal maintenance of microchimeric maternal origin cells in offspring, expanded accumulation of immune suppressive regulatory T cells with NIMA-specificity, along with teleological benefits and immunological consequences of NIMA- specific tolerance conserved across mammalian species.

Introduction, pioneering observations on bull derived from a fraternal twin litter failed to trans- immunological tolerance by Dr. Ray Owen mit his phenotypic blood group antigens in up to 20 of Each individual among outbred populations is immu- his sired next generation progeny. Reflecting on ana- nologically unique – defined by inherited maternal and tomical vascular anastomoses between bovine twin paternal genes that encode distinctive MHC haplotype embryos,2 these observations were pieced together to alleles along with other minor alloantigens. This estab- postulate co-existence of shared blood cells between lished definition of immunological identity, with ensu- genetically non-identical twin cattle throughout adult ing implications for tolerance based on binary self life.1 More importantly, Owen recognized the revolu- versus non-self antigen distinction is based on pioneer- tionary cross-disciplinary implications of these findings ing observations by Dr. Ray D. Owen comparing anti- for immunology and genetics, articulating arguably the gen diversity among fraternal twin cattle. In a seminal first definitive example of persistent immunological tol- paper published in 1945, only 5 paragraphs were needed erance to genetically foreign antigens. to articulate and re-conceptualize foundational concepts Although this seminal characterization of tolerance regardingimmunologicalidentityandtolerance.1 between fraternal twin cattle establishing the existence Dr. Owen’s roots were in farming dairy cattle. In this of acquired immunological tolerance is most widely rec- context, he made the intriguing observation that a ognized, other related contributions have had equally majority of fraternal twin cattle had compatible blood sustained impacts shaping research on immunological types despite a diversity of at least 40 distinct genetically responsiveness to developmentally pertinent genetically controlled antigens known for this species.1 This unex- foreign antigens. This brief review written to commem- pected finding persisted even in cases of superfecunda- orate Dr. Owen’s100th birthday contains a snapshot of tion, involving twins in the same pregnancy sired by past and ongoing work on immune tolerance to nonin- genetically distinct fathers. Owen also noted that one herited maternal antigen directly stemming from

CONTACT Sing Sing Way [email protected] © 2016 Jeremy M. Kinder, Tony T. Jiang, James M. Ertelt, Lijun Xin, Beverly S. Strong, Aimen F. Shaaban, and Sing Sing Way. Published with license by Taylor & Francis. This is an Open Access article distributed under the terms of the Creative Commons Attribution-Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0/), which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited. The moral rights of the named author(s) have been asserted. CHIMERISM 9

‘Evidence for actively acquired tolerance to Rh antigens in these clinical contexts has been further evaluated. reported by Owens and colleagues 60 years ago.’3 Functional consequences of NIMA-specifictolerance were demonstrated in individuals who received multiple blood transfusions and normally developed anti- Human immunological tolerance to bodies against almost all HLA alloantigens. noninherited maternal antigens Interestingly however, a majority of transfusion depen- Despite the pervasive immunological implications dent individuals broadly exposed to foreign HLA stemming from tolerance to discordant cells in the alleles were found to selectively lack antibodies with somewhat obscure setting of fraternal twinning, far specificity to NIMA, compared with noninherited more common is compulsory exposure of each indi- paternal antigen (NIPA) HLA haplotypes.5 Along with vidual during in utero fetal development to genetically critical implications these data have on organ donor foreign maternal cells and tissues that express nonin- selection prior to transplantation, NIMA-specific toler- herited maternal antigens (NIMA). Here, Owen was ance was also recognized to protect against allograft the first to recognize that physiological exposure to rejection following transplantation. In a landmark ret- discordant maternal antigens in this developmental rospective analysis of 205 kidney transplant recipients context can confer sustained immunological tolerance from HLA mis-matched sibling donors, long-term allo- to NIMA in offspring.3 Investigating the heterogeneity graft survival was markedly improved if mismatched of sensitization to erythrocyte rhesus (Rh) antigen for NIMA compared to NIPA HLA haplotypes.6 In among Rh-negative women during pregnancies with fact, 5 and 10 year survival of NIMA-mismatched kid- Rh-positive male partners, Owen postulated that early neys was nearly identical to survival rates of HLA- developmental stimulation by Rh antigen among identical grafts. Together, these classical studies not women born to Rh-positive mothers might confer per- only reaffirm Owen’s initial observation of persistent sistent tolerance to Rh sensitization. In other words, postnatal NIMA-specific tolerance, but also highlight exposure to noninherited antigens expressed by the exciting translational opportunities for exploiting toler- fi maternal grandmother may have bene cial impacts in ance to NIMA naturally imprinted during early devel- women on the outcome of next-generation pregnan- opment for improved outcomes after transplantation. cies. This hypothesis was addressed by comparing the More variable degrees of protection have been maternal Rh status of women categorized as either described after haematopoietic stem cell transplanta- Rh-tolerant or Rh-intolerant following repeated stim- tion exploiting NIMA-specific tolerance to protect ulation by concepti bearing this genetically foreign against graft-vs.-host disease (GVHD) in HLA-discor- paternal antigen. Remarkably, a significant majority dant donor-recipient pairings.7,8 Among HLA-haploi- (78%; 32 of 41) of Rh-tolerant women were shown to dentical sibling-to-sibling donor-recipient pairs, van have Rh-positive mothers, whereas this maternal Rh Rood and colleagues reported a 1.fold9- reduced risk skewing was eliminated among Rh-intolerant women of acute GVHD in NIMA-mismatched compared (48%; 27 of 56 born to Rh-positive mothers).3 Thus, a with NIPA-mismatched bone marrow transplants.9 critical association between early developmental expo- Ichinohe and colleagues described 9.fold9- reduced sure to Rh antigen for reproductive age women from rates of severe grade III/IV acute GVHD after haema- their own mothers, and protection against Rh sensiti- topoietic stem cell transplantation between NIMA- zation during next generation pregnancies was mismatched family members.10 In addition, long-term recognized. Interestingly however, the incidence of follow up of the same cohort by Kanda and colleagues erythroblastosis fetalis or newborn hemolytic disease revealed immunosuppressive therapy could be suc- remained similar among Rh-tolerant and Rh-intolerant cessfully withdrawn for NIMA-mismatched transplant mothers suggesting NIMA-specific tolerance to this recipients with mild GVHD.11 single alloantigen alone was not sufficient to confer Despite these remarkable protective benefits from survival benefits to next-generation offspring.3,4 initial retrospective analysis, individual case reports of As the need for immunological tolerance to geneti- prospectively selected NIMA-mismatched donor- cally foreign antigens became increasingly recognized recipient pairs showed less promising results. In in transfusion medicine and transplantation over the one case series of 3 patients selected to receive next 30 years, applicability of NIMA-specific tolerance NIMA-mismatched donor haematopoietic stem cells, 10 J. M. KINDER ET AL. acute GVHD occurred in 2 recipients and graft rejec- definitively illustrated by improved outcomes among tion occurred in the third.12 For these recipients of allo- breast-fed individuals receiving maternal donor kid- geneic donor stem cells, the hypothesis that more ney allografts.16 In retrospective inquires on breast- intense immune suppressive conditioning therapies feeding, functional survival of the maternal allograft may override protection conferred by NIMA-specific was significantly increased among breast-fed com- tolerance is consistent with comparable survival rates pared with non-breast fed renal transplant recipient of NIMA-mismatched solid organ allografts in retro- offspring. These protective benefits of breastfeeding spective cohorts when increased potency immune sup- were specific to maternal allograft tissue bearing pression with cyclosporine is used to avert rejection.6,13 NIMA, since no differences in paternal donor allograft However, the lack of significant protective benefits survival were identified.16 These remarkable benefits in these isolated contexts does not negate the need in human transplantation strongly implicate potent for further investigating how naturally engrained antigen-specific immune modulatory properties of NIMA-specific tolerance can be therapeutically soluble maternal-HLA in breast milk.32,33 In turn, exploited in transplantation and other clinical areas direct associations between breastfeeding, increased that require more stringent immunological regulation postnatal persistence of microchimeric maternal cells (e.g. allergy, autoimmunity, maternal-fetal tolerance). and NIMA-specific tolerance shown in complemen- Another more intriguing explanation for incomplete tary animal studies suggest breast milk may contain a and discordant phenotypes of NIMA-specifictolerance critical source of maternal cells that establish micro- may be related to variations in postnatal exposure to chimerism in offspring.14,15,17,29,33 Alternatively, maternal cells bearing NIMA through breastfeeding, another interpretation of increased maternal cell and ensuing differences in levels of maternal cell microchimerism in breast fed individuals is that microchimerism.14-17 Postnatal persistence of geneti- NIMA-specific tolerance prevents rejection of genetically foreign chimeric maternal cells in offspring was cally foreign cells transferred through breast milk. originally described in infants with severe combined Additional clues on the developmental ontogeny fi 18-23 immune de ciency. In a study of 121 infants with and adaptive immune components responsible for defective T and B lymphocyte development, 40% had human NIMA-specific tolerance were unveiled in pio- engrafted maternal T cells and a similar proportion neering analysis of latent anti-maternal immunity in T developed clinically apparent GVHD caused by anti- cells of fetal and adult offspring.34 Recognizing the fetal allo-immunity.24 Similarly in cases of maternal need for active suppression among maturing fetal T malignancy during pregnancy, transplacental metasta- cells during in utero development to avert potentially ses of immune evasive tumor cells has been described harmful anti-maternal immunity, Mold, McCune and in melanoma, lymphoblastic leukemia, and lung ade- colleagues showed tolerogenic fetal immune suppres- nocarcinoma.25-27 With more sensitive techniques sive regulatory T cells develop from exposure to geneti- allowing detection of potentially rarer maternal cells or cally foreign maternal alloantigens.35 This dedicated their DNA (e.g., fluorescence in situ hybridization, fetal CD4+ T cell subset identified by expression of the quantitative polymerase chain reaction), microchimer- Foxp3 transcriptional regulator or high-affinity IL-2 ism of maternal origin is increasingly recognized to cytokine receptor, CD25, was shown to selectively sup- occur near ubiquitously among offspring.28-30 Maternal press anti-maternal responses. In elegant co-culture cell-specific DNA can be detected in up to 30% of cord assays between purified fetal T cells and antigen pre- blood specimens at a median concentration of 0.3% senting cells from the biological mother or non-related among fetal cells,31 whereas DNA encoding maternal adult donors, selective suppression of anti-maternal MHC haplotype alleles are found in 22% to 55% of T cell proliferation by fetal CD25+ T cells was demon- healthy adult individuals after analysis of less than strated.35 Expansion of CD25+ or Foxp3+ regulatory 2 mg of peripheral blood DNA.28,29 Thus, vertical T cells that suppress anti-maternal immunity also par- transfer and engraftment of maternal cells in offspring alleled microchimeric maternal cells in fetal lymph is likely an unavoidable by-product of in utero devel- node tissue. Thus for human infants with numerically opment through a porous placental interface. replete T cells at the time of birth, NIMA-specific toler- The additional immune modulatory properties of ance is likely essential for restraining harmful anti- breastfeeding on NIMA-specific tolerance are most maternal immunity during in utero and early postnatal CHIMERISM 11 development when exposure to foreign maternal anti- faithfully reproduced and further dissected in animals. gens is unavoidable.34,35 With this reasoning, postnatal Using an elegant F1 backcross breeding strategy to persistence of NIMA-specific tolerance through adult- generate genetically identical mice discordantly hood can be viewed as a developmental remnant of exposed to defined MHC haplotype alleles as surro- immune suppressive pathways essential for in utero gate NIMA or NIPA, Burlingham and colleagues fetal-maternal co-habitation. Together, this emerging showed remarkable protective benefits against rejec- body of human retrospective and experimental data tion of fully allogeneic NIMA-compatible donor heart sparked by Owen’s initial characterization of maternal grafts.14,37 In parallel with the aforementioned Rh ancestral phenotype highlight profound immune improved survival of human maternal kidney allog- modulatory properties engrained through developmen- rafts in breast-fed compared with non-breast fed tal NIMA exposure. However, this intriguing immuno- recipient offspring,16 complementary cross fostering logical association between NIMA-specific tolerance, nursing studies in mice show maternal antigen expo- maternal cell microchimerism, and expanded fetal reg- sure both in utero and through oral breast milk inges- ulatory T cells that actively suppress anti-maternal tion are simultaneously essential for improved immunity also raises provocative new questions with survival of NIMA-mismatched cardiac allografts.14 regards to why this biological phenomenon is pro- Similarly in animal models of haematopoietic stem grammed to persist through adulthood. cell transplantation, GVHD was attenuated in irradiated recipient mice reconstituted with allogeneic NIMA-mismatched donor splenocytes.38 Postnatal Animal models of immune tolerance with early exposure to NIMA through breastfeeding also enhan- developmental antigen exposure ces protection against GVHD for immune progenitor Nearly 10 y after Owen’s seminal description of cells transferred into NIMA-mismatched irradiated immunological tolerance between genetically discor- recipient mice,15 and these beneficial impacts are dant fraternal twin cattle, Dr. Peter Medawar’s analy- directly linked with postnatal persistence of microchi- 17

sis of skin graft survival provided pivotally important meric maternal cells in offspring. In turn, analysis of experimental evidence supporting the existence of individual mice after solid organ or haematopoietic actively acquired tolerance to genetically foreign tis- stem cell transplantation have identified direct associ- sues.36 Employing unique strains of highly inbred ations between NIMA-tolerant phenotypes, levels of genetically identical mice, Medawar’s classical experi- maternal cell microchimerism and expanded accumu- ments showed in utero exposure to cells from discor- lation of CD25+, Foxp3+, or transforming growth fac- dant mouse strains can confer tolerance to skin tor-b producing regulatory CD4+ T cells.15,17,37,39 grafts that persists through adulthood. Tolerance Thus, animal models of NIMA-specific tolerance ame- to skin grafting in chickens was similarly observed nable to experimental investigation have been instru- after embryonic cell transfer between unique inbred mental in verifying, as well as further establishing the strains identified by distinctive feather coloration.36 immunological cellular and molecular mechanisms Although the use of genetically homozygous inbred responsible for NIMA-specific tolerance.39 animals in these studies precludes analysis of NIMA- The study of in utero transplantation of genetically specific tolerance, these results nonetheless clearly foreign cells also exploits the tolerogenic properties established antigen-specific tolerogenic properties unique to fetal development and provides important stemming from in utero and early developmental anti- mechanistic clues on NIMA-specific tolerance.40-42 gen exposure. Considering immunological tolerance The theoretical advantage of in utero transplantation to developmentally irrelevant alloantigens can be is that therapeutic introduction of genetically foreign primed by in utero stimulation, physiological exposure cells into the fetal recipient prior to maturation of of the fetus to semi-allogeneic maternal tissues would adaptive immune components can induce long-term be expected to confer similar, if not more profound, donor-specific tolerance without the need for toxic immunological tolerance to NIMA. myeloablative conditioning. Animal models of in There is now definitive evidence that NIMA-spe- utero haematopoietic cell transplantation highlight cific tolerance initially described in retrospective anal- the critical importance of a minimum threshold of ysis of human transplantation outcomes can be antigen exposure necessary to establish and maintain 12 J. M. KINDER ET AL. allo-specific tolerance.40,43-48 Therefore, tolerance to peptide, between NIMA exposed and naive control NIMA that parallels persistence of maternal origin mice suggest thymic deletion of NIMA responsive cells microchimeric cells also likely hinges on a minimum play less important roles in immunological tolerance level of exposure to microchimeric maternal cells.17,44 to NIMA. (ref. 51 and unpublished data) Thus, early Further study is needed however, to determine how developmental exposure to NIMA primes in offspring the level of maternal microchimerism may dictate an expanded pool of peripherally induced immune alternate outcomes of autoimmunity or NIMA-spe- suppressive regulatory T cells with NIMA-specificity. cific tolerance. Additionally, given that in utero trans- Sustained expansion of NIMA-specific regulatory T plantation of genetically foreign cells to the fetus does cells in offspring also paralleled postnatal retention of not occur in isolation from the immunologically com- microchimeric maternal cells. OVA encoding DNA petent mother, maternal allo-sensitization can result specific to genetically discordant maternal cells was from the introduction of discordant third-party identified in vital organs (e.g. liver, heart) of NIMA alloantigens into the fetus.49,50 Thus, discordance in exposed mice, at levels corresponding to 1 maternal protective benefits of NIMA-specific tolerance after cell in 105 to 106 offspring cells in agreement with transplantation may also reflect transfer of maternal other studies using complementary tools for estimating adaptive immune components that have undergone levels of maternal microchimerism in mice and non- sensitization to fetal-specific antigen.6,9-13 human primates.17,47,48,51,55,56 One interpretation of To more definitively identify the specificity of these data is that postnatal persistence of NIMA-specific immune suppressive regulatory T cells that expand tolerance is a developmental remnant that protects with developmental NIMA exposure, we developed a genetically foreign microchimeric maternal cells from breeding strategy that uniquely transforms defined rejection in offspring. Alternatively, retained microchi- model antigens into surrogate NIMA.51 Specifically, meric maternal cells may have themselves adapted tol- female mice heterozygous for a defined transgene that erogenic properties required for driving expanded encodes cell surface expression of a recombinant pro- accumulation of NIMA-specificregulatoryTcellsand

tein containing ovalbumin plus the 2W1S variant of therefore promote their own survival. mouse I-Ea peptide were used for mating with non- To definitively investigate the cause and effect rela- transgenic males.52,53 This approach that simulta- tionship between the interrelated phenomena of b fi neously transforms the MHC class II I-A :2W1S55-68 expanded accumulation of NIMA-speci c regulatory peptide plus ovalbumin into surrogate NIMA in half T cells and microchimeric maternal cells simulta- the offspring, combined with tetramer staining and neously retained in offspring, the impacts of selectively bead enrichment tools for precisely identifying rare I- depleting microchimeric maternal cells based on co- b fi + A :2W1S-speci c CD4 T cells, provided a unique expression of ovalbumin protein with 2W1S55-58 pep- opportunity to investigate the differentiation of tide in NIMA exposed mice was evaluated. Remark- endogenous NIMA-specific cells.54 We found CD4+ T ably, expanded accumulation of NIMA-specific cells with surrogate-NIMA specificity in NIMA- regulatory T cells declined to background levels found exposed adult mice became highly enriched (~50%) in NIPA or naive control mice within the first 2 weeks for Foxp3 expression compared with CD4+ T cells of after depleting microchimeric 2W1S-OVA+ maternal the same specificity in NIPA exposed or control mice cells with anti-ovalbumin antibodies.51 These results without developmental 2W1S exposure.51 In agree- are consistent with the hypothesis that microchimeric ment with aforementioned human and mouse studies cells provide an essential source of cognate maternal highlighting the importance of postnatal stimulation antigen required for maintaining expanded accumula- by genetically foreign maternal cells through breast- tion of NIMA-specific regulatory T cells. In this feeding,15-17 expanded accumulation of NIMA-specific regard, numerical retention of memory regulatory T Foxp3+ regulatory T cells declined sharply in cross cells with NIMA-specificity appear to share with effec- fostered mice exposed to maternal tissues bearing tor CD4+ T cells of foreign microbial specificity the NIMA-2W1S during in utero development or through necessity for frequent, if not constant, cognate antigen breastfeeding in isolation. On the other hand, compa- exposure reminders. Based on these findings, it may rable absolute numbers of I-Ab:2W1S-specific CD4+ T be worthwhile to investigate if memory regulatory cells, and their similar avidity for cognate I-Ab:2W1S T cells described in other contexts (e.g., transient CHIMERISM 13 expression in the skin or after acute infection with more universal explanation for evolutionarily con- viral pathogens).57-60 represent bona fide memory like served NIMA-specific tolerance. This notion is sup- activated CD8+ T cells, or alternatively share a ported by our recent demonstration of expanded requirement for low-level exposure to cognate anti- NIMA-specific regulatory T cell accumulation in gen.61-65 In the broader scientific context, these results female compared with male NIMA-2W1S littermate illustrate how dissecting the fundamental immunology offspring, and correspondingly enriched levels of responsible for NIMA-specific tolerance can continue microchimeric maternal cell DNA retained in female to reveal hidden immunological secrets engrained in gender specific reproductive tissue.51 mammalian reproduction. To further investigate this hypothesis, susceptibility to complications during allogeneic pregnancy stemming from disruptions in fetal tolerance were Teleological benefits and immunological evaluated in genetically identical female mice develop- consequences of NIMA-specific tolerance mentally exposed to discordant MHC haplotypes as Despite the primary use of animal models to verify surrogate NIMA. Remarkably, this analysis showed the existence and further establish immunological NIMA-specific tolerance confers profound resiliency mechanisms responsible for NIMA-specific tolerance, against fetal wastage normally triggered by infection comparison of NIMA-specific tolerance across mam- with the prenatal bacterial pathogen Listeria monocy- malian species can also provide critical insights on the togenes or partial transient depletion of bulk maternal evolutionary ontogeny of this highly engrained immu- regulatory T cells.51,70-72 These protective benefits nological phenomena. For human infants with numer- occurred in an antigen-specific fashion requiring com- ically replete adaptive immune components at the monality between NIMA and paternal-fetal expressed time of birth, tolerance to genetically foreign maternal antigens since susceptibility to fetal wastage cells and tissues begins in utero with suppressed rebounded when third-party male mice bearing irrele- activation of maturing immune cells with NIMA- vant MHC haplotype alleles were used for mating fi 34,35 speci city. However, this reasoning does not with NIMA-exposed female mice. Thus, expanded explain why tolerance imprinted by exposure to for- accumulation of regulatory T cells with fetal specific- eign antigens during early development is widely con- ity, primed by either developmental NIMA stimula- served across mammalian species (e.g. non-human tion or prior pregnancy, efficiently overrides primates, ruminants, rodents) with sharply delayed susceptibility to invasive infection with prenatal adaptive immune cell maturation relative to parturi- pathogens like Listeria monocytogenes conferred by tion.34,66 For example, prolonged survival of NIMA- increased non-specific immune suppression from matched allografts and expanded NIMA-specific regu- accumulation of bulk maternal regulatory T cells.71,73 latory T cells in human adults is consistently repro- Given the pivotal importance of decidual infiltration duced in adult mice despite the absence of peripheral by activated fetal-specific CD8+ T cells in the T cells at the time of birth for this species.14,34,37 The immune-pathogenesis of fetal wastage that occurs preservation of NIMA-specific tolerance in mamma- with prenatal infection,70 dissecting the anatomical lian species born without functional adaptive immune and molecular details whereby fetal-specific regulatory components suggests the existence of more universal CD4+ T cells efficiently reinforce fetal tolerance are biological benefits driving conserved tolerance to critically important areas for future investigation with NIMA in placental mammals. direct translational implications for improving human An important clue is postnatal persistence of pregnancy outcomes. NIMA-specific tolerance through adulthood that is Cross-generational protection against fetal wastage actively maintained by maternal cells that established in animal pregnancy models are also in agreement microchimerism in offspring.51 In turn, given the with the ‘grandmother effect’ reported by Gammill, necessity for expanded tolerance that encompasses Nelson and colleagues where reduced rates of preg- immunologically foreign paternal-fetal antigens in nancy complications stemming from disruptions in successful pregnancy shared by all eutherian placental fetal tolerance (e.g., preeclampsia, recurrent miscar- mammals,67-69 we reasoned reinforced fetal tolerance riage) in women parallel increased levels of microchi- during next-generation pregnancies may represent a meric cells retained from their mothers.74-76 Given the 14 J. M. KINDER ET AL. necessity for overlap between NIMA and fetal- NIMA-specific tolerance underscores the remarkably expressed antigen during next-generation pregnancies engrained drive for genetic fitness in each female indi- for reinforced fetal tolerance in mice,51 important vidual. In addition to transmitting half of homologous next-steps are to investigate if similar overlap aug- chromosomes through Mendelian inheritance, verti- ments resiliency against complications in human cally transferred maternal cells that establish micro- pregnancy. In this regard, while developmental expo- chimerism in offspring selectively enforce fetal sure to the single minor Rh alloantigen initially tolerance during next generation pregnancies that described by Owen was insufficient to prevent hemo- promotes conservation of NIMA.51 However in lytic disease of the newborn,3,4 these data suggest nature, NIMA-specific tolerance among dominant broader non-inherited antigenic overlap between female individuals is likely counterbalanced by patho- maternal grandmother and offspring that encom- gen-mediated selection for MHC diversity among passes MHC haplotype alleles may more efficiently homologous chromosomes within individuals, and confer cross-generational reproductive benefits. between individuals across outbred populations.30,84,85 These findings also highlight striking commonality Together, these findings highlight the need for between reproductive benefits conferred by expanded more extended cross-generational analysis to resolve regulatory T cells with NIMA-specificity retained in the ongoing controversy regarding how MHC hap- females from developmental exposure to genetically lotype similarity impacts mate selection and preg- foreign maternal antigens, and regulatory T cells with nancy outcomes.86-88 Here, the efficiency whereby fetal specificity retained in mothers after prior preg- NIMA-specific tolerance retained in adult female nancy.73 In each case, memory regulatory T cells in mice protects against disruptions in fetal tolerance reproductive age females re-accumulate with sharply during next generation pregnancies strongly suggests accelerated tempo in response to cognate fetal antigen enhanced protection against rejecting the fetal allo- stimulation that protects against disruptions in fetal graft in addition to other selective benefits promoting tolerance. The actively retained enriched pool of pathogen resistance drives preservation of this immu- fi memory maternal regulatory T cells with speci city to nological phenomena at least in some placental pre-existing fetal antigen provides an intriguing scien- mammalian species. tific framework that explains human partner-specific On the other hand, cross-generational reproductive protective benefits of prior pregnancy against compli- advantages that preserve postnatal retention of micro- cations in subsequent pregnancies.77,78 Given the chimeric maternal cells may also perpetuate auto- necessity for cognate antigen reminders in the form inflammatory or autoimmune diseases in off- of microchimeric maternal cells in maintaining spring.89,90 These potentially harmful consequences of expanded accumulation of NIMA-specific regulatory retained microchimeric maternal cells have been best T cells,51 it is tantalizing to hypothesize that the pro- characterized in individuals with scleroderma where tective subset of mother’s little helpers in the form of increased levels of maternal microchimerism have maternal regulatory T cells with pre-existing fetal been identified.28,29 Enriched microchimeric maternal specificity are similarly maintained by fetal cells that cells are found in the peripheral blood and pancreatic establish persistent microchimerism in mothers after tissue of individuals with type 1 diabetes.91 For indi- parturition.79-83 In other words, microchimeric mater- viduals with rheumatoid arthritis, remarkable links nal cells that promote cross-generational reproductive between noninherited maternal HLA-DR alleles asso- fitness, and fetal cells that establish microchimerism ciated with disease susceptibility and resistance have in mothers after pregnancy can each be more accu- each been described.92-94 Similarly, maternal cell rately viewed as mother’s little genetic helpers. Thus, microchimerism has also been recently shown to be establishing functional similarities and potential increased among premature offspring in animal mod- differences between how maternal compared with fetal els of inflammation-induced preterm birth.95 microchimeric cells prime and sustain expanded Sharply increased levels of microchimeric cells have memory regulatory T cells represent critically impor- also been described in the target tissue of infants and tant areas for future investigation. children with various autoimmune disorders. For In the larger biological context, reproductive fitness example, significantly increased levels of female chi- during next generation pregnancies conferred by meric cells, presumably of maternal origin, were CHIMERISM 15 identified in muscle biopsy specimens of children with persistent microchimerism in offspring. In turn, the juvenile dermatomyositis or other idiopathic inflam- recognition of individuals in outbred populations as matory myopathies.96,97 Maternal chimeric cells were being ‘constitutively chimeric,’ with non-inherited leg- also uniformly identified in 15 cardiac biopsy samples acy of tolerogenic microchimeric cells, forces reconsid- from infants with neonatal lupus syndrome,98 and 2 eration of immunological identity beyond binary independent case series of liver biopsy specimens definitions of self versus non-self antigen distinction – from infants with biliary atresia.94,99 Together, these that incorporates transmission of maternal attributes clinical observations support the intriguing possibility through matrilineal non-Mendelian heredity.103 Along that allo-reactivity either against or initiated by geneti- with improved outcomes after transplantation, further cally foreign microchimeric cell could represent an mining immunological secrets engrained within mam- important trigger for human autoimmune and autoin- malian reproduction initially recognized by Owen may flammatory disorders.89,100 hold exciting new keys for more effective therapeutic Interestingly however, there is also compelling data strategies for preventing pregnancy complications and for improved survival of maternal compared with reversing autoimmunity. paternal hepatic allografts for infants with liver failure secondary to biliary atresia.96,97,101 Therefore, a defini- Disclosure of potential conflicts of interest tive pathological role for microchimeric maternal cells No potential conflicts of interest were disclosed. in triggering autoimmunity will require additional investigation since enriched chimeric cells in damaged References or diseased tissues may also reflect their participation in tissue repair and regeneration. Furthermore, quali- [1] Owen RD. Immunogenetic Consequences of Vascular Anastomoses between Bovine Twins. Science 1945; tative shifts in the molecular phenotype of microchi- fl 102:400-1; PMID:17755278; http://dx.doi.org/10.1126/ meric cells, timing and in ammatory context of science.102.2651.400 developmental exposure to maternal tissues are each [2] Lillie FR. The Theory of the Free-Martin. Science 1916;

also likely to play profound roles in controlling 43:611-3; PMID:17756274; http://dx.doi.org/10.1126/ whether tolerance or sensitization to NIMA devel- science.43.1113.611 ops.39,40,44,102 Given the conserved nature of NIMA- [3] Owen RD, Wood HR, Foord AG, Sturgeon P, Baldwin LG. Evidence for actively acquired tolerance to Rh anti- specific tolerance and long-term retention of micro- gens. Proc Natl Acad Sci USA 1954; 40:420-4; PMID: chimeric maternal cells in humans and mice, further 16589498; http://dx.doi.org/10.1073/pnas.40.6.420 studies using representative animal models of autoim- [4] Booth PB, Dunsford I, Grant J, Murray S. Haemo- munity that bypass limitations in human tissue avail- lytic disease in first-born infants. Br Med J 1953; ability are likely to be highly informative in dissecting 2:41-2; PMID:13051571; http://dx.doi.org/10.1136/ the beneficial and detrimental impacts of increasingly bmj.2.4826.41-b [5] Claas FH, Gijbels Y, van der Velden-de Munck J, van recognized constitutive chimerism among individuals. Rood JJ. Induction of B cell unresponsiveness to noninherited maternal HLA antigens during fetal life. Science Concluding perspectives 1988; 241:1815-7; PMID:3051377; http://dx.doi.org/ 10.1126/science.3051377 Sixty years since the initial description of actively [6] Burlingham WJ, Grailer AP, Heisey DM, Claas FH, acquired tolerance to Rh antigens by Dr. Ray Owen Norman D, Mohanakumar T, Brennan DC, de Fijter H, van Gelder T, Pirsch JD, et al. The effect of tolerance to have witnessed a dramatic explosion of new data fi noninherited maternal HLA antigens on the survival of highlighting not only the existence of NIMA-speci c renal transplants from sibling donors. N Engl J Med tolerance, but also the translational applicability of this 1998; 339:1657-64; PMID:9834302; http://dx.doi.org/ engrained immunological phenomenon in human 10.1056/NEJM199812033392302 solid organ and stem cell transplantation. With new [7] Eikmans M, van Halteren AG, van Besien K, van Rood JJ, technology for identifying exceptionally rare microchi- Drabbels JJ, Claas FH. Naturally acquired microchimerism: implications for transplantation outcome and novel meric maternal cells and transgenic mouse tools for fi methodologies for detection. Chimerism 2014; 5:24-39; tracking NIMA-speci c immune components, NIMA- PMID:24762743; http://dx.doi.org/10.4161/chim.28908 specific tolerance is now recognized to occur with verti- [8] Hirayama M, Azuma E, Komada Y. Tolerogenic effect cal transmission of maternal cells that establish of non-inherited maternal antigens in hematopoietic 16 J. M. KINDER ET AL.

stem cell transplantation. Front Immunol 2012; 3:135; [17] Dutta P, Molitor-Dart M, Bobadilla JL, Roenneburg DA, PMID:22654885; http://dx.doi.org/10.3389/fimmu. Yan Z, Torrealba JR, Burlingham WJ. Microchimerism 2012.00135 is strongly correlated with tolerance to noninherited [9] van Rood JJ, Loberiza FR, Jr., Zhang MJ, Oudshoorn M, maternal antigens in mice. Blood 2009; 114:3578-87; Claas F, Cairo MS, Champlin RE, Gale RP, Ringden O, PMID:19700665; http://dx.doi.org/10.1182/blood-2009- Hows JM, et al. Effect of tolerance to noninherited 03-213561 maternal antigens on the occurrence of graft-vs.-host [18] Flomenberg N, Dupont B, O’Reilly RJ, Hayward A, Pol- disease after bone marrow transplantation from a par- lack MS. The use of T cell culture techniques to establish ent or an HLA-haploidentical sibling. Blood 2002; the presence of an intrauterine-derived maternal T cell 99:1572-7; PMID:11861270; http://dx.doi.org/10.1182/ graft in a patient with severe combined immunodefi- blood.V99.5.1572 ciency (SCID). Transplantation 1983; 36:733-5; [10] Ichinohe T, Uchiyama T, Shimazaki C, Matsuo K, Tam- PMID:6606881; http://dx.doi.org/10.1097/00007890- aki S, Hino M, Watanabe A, Hamaguchi M, Adachi S, 198336060-00031 Gondo H, et al. Feasibility of HLA-haploidentical [19] Geha RS, Reinherz E. Identification of circulating hematopoietic stem cell transplantation between nonin- maternal T and B lymphocytes in uncomplicated herited maternal antigen (NIMA)-mismatched family severe combined immunodeficiency by HLA typing of members linked with long-term fetomaternal microchi- subpopulations of T cells separated by the fluores- merism. Blood 2004; 104:3821-8; PMID:15280193; cence-activated cell sorter and of Epstein Barr virus- http://dx.doi.org/10.1182/blood-2004-03-1212 derived B cell lines. J Immunol 1983; 130:2493-5; [11] Kanda J, Ichinohe T, Shimazaki C, Hamaguchi M, PMID:6304187. Watanabe A, Ishida H, Yoshihara T, Morimoto A, [20] Kadowaki J, Thompson RI, Zuelzer WW, Woolley Uoshima N, Adachi S, et al. Long-term survival after PV, Jr., Brough AJ, Gruber D. XX-XY lymphoid chi- HLA-haploidentical SCT from noninherited maternal maerism in congenital immunological deficiency syn- antigen-mismatched family donors: impact of chronic dromewiththymicalymphoplasia.Lancet1965; GVHD. Bone Marrow Transplantation 2009; 44:327-9; 2:1152-6; PMID:4158909; http://dx.doi.org/10.1016/ PMID:19219076; http://dx.doi.org/10.1038/bmt.2009.18 S0140-6736(65)92559-6 [12] Okumura H, Yamaguchi M, Kotani T, Sugimori N, [21] Pollack MS, Kapoor N, Sorell M, Kim SJ, Christiansen Sugimori C, Ozaki J, Kondo Y, Yamazaki H, Chuhjo T, FT, Silver DM, Dupont B, O’Reilly RJ. DR-positive

Takami A, et al. Graft rejection and hyperacute graft- maternal engrafted T cells in a severe combined immu- versus-host disease in stem cell transplantation from nodeficiency patient without graft-versus-host disease. non-inherited maternal-antigen-complementary HLA- Transplantation 1980; 30:331-4; PMID:6450472; http:// mismatched siblings. Eur J Haematol 2007; 78:157-60; dx.doi.org/10.1097/00007890-198011000-00004 PMID:17313562; http://dx.doi.org/10.1111/j.1600- [22] Pollack MS, Kirkpatrick D, Kapoor N, Dupont B, 0609.2006.00797.x O’Reilly RJ. Identification by HLA typing of intra- [13] Opelz G. The effect of tolerance to noninherited mater- uterine-derived maternal T cells in four patients with nal HLA antigens on the survival of renal transplants severe combined immunodeficiency. N Eng J Med from sibling donors. N Eng J Med 1999; 340:1369-70; 1982; 307:662-6; http://dx.doi.org/10.1056/NEJM http://dx.doi.org/10.1056/NEJM199904293401715 198209093071106 [14] Andrassy J, Kusaka S, Jankowska-Gan E, Torrealba JR, [23] Thompson LF, O’Connor RD, Bastian JF. Phenotype Haynes LD, Marthaler BR, Tam RC, Illigens BM, Ano- and function of engrafted maternal T cells in patients sova N, Benichou G, et al. Tolerance to noninherited with severe combined immunodeficiency. J Immunol maternal MHC antigens in mice. J Immunol 2003; 1984; 133:2513-7; PMID:6090535. 171:5554-61; PMID:14607963; http://dx.doi.org/ [24] Muller SM, Ege M, Pottharst A, Schulz AS, Schwarz K, 10.4049/jimmunol.171.10.5554 Friedrich W. Transplacentally acquired maternal T lym- [15] Aoyama K, Koyama M, Matsuoka K, Hashimoto D, phocytes in severe combined immunodeficiency: a study Ichinohe T, Harada M, Akashi K, Tanimoto M, of 121 patients. Blood 2001; 98:1847-51; PMID: Teshima T. Improved outcome of allogeneic bone mar- 11535520; http://dx.doi.org/10.1182/blood.V98.6.1847 row transplantation due to breastfeeding-induced toler- [25] Alexander A, Samlowski WE, Grossman D, Bruggers ance to maternal antigens. Blood 2009; 113:1829-33; CS, Harris RM, Zone JJ, Noyes RD, Bowen GM, Leach- PMID:19124834; http://dx.doi.org/10.1182/blood-2008- man SA. Metastatic melanoma in pregnancy: risk of 05-155283 transplacental metastases in the infant. J Clin Oncol [16] Campbell DA, Jr., Lorber MI, Sweeton JC, Turcotte JG, 2003; 21:2179-86; PMID:12775744; http://dx.doi.org/ Niederhuber JE, Beer AE. Breast feeding and maternal- 10.1200/JCO.2003.12.149 donor renal allografts. Possibly the original donor-spe- [26] Isoda T, Ford AM, Tomizawa D, van Delft FW, De Cas- cific transfusion. Transplantation 1984; 37:340-4; tro DG, Mitsuiki N, Score J, Taki T, Morio T, Takagi M, PMID:6369661; http://dx.doi.org/10.1097/00007890- et al. Immunologically silent cancer clone transmission 198404000-00004 from mother to offspring. Proc Natl Acad Sci USA CHIMERISM 17

2009; 106:17882-5; PMID:19822752; http://dx.doi.org/ PMID:17982065; http://dx.doi.org/10.4049/jimmunol. 10.1073/pnas.0904658106 179.10.6749 [27] Walker JW, Reinisch JF, Monforte HL. Maternal pul- [38] Araki M, Hirayama M, Azuma E, Kumamoto T, Iwa- monary adenocarcinoma metastatic to the fetus: first moto S, Toyoda H, Ito M, Amano K, Komada Y. Pre- recorded case report and literature review. Pediatr diction of reactivity to noninherited maternal antigen in Pathol Mol Med 2002; 21:57-69; PMID:11858176; MHC-mismatched, minor histocompatibility antigen- http://dx.doi.org/10.1080/15227950252774174 matched stem cell transplantation in a mouse model. J [28] Lambert NC, Erickson TD, Yan Z, Pang JM, Guthrie Immunol 2010; 185:7739-45; PMID:21078914; http://dx. KA, Furst DE, Nelson JL. Quantification of maternal doi.org/10.4049/jimmunol.1001226 microchimerism by HLA-specific real-time polymerase [39] Dutta P, Burlingham WJ. Microchimerism: tolerance vs. chain reaction: studies of healthy women and women sensitization. Curr Opin Organ Transplant 2011; with scleroderma. Arthritis Rheumatism 2004; 50:906- 16:359-65; PMID:21666480; http://dx.doi.org/10.1097/ 14; PMID:15022334; http://dx.doi.org/10.1002/art. MOT.0b013e3283484b57 20200 [40] Alhajjat AM, Lee AE, Strong BS, Shaaban AF. NK cell [29] Maloney S, Smith A, Furst DE, Myerson D, Rupert K, tolerance as the final endorsement of prenatal tolerance Evans PC, Nelson JL. Microchimerism of maternal ori- after in utero hematopoietic cellular transplantation. gin persists into adult life. J Clin Invest 1999; 104:41-7; Front Pharmacol 2015; 6:51; PMID:25852555; http://dx. PMID:10393697; http://dx.doi.org/10.1172/JCI6611 doi.org/10.3389/fphar.2015.00051 [30] Spurgin LG, Richardson DS. How pathogens drive [41] Merianos D, Heaton T, Flake AW. In utero hemato- genetic diversity: MHC, mechanisms and misunder- poietic stem cell transplantation: progress toward standings. Proc Biol Sci 2010; 277:979-88; PMID: clinical application. Biol Blood Marrow Transplant 20071384; http://dx.doi.org/10.1098/rspb.2009.2084 2008; 14:729-40; PMID:18541191; http://dx.doi.org/ [31] Lo YM, Lau TK, Chan LY, Leung TN, Chang AM. 10.1016/j.bbmt.2008.02.012 Quantitative analysis of the bidirectional fetomaternal [42] Derderian SC, Jeanty C, Walters MC, Vichinsky E, transfer of nucleated cells and plasma DNA. Clin Chem MacKenzie TC. In utero hematopoietic cell transplanta- 2000; 46:1301-9; PMID:10973858. tion for hemoglobinopathies. Front Pharmacol 2014; [32] Molitor ML, Haynes LD, Jankowska-Gan E, Mulder A, 5:278; PMID:25628564. Burlingham WJ. HLA class I noninherited maternal [43] Alhajjat AM, Strong BS, Lee AE, Turner LE, Wadhwani

antigens in cord blood and breast milk. Hum Immunol RK, Ortaldo JR, Heusel JW, Shaaban AF. Prenatal Allo- 2004; 65:231-9; PMID:15041161; http://dx.doi.org/ speciﬁc NK Cell Tolerance Hinges on Instructive Allor- 10.1016/j.humimm.2003.12.006 ecognition through the Activating Receptor during [33] Zhou L, Yoshimura Y, Huang Y, Suzuki R, Yokoyama Development. J Immunol 2015; 195(4):1506-16. M, Okabe M, Shimamura M. Two independent path- [44] Durkin ET, Jones KA, Rajesh D, Shaaban AF. Early chi- ways of maternal cell transmission to offspring: through merism threshold predicts sustained engraftment and placenta during pregnancy and by breast-feeding after NK-cell tolerance in prenatal allogeneic chimeras. Blood birth. Immunol 2000; 101:570-80; http://dx.doi.org/ 2008; 112:5245-53; PMID:18796629; http://dx.doi.org/ 10.1046/j.1365-2567.2000.00144.x 10.1182/blood-2007-12-128116 [34] Mold JE, McCune JM. Immunological tolerance during [45] Kroemer A, Edtinger K, Li XC. The innate natural killer fetal development: from mouse to man. Adv Immunol cells in transplant rejection and tolerance induction. 2012; 115:73-111; PMID:22608256; http://dx.doi.org/ Curr Opin Organ Transplant 2008; 13:339-43; 10.1016/B978-0-12-394299-9.00003-5 PMID:18685327; http://dx.doi.org/10.1097/MOT. [35] Mold JE, Michaelsson J, Burt TD, Muench MO, Becker- 0b013e3283061115 man KP, Busch MP, Lee TH, Nixon DF, McCune JM. [46] Ashizuka S, Peranteau WH, Hayashi S, Flake AW. Maternal alloantigens promote the development of tol- Busulfan-conditioned bone marrow transplantation erogenic fetal regulatory T cells in utero. Science 2008; results in high-level allogeneic chimerism in mice made 322:1562-5; PMID:19056990; http://dx.doi.org/10.1126/ tolerant by in utero hematopoietic cell transplantation. science.1164511 Exp Hematol 2006; 34:359-68; PMID:16543070; http:// [36] Billingham RE, Brent L, Medawar PB. Actively dx.doi.org/10.1016/j.exphem.2005.11.011 acquired tolerance of foreign cells. Nature 1953; [47] Kim HB, Shaaban AF, Milner R, Fichter C, Flake AW. 172:603-6; PMID:13099277; http://dx.doi.org/10.1038/ In utero bone marrow transplantation induces donor- 172603a0 speciﬁc tolerance by a combination of clonal deletion [37] Molitor-Dart ML, Andrassy J, Kwun J, Kayaoglu HA, and clonal anergy. J Pediatr Surg 1999; 34:726-9; discus- Roenneburg DA, Haynes LD, Torrealba JR, Bobadilla JL, sion 9-30; PMID:10359172; http://dx.doi.org/10.1016/ Sollinger HW, Knechtle SJ, et al. Developmental expo- S0022-3468(99)90364-0 sure to noninherited maternal antigens induces CD4+ T [48] Kim HB, Shaaban AF, Yang EY, Liechty KW, Flake AW. regulatory cells: relevance to mechanism of heart Microchimerism and tolerance after in utero bone mar- allograft tolerance. J Immunol 2007; 179:6749-61; row transplantation in mice. J Surg Res 1998; 77:1-5; 18 J. M. KINDER ET AL.

PMID:9698523; http://dx.doi.org/10.1006/jsre.1997. [59] Gratz IK, Rosenblum MD, Maurano MM, Paw JS, 5255 Truong HA, Marshak-Rothstein A, Abbas AK. Cutting [49] Merianos DJ, Tiblad E, Santore MT, Todorow CA, Laje edge: Self-antigen controls the balance between effector P, Endo M, Zoltick PW, Flake AW. Maternal alloanti- and regulatory T cells in peripheral tissues. J Immunol bodies induce a postnatal immune response that lim- 2014; 192:1351-5; PMID:24442443; http://dx.doi.org/ its engraftment following in utero hematopoietic cell 10.4049/jimmunol.1301777 transplantation in mice. J Clin Invest 2009; [60] Sanchez AM, Zhu J, Huang X, Yang Y. The development 119:2590-600; PMID:19652363. and function of memory regulatory T cells after acute [50] Jeanty C, Derderian SC, Mackenzie TC. Maternal- viral infections. J Immunol 2012; 189:2805-14; fetal cellular trafficking: clinical implications and PMID:22855712; http://dx.doi.org/10.4049/jimmunol. consequences. Curr Opin Pediatr 2014; 26:377-82; 1200645 PMID:24759226; http://dx.doi.org/10.1097/MOP. [61] Homann D, Teyton L, Oldstone M. Differential regula- 0000000000000087 tion of antiviral T-cells immunity results stable CD8+ [51] Kinder JM, Jiang TT, Ertelt JM, Xin L, Strong BS, Shaa- by declining CD4+ T-cell memory. Nat Med 2001; ban AF, Way SS. Cross-generational reproductive fit- 7:913-9; PMID:11479623; http://dx.doi.org/10.1038/ ness enforced by microchimeric maternal cells. Cell 90950 2015; 162:505-15; PMID:26213383. [62] Nelson RW, McLachlan JB, Kurtz JR, Jenkins MK. CD4 [52] Moon JJ, Dash P, Oguin TH, 3rd, McClaren JL, Chu + T cell persistence and function after infection are HH, Thomas PG, Jenkins MK. Quantitative impact of maintained by low-level peptide:MHC class II presenta- thymic selection on Foxp3+ and Foxp3- subsets of self- tion. J Immunol 2013; 190:2828-34; PMID:23382562; peptide/MHC class II-specific CD4+ T cells. Proc Natl http://dx.doi.org/10.4049/jimmunol.1202183 Acad Sci USA 2011; 108:14602-7; PMID:21873213; [63] Seder R, Ahmed R. Similarities and differences in CD4+ http://dx.doi.org/10.1073/pnas.1109806108 and CD8 effector and memory T cells generation. Nat [53] Rees W, Bender J, Teague TK, Kedl RM, Crawford F, Immunol 2003; 4:835-42; PMID:12942084; http://dx. Marrack P, Kappler J. An inverse relationship between doi.org/10.1038/ni969 T cell receptor affinity and antigen dose during CD4(+) [64] Uzonna JE, Wei G, Yurkowski D, Bretscher P. Immune T cell responses in vivo and in vitro. Proc Natl Acad Sci elimination of Leishmania major in mice: implications USA 1999; 96:9781-6; PMID:10449771; http://dx.doi. for immune memory, vaccination, and reactivation dis-

org/10.1073/pnas.96.17.9781 ease. J Immunol 2001; 167:6967-74; PMID:11739516; [54] Moon J, Chu H, Pepper M, McSorley S, Jameson S, Kedl http://dx.doi.org/10.4049/jimmunol.167.12.6967 R, Jenkins M. Naive CD4(+) T cell frequency varies for [65] Williams MA, Holmes BJ, Sun JC, Bevan MJ. Develop- different epitopes and predicts repertoire diversity and ing and maintaining protective CD8+ memory T cells. response magnitude. Immunity 2007; 27:203-13; Immunol Rev 2006; 211:146-53; PMID:16824124; PMID:17707129; http://dx.doi.org/10.1016/j.immuni. http://dx.doi.org/10.1111/j.0105-2896.2006.00389.x 2007.07.007 [66] Adkins B, Leclerc C, Marshall-Clarke S. Neonatal adap- [55] Bakkour S, Baker CA, Tarantal AF, Wen L, Busch MP, tive immunity comes of age. Nat Rev Immunol 2004; Lee TH, McCune JM. Analysis of maternal microchi- 4:553-64; PMID:15229474; http://dx.doi.org/10.1038/ merism in rhesus monkeys (Macaca mulatta) using nri1394 real-time quantitative PCR amplification of MHC poly- [67] Erlebacher A. Mechanisms of T cell tolerance towards morphisms. Chimerism 2014; 5:6-15; PMID:24451553; the allogeneic fetus. Nat Rev Immunol 2013; 13:23-33; http://dx.doi.org/10.4161/chim.27778 PMID:23237963; http://dx.doi.org/10.1038/nri3361 [56] Shaaban AF, Kim HB, Gaur L, Liechty KW, Flake AW. [68] Munoz-Suano A, Hamilton AB, Betz AG. Gimme shel- Prenatal transplantation of cytokine-stimulated marrow ter: the immune system during pregnancy. Immunol improves early chimerism in a resistant strain combina- Rev 2011; 241:20-38; PMID:21488887; http://dx.doi. tion but results in poor long-term engraftment. Exp org/10.1111/j.1600-065X.2011.01002.x Hematol 2006; 34:1278-87; PMID:16939821. [69] Samstein RM, Josefowicz SZ, Arvey A, Treuting PM, [57] Brincks E, Roberts A, Cookenham T, Sell S, Kohlmeier J, Rudensky AY. Extrathymic generation of regulatory T Blackman M, Woodland D. Antigen-specific memory cells in placental mammals mitigates maternal-fetal regulatory CD4+FoxP3+ T cells control memory conflict. Cell 2012; 150:29-38; PMID:22770213; http:// responses to influenza virus infection. J Immunol 2013; dx.doi.org/10.1016/j.cell.2012.05.031 190:3438-46; PMID:23467933; http://dx.doi.org/ [70] Chaturvedi V, Ertelt JM, Jiang TT, Kinder JM, Xin L, 10.4049/jimmunol.1203140 OwensKJ,JonesHN,WaySS.CXCR3blockadepro- [58] Rosenblum M, Gratz I, Paw J, Lee K, Marshak-Roth- tects against Listeria monocytogenes infection- stein A, Abbas A. Response to self antigen imprints induced fetal wastage. J Clin Invest 2015; 125:1713- regulatory memory in tissues. Nature 2011; 480:538- 25; PMID:25751061; http://dx.doi.org/10.1172/ 42; PMID:22121024. JCI78578 CHIMERISM 19

[71] Rowe JH, Ertelt JM, Aguilera MN, Farrar MA, Way Microchimerism in female bone marrow and bone dec- SS. Foxp3(+) regulatory T cell expansion required ades after fetal mesenchymal stem-cell trafﬁcking in for sustaining pregnancy compromises host defense pregnancy. Lancet 2004; 364:179-82; PMID:15246731; against prenatal bacterial pathogens. Cell Host http://dx.doi.org/10.1016/S0140-6736(04)16631-2 Microbe 2011; 10:54-64; PMID:21767812; http://dx. [84] Goulder PJ, Watkins DI. Impact of MHC class I diver- doi.org/10.1016/j.chom.2011.06.005 sity on immune control of immunodeﬁciency virus rep- [72] Rowe JH, Ertelt JM, Xin L, Way SS. Listeria monocy- lication. Nat Rev Immunol 2008; 8:619-30; togenes cytoplasmic entry induces fetal wastage by PMID:18617886; http://dx.doi.org/10.1038/nri2357 disrupting maternal FoxP3+ regulatory cell-sustained [85] Oliver MK, Telfer S, Piertney SB. Major histocom- fetal tolerance. PLoS Pathog 2012; 8:e1002873; patibility complex (MHC) heterozygote superiority to PMID:22916020; http://dx.doi.org/10.1371/journal. natural multi-parasite infections in the water vole ppat.1002873 (Arvicola terrestris). Proc Biol Sci 2009; 276: [73] Rowe JH, Ertelt JM, Xin L, Way SS. Pregnancy imprints 1119-28; PMID:19129114; http://dx.doi.org/10.1098/ regulatory memory that sustains anergy to fetal antigen. rspb.2008.1525 Nature 2012; 490:102-6; PMID:23023128; http://dx.doi. [86] Israeli M, Kristt D, Nardi Y, Klein T. Genetic Consider- org/10.1038/nature11462 ations in Human Sex-Mate Selection: Partners Share [74] Gammill HS, Adams Waldorf KM, Aydelotte TM, Lucas Human Leukocyte Antigen but not Short-Tandem- J, Leisenring WM, Lambert NC, Nelson JL. Pregnancy, Repeat Identity Markers. Am J Reprod Immunol 2014; microchimerism, and the maternal grandmother. PLoS 71:467-71; PMID:24589062; http://dx.doi.org/10.1111/ One 2011; 6:e24101; PMID:21912617; http://dx.doi.org/ aji.12213 10.1371/journal.pone.0024101 [87] Chaix R, Cao C, Donnelly P. Is mate choice in [75] Gammill HS, Stephenson MD, Aydelotte TM, Nelson humans MHC-dependent? PLoS Genet 2008; 4: JL. Microchimerism in recurrent miscarriage. Cell Mol e1000184; PMID:18787687; http://dx.doi.org/10.1371/ Immunol 2014; 11:589-94; PMID:25242272; http://dx. journal.pgen.1000184 doi.org/10.1038/cmi.2014.82 [88] Ober C, Weitkamp LR, Cox N, Dytch H, Kostyu D, [76] Gammill HS, Stephenson MD, Aydelotte TM, Nelson Elias S. HLA and mate choice in humans. Am J Hum JL. Microchimerism in women with recurrent miscar- Genet 1997; 61:497-504; PMID:9326314 http://dx.doi. riage. Chimerism 2014; 5:103-5; PMID:25779348; org/10.1086/515511

http://dx.doi.org/10.1080/19381956.2015.1017241 [89] Nelson JL. The otherness of self: microchimerism in [77] Campbell D, MacGillivray I, Carr-Hill R. Pre-eclampsia health and disease. Trends Immunol 2012; 33:421-7; in second pregnancy. Br J Obstet Gynaecol 1985; PMID:22609148; http://dx.doi.org/; http://dx.doi.org/ 92:131-40; PMID:3970893; http://dx.doi.org/10.1111/ 10.1016/j.it.2012.03.002 j.1471-0528.1985.tb01064.x [90] Stevens AM. Microchimeric cells in systemic lupus [78] Trupin L, Simon L, Eskenazi B. Change in paternity: erythematosus: targets or innocent bystanders? Lupus a risk factor for preeclampsia in multiparas. Epide- 2006; 15:820-6; PMID:17153857; http://dx.doi.org/ miol 1996; 7:240-4; http://dx.doi.org/10.1097/ 10.1177/0961203306070068 00001648-199605000-00004 [91] Nelson JL, Gillespie KM, Lambert NC, Stevens AM, [79] Kinder JM, Jiang TT, Clark DR, Chaturvedi V, Xin L, Loubiere LS, Rutledge JC, Leisenring WM, Erickson Ertelt JM, Way SS. Pregnancy-induced maternal regula- TD, Yan Z, Mullarkey ME, et al. Maternal microchimer- tory T cells, bona ﬁde memory or maintenance by anti- ism in peripheral blood in type 1 diabetes and pancre- genic reminder from fetal cell microchimerism? atic islet b cell microchimerism. PNAS 2007; 104:1637- Chimerism 2014; 5:16-9; PMID:24553046; http://dx.doi. 42; PMID:17244711; http://dx.doi.org/10.1073/ org/10.4161/chim.28241 pnas.0606169104 [80] Bianchi D, Zickwolf G, Weil G, Sylvester S, DeMaria M. [92] Feitsma AL, Worthington J, van der Helm-van Mil Male fetal progenitor cells persist in maternal blood for AH, Plant D, Thomson W, Ursum J, van Schaarden- as long as 27 years postpartum. PNAS 1996; 93:705-8; burg D, van der Horst-Bruinsma IE, van Rood JJ, PMID:8570620; http://dx.doi.org/10.1073/pnas.93.2.705 Huizinga TW, et al. Protective effect of noninherited [81] Trowsdale J, Betz A. Mother’s little helpers: mechanisms maternal HLA-DR antigens on rheumatoid arthritis of maternal-fetal tolerance. Nat Immunol 2006; 7:241-6; development. Proc Natl Acad Sci USA 2007; PMID:16482172; http://dx.doi.org/10.1038/ni1317 104:19966-70; PMID:18077428; http://dx.doi.org/ [82] Adams KM, Lambert NC, Heimfeld S, Tylee TS, Pang 10.1073/pnas.0710260104 JM, Erickson TD, Nelson JL. Male DNA in female [93] Rak JM, Maestroni L, Balandraud N, Guis S, Boudinet donor apheresis and CD34-enriched products. Blood H, Guzian MC, Yan Z, Azzouz D, Auger I, Roudier C, 2003; 102:3845-7; PMID:12869496; http://dx.doi.org/ et al. Transfer of the shared epitope through microchi- 10.1182/blood-2003-05-1570 merism in women with rheumatoid arthritis. Arthritis [83] O’Donoghue K, Chan J, de la Fuente J, Kennea N, San- Rheum 2009; 60:73-80; PMID:19117368; http://dx.doi. dison A, Anderson JR, Roberts IA, Fisk NM. org/10.1002/art.24224 20 J. M. KINDER ET AL.

[94] van der Horst-Bruinsma IE, Hazes JM, Schreuder block. Lancet 2003; 362:1617-23; PMID:14630442; GM, Radstake TR, Barrera P, van de Putte LB, Mus- http://dx.doi.org/10.1016/S0140-6736(03)14795-2 tamu D, van Schaardenburg D, Breedveld FC, de [99] Suskind DL, Rosenthal P, Heyman MB, Kong D, Vries RR. Influence of non-inherited maternal HLA- Magrane G, Baxter-Lowe LA, Muench MO. Maternal DR antigens on susceptibility to rheumatoid arthritis. microchimerism in the livers of patients with biliary Annal rheum Dis 1998; 57:672-5; http://dx.doi.org/ atresia. BMC Gastroenterol 2004; 4:14; PMID:15285784; 10.1136/ard.57.11.672 http://dx.doi.org/10.1186/1471-230X-4-14 [95] Wegorzewska M, Le T, Tang Q, MacKenzie TC. [100] Gammill H, Nelson J. Naturally acquired microchimer- Increased maternal T cell microchimerism in the alloge- ism. Int J Dev Biol 2010; 54:531-43; PMID:19924635; neic fetus during LPS-induced preterm labor in mice. http://dx.doi.org/10.1387/ijdb.082767hg Chimerism 2015:1-6; PMID:25779065; http://dx.doi. [101] Nijagal A, Fleck S, Hills NK, Feng S, Tang Q, Kang org/10.1080/19381956.2014.1002703 SM, Rosenthal P, MacKenzie TC. Decreased risk of [96] Reed AM, Picornell YJ, Harwood A, Kredich DW. Chi- graft failure with maternal liver transplantation in merism in children with juvenile dermatomyositis. Lan- patients with biliary atresia. Am J Transplant 2012; cet 2000; 356:2156-7; PMID:11191546; http://dx.doi. 12:409-19; PMID:22221561; http://dx.doi.org/ org/10.1016/S0140-6736(00)03500-5 10.1111/j.1600-6143.2011.03895.x [97] Artlett CM, Ramos R, Jiminez SA, Patterson K, Miller [102] Dutta P, Dart M, Roenneburg DA, Torrealba JR, FW, Rider LG. Chimeric cells of maternal origin in juve- Burlingham WJ. Pretransplant immune-regulation nile idiopathic inflammatory myopathies. Childhood predicts allograft tolerance. Am J Transplant 2011; Myositis Heterogeneity Collaborative Group. Lancet 11:1296-301; PMID:21449933; http://dx.doi.org/ 2000; 356:2155-6; PMID:11191545; http://dx.doi.org/ 10.1111/j.1600-6143.2011.03484.x 10.1016/S0140-6736(00)03499-1 [103] Maurel MC, Kanellopoulos-Langevin C. Heredity-ventur- [98] Stevens AM, Hermes HM, Rutledge JC, Buyon JP, Nel- ing beyond genetics. Biol Reprod 2008; 79:2-8; son JL. Myocardial-tissue-specific phenotype of mater- PMID:18401011; http://dx.doi.org/10.1095/biolreprod.107. nal microchimerism in neonatal lupus congenital heart 065607 Immunity Previews

determine whether glycobiological status deficiency aggravated immunopathology Baumeister, S.H., Freeman, G.J., Dranoff, G., and maintains T cell homeostasis through and increased mortality in the Cl13 Sharpe, A.H. (2016). Annu. Rev. Immunol., Pub- lished online February 25, 2016. http://dx.doi.org/ + modification of multiple immune-inhibi- model in a CD4 -T-cell-dependent 10.1146/annurev-immunol-032414-112049. tory receptors. In keeping with this hy- fashion. Discrimination between PSGL-1 pothesis, several immune checkpoint glycoforms on CD4+ versus on CD8+ Carlow, D.A., Gossens, K., Naus, S., Veerman, K.M., Seo, W., and Ziltener, H.J. (2009). Immunol. proteins, including PD-1, are glycosylated T cells could enable the development of Rev. 230, 75–96. or possess consensus sequences for less toxic, CD8+-T-cell-specific therapeu- glycosylation. Moreover, PD-1 modulates tics. Confirmation of PSGL-1 immunoreg- Chang, C.H., Qiu, J., O’Sullivan, D., Buck, M.D., glycolysis and resultant glycan substrate ulatory function in humans will be Noguchi, T., Curtis, J.D., Chen, Q., Gindin, M., Gubin, M.M., van der Windt, G.J., et al. (2015). availability (Chang et al., 2015), as well important, especially given that human Cell 162, 1229–1241. as motility and effector function of CD4+ and murine PSGL-1 vary in amino acid and CD8+ T cells (Zinselmeyer et al., sequence, glycan architecture, and Matsumoto, M., Miyasaka, M., and Hirata, T. (2009). J. Immunol. 183, 7204–7211. 2013). Consequently, PD-1’s glycosyla- sulfation, particularly at the N-terminal tion state might influence its biological selectin-binding region (Barthel et al., Tinoco, R., Carrette, F., Barraza, M.L., Otero, D.C., activity. 2007). Although further investigation Magan˜ a, J., Bosenberg, M.W., Swain, S.L., and 44 Because PSGL-1 deficiency promotes will be required to determine the role Bradley, L.M. (2016). Immunity , this issue, 1190–1203. antiviral and tumor-specific immunity of PSGL-1 in immune checkpoint regula- through suppression of several T-cell- tion in humans, the findings by Tinoco Veerman, K.M., Williams, M.J., Uchimura, K., inhibitory receptors, this work has exciting et al. (2016) provide impetus to examine Singer, M.S., Merzaban, J.S., Naus, S., Carlow, D.A., Owen, P., Rivera-Nieves, J., Rosen, S.D., translational implications. For instance, the therapeutic potential of combining and Ziltener, H.J. (2007). Nat. Immunol. 8, combination checkpoint blockade has PSGL-1 inhibition with immune check- 532–539. yielded unprecedented response rates point blockade. in patients with cancer (Baumeister et al., Veerman, K.M., Carlow, D.A., Shanina, I., Priatel, J.J., Horwitz, M.S., and Ziltener, H.J. (2012). 2016). Thus, PSGL-1 inhibitors could J. Immunol. 188, 1638–1646. potentially improve T cell responses to REFERENCES cancer and other immune-associated dis- Zinselmeyer, B.H., Heydari, S., Sacrista´ n, C., Barthel, S.R., Gavino, J.D., Descheny, L., and Di- Nayak, D., Cammer, M., Herz, J., Cheng, X., Davis, orders by inhibiting multiple checkpoints mitroff, C.J. (2007). Expert Opin. Ther. Targets S.J., Dustin, M.L., and McGavern, D.B. (2013). concurrently. However, general PSGL-1 11, 1473–1491. J. Exp. Med. 210, 757–774.

Offspring’s Tolerance of Mother Goes Viral

Jeremy M. Kinder,1 Tony T. Jiang,1 and Sing Sing Way1,* 1Division of Infectious Diseases and Perinatal Institute, Cincinnati Children’s Hospital, 3333 Burnet Avenue, Cincinnati, OH 45229, USA *Correspondence: [email protected] http://dx.doi.org/10.1016/j.immuni.2016.04.021

Pregnancy uniquely allows genetically discordant tissues of the mother and child to intimately coexist in harmony. In this issue of Immunity, Ou and colleagues show that hepatitis B virus exploits these naturally occurring immune tolerance pathways to establish persistent postnatal infection in offspring.

An elaborate assortment of immuno- developmental window. Given the pre- nity allowing for potentially more severe logical shifts takes place during preg- dominant importance of reproduction or persistent infections. Does tolerance nancy to accommodate the intimate for species survival, it is perhaps not to innocuous non-inherited maternal approximation of genetically discordant surprising that many non-overlapping antigens in offspring extend to maternal and fetal tissues. Although mechanisms are in place to sustain this foreign microbes transmitted from the this is typically studied from the view- bi-directional need for expanded immu- mother? Could these tolerogenic path- point of expanded maternal tolerance nological tolerance. However, this ways promote susceptibility in offspring to foreign paternal antigens expressed assortment of immunological shifts de- to vertically transmitted microbial in- by the developing fetus, the fetus is signed to ensure tolerance to foreign vaders? In this issue of Immunity, Tian reciprocally exposed to an equally vast beneﬁcial antigens can also create holes et al. (2016) uncover one mechanism array of genetically foreign non-in- in host defense, which can be exploited whereby perinatal exposure to hepatitis herited maternal antigens in this unique by microbial pathogens to evade immu- B virus (HBV) might potentiate persistent

Immunity 44, May 17, 2016 1085 Immunity Previews

increased frequency of hepatic macro- In utero hepatitis B exposure No congenital exposure phage cells, known as Kupffer cells, upre- TNF-α gulate expression of the primary PD-1 ligand, PD-L1. This interaction proved to Macrophage IL-10 Macrophage IL-1β be essential, because either blockade of iNOS PD-1-PD-L1 interactions with neutralizing antibody or bulk depletion of macro- HBV HBV phages each efficiently overturned the susceptibility to persistent infection of offspring sensitized by in utero HBV pMHCI PD-L1 pMHCI exposure. Together, these findings high- PD-1 TCR TCR light how in utero sensitization with CD8+ T cell microbial antigens can drive drastic shifts in the differentiation and co-stimulatory CD8+ T cell potential of macrophage cells within IFN-γ target tissues, which ultimately dictates the outcome of infection. Along with add- IFN-γ ing ‘‘bad macrophage cells’’ to the list of potential therapeutic targets for re-invigo- Granzyme B Granzyme B rating the activation of exhausted CD8 T cells during persistent HBV infection, these results demonstrating functional Persistent viral infection Self-resolving acute infection exhaustion of microbe-specific CD8+ T cells controlled by tissue resident Figure 1. Prenatal Exposure to Hepatitis B Virus-Infected Mothers Promotes Viral macrophage cells also add an exciting Persistence in Offspring new dimension whereby antigenic ex- In utero hepatitis B virus exposure from infected mothers dampens the anti-viral response in offspring, posure in early development engenders resulting in persistent viral infection. This is accomplished by skewing neonatal macrophages toward a immunological tolerance. tolerant phenotype (IL-10 production, arginase-1 [Arg1], and mannose receptor 1 [Mrc1] expression) and expression of PD-L1 to maintain a functionally exhausted phenotype for HBV-specific CD8+ T cells (left- While this unfortunate outcome sided panel). By contrast, when congenital hepatitis B virus exposure is eliminated in genetically identical might reflect engagement of bystander offspring, a robust innate inflammatory response and T cell receptor (TCR) engagement by peptide + MHC tolerance in place to suppress harm- (pMHCI) drives expansion of activated viral-specific CD8+ T cells resulting in self-resolving acute infection (right-sided panel). ful neonatal responses to genetically foreign maternal antigens, Tian et al. demonstrate a far more deliberate and infection in offspring through immuno- made offspring drastically more suscepti- sinister plot by showing that HBV inten- logical shifts driven by the conserved ble to postnatal HBV infection, mimicked tionally exploits these immune tolerance viral e antigen (HBeAg). by injection of HBV genomic DNA, sug- pathways for its benefit. Specifically, the Acquisition of HBV in adults usually gesting that in utero sensitization drives authors show that shifts in Kupffer cell causes only short-lived acute infection. immunological shifts that favor long-term polarization with in utero sensitization However, perinatally acquired HBV HBV persistence. is mediated by the conserved HBV spe- carries a 90% risk of persistent infection, In contrast to suppressed responsive- cific protein HBeAg, whose biological which in large part accounts for the ness to non-inherited maternal cellular role had been previously poorly defined. majority of the estimated 250 million antigens that occurs with accumula- Susceptibility to persistent viral infection chronically infected individuals (Schweit- tion of immune-suppressive regulatory was overturned when HBeAg exposure zer et al., 2015). To investigate whether CD4+ T cells (Kinder et al., 2015; Mold was eliminated either during in utero this discordance in viral control stems et al., 2008), in utero sensitization to sensitization or postnatal infection. In from bystander immune tolerance HBV in offspring was associated with turn, protective immunity against persis- engendered during pregnancy, the au- blunted expansion of HBV-specific CD8+ tent infection caused by HBV strains thors developed an innovative preclinical T cells, and their diminished production lacking only the HBeAg paralleled approach for exposing offspring to HBV- of molecules associated with cytolytic restored production of pro-inflammatory infected mothers. By mating female mice activity such as interferon-g (IFN-g) molecules (e.g., nitric oxide synthase, encoding HBV genomic DNA on only and granzyme B (Figure 1). Functional TNF-a,andIL-1b), with reciprocally one of two homologous chromosomes quiescence of HBV-specific CD8+ T cells blunted expression of anti-inflammatory with non-transgenic males, half of the also paralleled sharply increased expres- molecules (e.g., arginase-1, mannose offspring do not inherit HBV encoding sion of the inhibitory coreceptor PD-1, receptor-1, IL-10) by hepatic macro- DNA but were exposed to a full comple- a molecule associated with an exhau- phage cells (Figure 1). Together, these ment of HBV antigens from their mothers sted phenotype among T cells during results indicate that the conservation during in utero development. Remarkably, persistent infection and cancer (Pauken of HBeAg is maintained to precondition prenatal exposure to HBV in this context and Wherry, 2015). Additionally, an a more tolerant-inducing phenotype for

1086 Immunity 44, May 17, 2016 Immunity Previews

neonatal macrophage cells that pro- considering the efficiency whereby microbes that selectively cause more motes persistent infection following vaccination in the early newborn period severe or persistent infections when ac- perinatal HBV acquisition. protects against perinatal HBV trans- quired in early infancy. In turn, while Ou An interesting extension of these mission (Poland and Jacobson, 2004), and colleagues show that disruption of findings is the memory-like tolerant this tolerance might alternatively be dis- the PD-1-PD-L1 interaction or deple- phenotype of classical innate immune rupted with cognate antigen encounter tion of macrophages each promotes components such as macrophage cells in a highly inflammatory milieu. None- the resolution of persistent HBV infec- primed by HBeAg. While the ability theless, these findings underscore the tion, these results also highlight the to ‘‘remember’’ has traditionally been remarkable ability whereby prenatal remarkable detrimental potential when thought to be a unique property of adap- stimulation can potently shape the func- naturally existing immune tolerance dur- tive immune components (T and B cells), tional repertoire of both antigen-spe- ing pregnancy goes viral. there is emerging evidence now that cific adaptive and non-antigen-specific non-antigen-specific immune compo- innate immune components. nents can also be similarly ‘‘trained’’ Tian et al. highlight a mechanism REFERENCES (Levy and Wynn, 2014). This education whereby maternal HBV infection actively or training of non-antigen-specific im- influences the development of immune Elahi, S., Ertelt, J.M., Kinder, J.M., Jiang, T.T., Zhang, X., Xin, L., Chaturvedi, V., Strong, B.S., mune components is likely to play espe- cells within offspring to promote its Qualls, J.E., Steinbrecher, K.A., et al. (2013). Na- cially prominent roles during in utero persistence. It is clear that HBV, through ture 504, 158–162. development and early infancy when the conserved expression of HBeAg, Kinder, J.M., Jiang, T.T., Ertelt, J.M., Xin, L., adaptive immune cells are under-devel- has evolved to exploit these natural tol- Strong, B.S., Shaaban, A.F., and Way, S.S. oped or intentionally biased toward a erogenic pathways, but do other micro- (2015). Cell 162, 505–515. more tolerant phenotype to handle the bial pathogens take advantage of these Levy, O., and Wynn, J.L. (2014). Neonatology 105, barrage of foreign innocuous maternal mechanisms to promote persistence 136–141. and commensal antigens (Elahi et al., after congenital or perinatal infection? Mold, J.E., and McCune, J.M. (2012). Adv. Immu- 2013; Levy and Wynn, 2014; Mold and In addition to HBV, maternal infection nol. 115, 73–111. McCune, 2012). with several other viral pathogens (e.g., Therefore, to establish persistent herpes simplex virus, cytomegalovirus, Mold, J.E., Michae¨ lsson, J., Burt, T.D., Muench, M.O., Beckerman, K.P., Busch, M.P., Lee, T.H., infection, HBV has apparently learned rubella, HIV) causes high risk of vertical Nixon, D.F., and McCune, J.M. (2008). Science how to co-op this tolerogenic pheno- transmission or perinatal infection with 322, 1562–1565. type of neonatal macrophage cells and often devastating consequences for the Pauken, K.E., and Wherry, E.J. (2015). Trends their ability to be trained. However, as child (Silasi et al., 2015). Therefore, Immunol. 36, 265–276. susceptibility to persistent infection investigating whether other perinatally fades when HBV is administered to older acquired pathogens exploit the cross- Poland, G.A., and Jacobson, R.M. (2004). N. Engl. J. Med. 351, 2832–2838. mice, at time points further removed talk between polarized macrophage from in utero sensitization (Tian et al., cells and functionally exhausted viral- Schweitzer, A., Horn, J., Mikolajczyk, R.T., Krause, 2016), trained immunity in this context specific CD8+ T cells described by Tian G., and Ott, J.J. (2015). Lancet 386, 1546–1555. is also not permanent—potentially re- et al. (2016), or potentially other toler- Silasi, M., Cardenas, I., Kwon, J.Y., Racicot, K., flecting a physiological shift to other ance pathways activated during preg- Aldo, P., and Mor, G. (2015). Am. J. Reprod. Immu- immune component layers with the nancy, have exciting potential to spur nol. 73, 199–213. progression of postnatal development the development of host-directed thera- Tian, Y., Kuo, C.F., Akbari, O., and Ou, J.H. (2016). (Mold and McCune, 2012). Furthermore, pies to combat this wide diversity of Immunity 44, this issue, 1204–1214.

Immunity 44, May 17, 2016 1087 Article

Cross-Generational Reproductive Fitness Enforced by Microchimeric Maternal Cells

Graphical Abstract Authors Jeremy M. Kinder, Tony T. Jiang, James M. Ertelt, ..., Beverly S. Strong, Aimen F. Shaaban, Sing Sing Way

Correspondence [email protected]

In Brief Selective accumulation of immune suppressive regulatory T cells in female offspring in response to maternal cell microchimerism enforces tolerance to overlapping fetal antigens during next- generation pregnancies. This highly conserved mechanism promotes reproductive ﬁtness by preserving conservation of non-inherited maternal traits.

Highlights d Microchimeric maternal cells drive postnatal persistence of NIMA-speciﬁc tolerance d NIMA-speciﬁc immune suppressive Tregs selectively accumulate in female offspring d Overlap between NIMA and fetal antigen during pregnancy accentuates fetal tolerance d Cross-generational protection against fetal wastage enforced by NIMA tolerance

Kinder et al., 2015, Cell 162, 1–11 July 30, 2015 ª2015 Elsevier Inc. http://dx.doi.org/10.1016/j.cell.2015.07.006 Please cite this article in press as: Kinder et al., Cross-Generational Reproductive Fitness Enforced by Microchimeric Maternal Cells, Cell (2015), http://dx.doi.org/10.1016/j.cell.2015.07.006 Article

Cross-Generational Reproductive Fitness Enforced by Microchimeric Maternal Cells

Jeremy M. Kinder,1 Tony T. Jiang,1 James M. Ertelt,1 Lijun Xin,1 Beverly S. Strong,2 Aimen F. Shaaban,2 and Sing Sing Way1,* 1Division of Infectious Diseases and Perinatal Institute 2Center for Fetal Cellular and Molecular Therapy Cincinnati Children’s Hospital. 3333 Burnet Avenue, Cincinnati, OH 45229, USA *Correspondence: [email protected] http://dx.doi.org/10.1016/j.cell.2015.07.006

SUMMARY 1954) and selective anergy to NIMA-specific HLA haplotypes among transfusion dependent individuals broadly exposed to Exposure to maternal tissue during in utero develop- foreign HLA (Claas et al., 1988). More recently, prolonged sur- ment imprints tolerance to immunologically foreign vival of NIMA-matched human allografts after solid organ trans- non-inherited maternal antigens (NIMA) that persists plantation (Burlingham et al., 1998) and reduced graft versus into adulthood. The biological advantage of this host disease among NIMA-matched stem cell transplants high- tolerance, conserved across mammalian species, re- light clinical benefits of NIMA-specific tolerance that persists in mains unclear. Here, we show maternal cells that individuals through adulthood (Ichinohe et al., 2004; Eikmans et al., 2014; van Rood et al., 2002). establish microchimerism in female offspring during In human development, tolerance to mother begins in utero development promote systemic accumulation of im- with suppressed activation of maturing immune cells with mune suppressive regulatory T cells (Tregs) with NIMA specificity for infants with a full numerical complement of NIMA specificity. NIMA-specific Tregs expand during adaptive immune components at the time of birth (Mold and pregnancies sired by males expressing alloantigens McCune, 2012; Mold et al., 2008). In this scenario, postnatal with overlapping NIMA specificity, thereby averting persistence of NIMA-specific tolerance represents an expend- fetal wastage triggered by prenatal infection and able developmental remnant of immune suppressive mecha- non-infectious disruptions of fetal tolerance. There- nisms essential for in utero survival. However, this reasoning fore, exposure to NIMA selectively enhances repro- does not explain why tolerance imprinted by exposure to foreign ductive success in second-generation females car- antigens in utero is widely conserved across mammalian species rying embryos with overlapping paternally inherited (e.g., non-human primates, ruminants, rodents) regardless of fetal adaptive immune cell maturation relative to parturition (Bill- antigens. These findings demonstrate that genetic ingham et al., 1953; Burlingham et al., 1998; Dutta and Burling- fitness, canonically thought to be restricted to Men- ham, 2011; Owen, 1945; Picus et al., 1985). For example, delian inheritance, is enhanced in female placental prolonged survival of NIMA-matched allografts in humans is mammals through vertically transferred maternal consistently reproduced in mice despite the absence of periph- cells that promote conservation of NIMA and enforce eral T cells at the time of birth in this species (Akiyama et al., cross-generational reproductive benefits. 2011; Andrassy et al., 2003; Araki et al., 2010; Matsuoka et al., 2006). These results illustrating highly engrained phylogenetic roots of NIMA tolerance in mammalian reproduction strongly INTRODUCTION suggest the existence of universal biological benefits driving conserved tolerance to NIMA that persists through adulthood. Reproductive health and pregnancy outcomes have traditionally Given the necessity for sustained maternal tolerance to foreign been characterized from the viewpoint of maternal tolerance to fetal antigens in successful pregnancies across all eutherian immunologically foreign paternal antigens expressed by the placental mammals (Samstein et al., 2012), postnatal NIMA-spe- fetus (Erlebacher, 2013; Munoz-Suano et al., 2011). However, cific tolerance may be evolutionarily preserved to promote repro- compulsory fetal exposure to an equally diverse array of discor- ductive fitness by reinforcing fetal tolerance in future generation dant non-inherited maternal antigens (NIMA) also occurs during pregnancies. To address this hypothesis, immunological tools in utero and early postnatal maturation. Maternal antigen stimu- that allow precise identification of T cells with NIMA-specificity lation in these developmental contexts imprints remarkably were uniquely combined with mouse models of allogeneic preg- persistent tolerance to genetically foreign NIMA in offspring nancy, and pregnancy complications stemming from disruptions (Dutta et al., 2009; Hirayama et al., 2012; Mold and McCune, in fetal tolerance (Chaturvedi et al., 2015; Rowe et al., 2011; 2012). Pioneering examples of tolerance to NIMA include Rowe et al., 2012b). Our data show obligatory developmental blunted sensitization to erythrocyte rhesus (Rh) antigen among exposure to foreign maternal tissue primes expanded accumula- Rh-negative women born to Rh-positive mothers (Owen et al., tion of NIMA-specific immune suppressive regulatory CD4+

Cell 162, 1–11, July 30, 2015 ª2015 Elsevier Inc. 1 Please cite this article in press as: Kinder et al., Cross-Generational Reproductive Fitness Enforced by Microchimeric Maternal Cells, Cell (2015), http://dx.doi.org/10.1016/j.cell.2015.07.006

T cells (Tregs) that reinforce fetal tolerance during next-genera- compared with each group of control mice (Thornton et al., 2010) tion pregnancies sired by males with overlapping MHC haplo- (Figure 1C), these results strongly suggest developmental type specificity. Expanded NIMA-specific Treg accumulation exposure to immunologically discordant maternal tissue primes requires ongoing postnatal cognate antigen stimulation by induced FOXP3 expression among NIMA-specific CD4+ T cells. maternal cells that establish microchimerism in offspring. In the Importantly, these shifts were restricted to CD4+ T cells with broader context, cross-generational reproductive benefits NIMA-specificity since expanded Tregs and diminished Helios conferred by tolerance to NIMA indicate genetic fitness is not expression were eliminated among bulk CD4+ T cells in each restricted only to transmitting homologous chromosomes by group of mice regardless of developmental 2W1S stimulation Mendelian inheritance but is enhanced through vertically trans- (Figures 1A and 1C). ferred tolerogenic cells that establish microchimerism in Considering exposure to maternal tissue begins in utero when offspring favoring preservation of non-inherited maternal alleles fetal immune components are undergoing maturation, related within a population. experiments addressed whether expanded NIMA-specific Tregs require antigen presentation by maternal cells. Here, the I-Ab

RESULTS restricted nature of 2W1S55-68 peptide presentation was exploited to compare NIMA-specific Tregs in genetically identical Developmental Exposure to Maternal Tissue Drives NIMA-2W1S offspring born to I-Ab/b or I-Ad/d mothers (Rees Expanded NIMA-Specific Regulatory T Cell et al., 1999)(Figure 1D). Interestingly, expanded proportions of Accumulation FOXP3+ Tregs and diminished cell-intrinsic Helios expression b To investigate the fundamental biology driving conserved persis- were each significantly reduced for I-A :2W1S55-68-specific tence of NIMA tolerance across mammalian species, an instruc- CD4+ T cells in NIMA-2W1S offspring born to I-Ad/d mothers to tive allogeneic mating strategy that transforms defined model levels comparable to naive control mice (Figure 1D). The results antigens into surrogate NIMA was developed to precisely track highlighting tolerogenic properties of maternal cells in mouse T cells with NIMA specificity. Female mice heterozygous for a offspring parallel the presence of maternal hematopoietic cells transgene encoding constitutive expression of a transmembrane in human fetal lymph nodes during in utero development (Mold recombinant protein containing ovalbumin (OVA) and the et al., 2008) and suggest compulsory developmental exposure

2W1S55-68 variant of I-Ea in all cells (behind the b-actin promoter) to genetically foreign maternal tissue actively primes expansion (Moon et al., 2011; Rees et al., 1999) were mated with non-trans- of immune suppressive Tregs with NIMA specificity. genic males—thereby transforming 2W1S55-68 and OVA into sur- Expanded NIMA-specific Treg accumulation may reflect in rogate NIMA in half the offspring (Figures S1A and S1B). This utero and/or postnatal exposure to foreign maternal antigen approach, taking advantage of MHC tetramer staining and (e.g., soluble maternal HLA alloantigens, intact maternal cells enrichment techniques for identifying endogenous CD4+ T cells in breast milk) (Molitor et al., 2004; Zhou et al., 2000). To disso- b with I-A :2W1S55-68 specificity (Moon et al., 2007), combined ciate the contributions of maternal antigen stimulation during with tools for manipulating OVA-expressing cells allows NIMA- each developmental context, the individual impacts of in utero responsive and NIMA-expressing cells to be simultaneously and early postnatal NIMA exposure through breastfeeding b + evaluated. To ensure shifts in I-A :2W1S55-68-specific CD4 were evaluated by cross-fostering offspring after birth with naive T cells reflect developmental exposure to maternal tissue as or 2W1S-OVA+ nursing mothers. In agreement with improved opposed to 2W1S-OVA+ concepti within the same litter, survival of NIMA-matched allografts in transplant recipients offspring from reciprocal mating between males heterozygous exposed to maternal antigen both in utero and through breast- for the 2W1S-OVA expression transgene and non-transgenic feeding (Andrassy et al., 2003; Campbell et al., 1984), maximal females that transform 2W1S55-68 peptide and OVA into NIMA-specific Treg expansion required in utero plus postnatal surrogate non-inherited paternal antigens (NIPA) were used as maternal antigen stimulation (Figure 1E). Comparatively, Helios controls along with genetically identical naive mice without expression remained at diminished levels with maternal tissue developmental 2W1S exposure (Figures S1C and S1D). exposure in utero or through breastfeeding suggesting maternal Sharply increased proportions of immune suppressive regula- antigen stimulation in either developmental context primes en- tory T cells (Tregs) identified by expression of the lineage defining riched proportions of NIMA-specific CD4+ T cells poised for FOXP3 transcriptional regulator (Fontenot et al., 2003; Hori et al., induced FOXP3 expression (Figure 1E). Taken together, these + b 2003) were found among CD4 T cells with I-A :2W1S55-68 spec- findings indicate immunologically foreign maternal antigen stim- ificity in adult NIMA-2W1S mice compared with age-matched ulation in utero and through breastfeeding work synergistically to naive mice, and additional control mice exposed to the identical promote expanded peripheral accumulation of NIMA-specific 2W1S-OVA recombinant protein as a surrogate NIPA or ubiqui- Tregs. tous ‘‘self’’ antigen (Figures 1A and S1). Interestingly, while the + b percentage and number of FOXP3 Tregs with I-A :2W1S55-68 Microchimeric Maternal Cells Maintain Expanded specificity were significantly increased in the spleen and periph- NIMA-Specific Tregs eral lymph nodes for NIMA-2W1S offspring, the total number of Persistence of NIMA-specific tolerance coincides with postnatal b + I-A :2W1S55-68-specific CD4 T cells remained similar regard- retention of microchimeric maternal cells in adult human and ro- less of developmental 2W1S stimulation (Figure 1B). Along with dent offspring (Dutta et al., 2009; Loubie` re et al., 2006; Maloney sharply reduced expression of Helios that marks thymus derived et al., 1999; Mold et al., 2008). Nonetheless, the immunological b Tregs among cells with I-A :2W1S55-68 specificity in NIMA-2W1S cause and effect relationship between these two interrelated

2 Cell 162, 1–11, July 30, 2015 ª2015 Elsevier Inc. Please cite this article in press as: Kinder et al., Cross-Generational Reproductive Fitness Enforced by Microchimeric Maternal Cells, Cell (2015), http://dx.doi.org/10.1016/j.cell.2015.07.006

A Naive NIMA NIPA Self B C Naive NIMA NIPA Self 200 77 38 55 70 *** ***

CD44 20 I-Ab:2W1S Helios :2W1S Tregs :2W1S Tregs per mouse 13 4646 22 20 b I-A 2 100 *** ** A PA Naive NIM NI Self 75 FOXP3 cells among +

CD25 hi 200 50

80 *** *** + 25 60 FOXP3 % Helios among + 20 0 T cells T + 40 I-Ab:2W1S + + + + :2W1S CD4 b

CD4 Naive NIMA NIPA Self

20 I-A T cells per mouse T

% FOXP3 2 0 A PA I-Ab:2W1S + + + + Naive NIM NI Self Naive NIMA NIPA Self E : : b D b/b b 80 80 Maternal I-A *** 60 60 T cells T T cells T + 2W1S + +/- among I-A d/dd/d OVA b/b among I-A 40 40

X + I-AA I-A + (H-22d/dd/d) (H-2b/b) 20 20 I-Ab:2W1S presentation by 2W1S CD4 maternal and fetal cells 0 2W1S CD4 0 % FOXP3 % FOXP3 100 100 Maternal I-Ad/d *** 75 75 among among hi

2W1S hi +/- 50 50 I-Ab/b X OVA I-Ad/d b/b d/d :2W1S Tregs :2W1S Tregs (H-2 ) (H-2 ) :2W1S Tregs b b 25 25 I-A I-A % Helios I-Ab:2W1S presentation % Helios exclusively by fetal cells 0 0 Maternal MHC I-Ab I-Ad I-Ab I-Ad In utero ++ haplotype Breastfeeding + + NIMA Naive

Figure 1. Developmental Exposure to Maternal Tissue Primes Expanded NIMA-Specific FOXP3+ Tregs (A) Representative plots showing the gating strategy used to identify I-Ab:2W1S specific among CD4+ T cells (top), FOXP3+ Tregs among I-Ab:2W1S-specific CD4+ T cells (middle), and composite data (bottom) for percent FOXP3+ among CD4+ T cells with I-Ab:2W1S specificity (filled) compared with bulk CD4+ T cells (open) in the spleen plus peripheral lymph nodes of naive (blue), NIMA-2W1S (red), NIPA-2W1S (green), or 2W1S-self (gray) 8-week-old adult mice. (B) Total number of I-Ab:2W1S-specific FOXP3+ Tregs (top) and CD4+ T cells (bottom) for each group of mice described in (A). (C) Percent Helioshi among I-Ab:2W1S-specific (red line) or bulk (gray shaded) FOXP3+ CD4+ T cells for each group of mice described in (A). b/b b (D) Mating strategy for generating genetically identical NIMA-2W1S offspring born to either MHC class II I-A (I-A :2W1S55-68 peptide presented by cells of both d/d b + maternal and fetal origin) or I-A (I-A :2W1S55-68 peptide only presented by cells of fetal origin) haplotype mothers, and composite data for percent FOXP3 among I-Ab:2W1S-specific CD4+ T cells and Helioshi among I-Ab:2W1S-specific FOXP3+ cells for each group of NIMA-2W1S (red) compared with naive (blue) mice. (E) Percent FOXP3+ among I-Ab:2W1S-specific CD4+ T cells, and Helioshi among I-Ab:2W1S-specific FOXP3+ cells for each group of cross-fostered offspring exposed to 2W1S-OVA in utero and/or postnatally through breastfeeding by 2W1S-OVA+ mothers. Each point represents the result from an individual female mouse, and these data are representative of at least three separate experiments each with similar results. Bars, mean ± 95% confidence interval. **p < 0.01, ***p < 0.001. See also Figure S1. phenomena engrained in mammalian reproduction remain unde- required for maintaining expanded Tregs with NIMA-specificity. fined. One possibility is that postnatal maintenance of NIMA- Having established early developmental exposure to 2W1S- specific tolerance actively prevents rejection of antigenically OVA+ maternal tissues imprints persistent accumulation of discordant maternal cells fostering their long-term survival in NIMA-2W1S-specific Tregs, analysis of cells expressing these offspring. Alternatively, retained microchimeric maternal cells model antigens was extended to investigate the necessity for may provide an essential postnatal source of cognate antigen postnatal stimulation by microchimeric 2W1S-OVA+ maternal

Cell 162, 1–11, July 30, 2015 ª2015 Elsevier Inc. 3 Please cite this article in press as: Kinder et al., Cross-Generational Reproductive Fitness Enforced by Microchimeric Maternal Cells, Cell (2015), http://dx.doi.org/10.1016/j.cell.2015.07.006

A B Figure 2. NIMA-Specific Treg Expansion 102 Heart 102 Liver Requires Persistent Postnatal Exposure to 2W1S-OVA+ Naive Microchimeric Maternal Cells 1 1 + cells 10 * cells 10 ** (A) Maternal 2W1S-OVA microchimeric cell en- 5 5 coding DNA levels in each tissue of naive (blue 0 0 10 10 filled), NIMA-2W1S (red filled), or NIMA-2W1S mice 10-1 10-1 treated with anti-OVA depleting antibody (red open). Est. Geq/10 L.O.D. Est. Geq/10 L.O.D. -2 -2 -OVA (B) Cell surface OVA expression levels among 10 10 antibody -OVA + -OVA + splenocytes from 2W1S-OVA+ compared with naive control mice after staining with anti-OVA (red Naive NIMA Naive NIMA line) or rabbit IgG isotype (gray shaded) antibodies. (C) Representative plots and composite data for C NIMA + b + Naive NIMA FOXP3 Tregs among I-A :2W1S-specific CD4 -OVA b

: T cells, and Helios expression among I-A :2W1S- 19 60 20 b 80 100 + *** *** speciﬁc FOXP3 cells for each group of mice described in (A). Each point represents the result 60 75

T cells T from an individual female mouse at 8 weeks of age, + among hi FOXP3

among I-A and these data are representative of at least three

+ 40 50 CD25 separate experiments each with similar results. :2W1S Tregs :2W1S Tregs 68 48 70 20 b 25 Bars, mean ± 95% conﬁdence interval. *p < 0.05, I-A % Helios

2W1S CD4 **p < 0.01, ***p < 0.001. L.O.D., limit of detection.

% FOXP3 0 0 See also Figure S2. -OVA + -OVA + Naive NIMA Naive NIMA Helios cells. We found OVA-encoding DNA, reflective of 2W1S-OVA+ notion is supported by highly enriched microchimeric 2W1S- maternal cells in systemic organs (e.g., heart, liver), at levels OVA+ maternal cells in female reproductive tissue (uterus) of ranging from 1 in 105 to 106 cells in NIMA offspring consistent NIMA offspring, and their conspicuous absence in analogous with quantities of microchimeric maternal cells identified using male reproductive tissue (prostate) (Figure 3A). In turn, NIMA- PCR for MHC haplotype alleles (Bakkour et al., 2014; Dutta specific Treg expansion and reduced Helios expression were et al., 2009; Mold et al., 2008)(Figures 2A and S2A). markedly more pronounced in female compared with male To definitively address the cause and effect relationship be- NIMA-2W1S littermate offspring (Figure 3B). Thus, gender- tween NIMA-specific tolerance and microchimeric maternal specific differences favoring more robust NIMA-specific Treg cells, anti-OVA antibody that uniformly binds 2W1S-OVA+ cells expansion in females parallel the selective accumulation of was used to deplete microchimeric 2W1S-OVA+ maternal cells microchimeric maternal cell in female reproductive tissue. (Figure 2B). In line with the efficiency whereby anti-OVA antibody To further investigate the reproductive significance for gender- depletes congenically marked 2W1S-OVA+ cells after adoptive specific differences in postnatal persistence of NIMA-specific transfer into non-transgenic recipients (Figure S2B), OVA encod- tolerance, shifts in NIMA-specific CD4+ T cells were evaluated ing DNA representative of microchimeric 2W1S-OVA+ cells in in females during pregnancy after cognate fetal antigen stimula- each organ of NIMA offspring declined sharply within 12 days tion. During allogeneic pregnancies sired by 2W1S-OVA+ trans- following in vivo anti-OVA antibody administration (Figure 2A). genic males, sharply accelerated expansion tempo occurred Remarkably, NIMA-2W1S-specific Treg accumulation and cell- among 2W1S-specific Tregs in NIMA-2W1S female mice intrinsic Helios downregulation both returned to background compared with naive control mice (7.2-fold compared with levels found in naive control mice after elimination of microchi- 3.4-fold expansion by midgestation in NIMA and naive mice, meric 2W1S-OVA+ cells within this time frame (Figure 2C). respectively [p = 0.004]) (Figure 4). Accelerated NIMA-specific Thus, microchimeric maternal cells provide an essential source Treg expansion tempo during pregnancy represents a targeted of cognate maternal antigen required for sustaining postnatal response to cognate 2W1S stimulation by shared fetal- NIMA-specific tolerance. expressed antigen because NIMA-2W1S-specific Tregs did not expand during pregnancies sired by non-transgenic male mice Selectively Enriched NIMA-Specific Treg Expansion in (Figure 4). Thus, mammalian females contain an enriched pool Female Offspring Accentuated during Pregnancy with of NIMA-specific Tregs poised for accelerated re-expansion NIMA-Matched Fetal Antigen Stimulation upon encounter with paternal-fetal antigen of overlapping spec- Given the necessity for expanded maternal tolerance that en- ificity during pregnancy. compasses immunologically foreign paternal-fetal antigens in successful pregnancy shared by all eutherian placental mam- Microchimeric Maternal Cells Enforce Cross- mals (Erlebacher, 2013; Munoz-Suano et al., 2011; Samstein Generational Protection against Fetal Wastage et al., 2012), we reasoned reinforced fetal tolerance that pro- Since the immunological identity of individuals is primarily motes reproductive fitness may represent a more universal defined by unique expression of MHC haplotype alleles (Zinker- evolutionary driver for conserved NIMA-specific tolerance. This nagel and Doherty, 1979), MHC haplotype alleles (e.g., H-2d,

4 Cell 162, 1–11, July 30, 2015 ª2015 Elsevier Inc. Please cite this article in press as: Kinder et al., Cross-Generational Reproductive Fitness Enforced by Microchimeric Maternal Cells, Cell (2015), http://dx.doi.org/10.1016/j.cell.2015.07.006

A Uterus/ Figure 3. Expanded NIMA-Speciﬁc Treg Accumula- Prostate Heart Liver tion in Female Offspring Parallels Discordant 102 102 102 Maternal Cell Microchimerism in Gender-Speciﬁc NIMANNIMA Reproductive Tissue 1 1 NIMA * 1 NaiveNNaive + cells 10 10 10

5 (A) 2W1S-OVA encoding DNA levels in each tissue among NIMANIMA NIMA-2W1S female (red circle), littermate 2W1S-NIMA male 100 100 100 NaiveNaive (red triangle), naive female (blue circle) and naive male (blue 10-1 10-1 10-1 triangle) mice.

Est. Geq/10 (B) Representative plots and composite data showing L.O.D. 10-2 10-2 10-2 I-Ab:2W1S-speciﬁc CD4+ T cells (top), FOXP3+ Tregs among I-Ab:2W1S-speciﬁc CD4+ T cells (middle), and Helios b

: expression among Tregs with I-A :2W1S speciﬁcity (red line) b B NIMA NIMA Naive Naive 80 ** or bulk speciﬁcity (gray shaded) among NIMA-2W1S female compared with NIMA-2W1S littermate male mice. Each point 60

T cells T represents the result from an individual mouse at 8 weeks of + age, and these data are representative of at least three among I-A

+ 40 separate experiments each with similar results. Bars, mean ± CD44 95% conﬁdence interval. *p < 0.05, **p < 0.01. L.O.D., limit of b 20 I-A :2W1S detection. 2W1S CD4

51 30 14 18 % FOXP3 0 100 ** 75 FOXP3 among CD25 hi 50 and fetal-expressed OVA antigen was avoided by exclusively using non-transgenic H-2d Balb/c :2W1S Tregs :2W1S Tregs 26 58 58 61 b 25 male mice to sire allogeneic pregnancy in H-2d- I-A % Helios 2W1S-OVA NIMA female mice (Figure S3). This 0 analysis showed depletion of microchimeric maternal cells prior to mating efficiently overturns protection against fetal resorption and in utero L. monocytogenes invasion in NIMA female mice (Figure 5A). Importantly, these reproductive bene- H-2k—along with 2W1S and OVA antigens), were transformed fits were not restricted to NIMA H-2d haplotype alleles since fetal into surrogate NIMA to investigate functional properties of resorption and in utero L. monocytogenes invasion were each tolerance in the setting of broader NIMA overlap (Figure S3). In similarly averted among NIMA H-2k female mice during alloge- turn, the protective properties of NIMA-specific tolerance were neic pregnancy sired by H-2k CBA/J male mice (Figure 5B). probed by infection with the prenatal bacterial pathogen, Listeria Conversely, protection against fetal wastage was lost in NIMA monocytogenes, which disrupts fetal tolerance with ensuing H-2k female mice if NIMA mismatched H-2d males were used fetal wastage (Chaturvedi et al., 2015; Mylonakis et al., 2002; to sire allogeneic pregnancy, or if H-2k-2W1S-OVA+ microchi- Rowe et al., 2012a). We found fetal resorption and in utero meric maternal cells were depleted using anti-OVA antibody L. monocytogenes invasion after prenatal infection in naive prior to mating with non-transgenic H-2k CBA/J male mice (Fig- mice bearing allogeneic pregnancy were eliminated by overlap ure 5B). Thus, persistent postnatal tolerance to NIMA protects between NIMA and paternal-fetal MHC haplotype antigens against fetal wastage triggered by prenatal infection. (NIMA H-2d females mated with H-2d Balb/c male mice) (Figures To further extend this analysis to non-infectious disruptions 5A and S3). Protection against fetal wastage occurred in an in fetal tolerance stemming from blunted expansion of maternal antigen-specific fashion requiring commonality between NIMA Tregs (e.g., spontaneous abortion, preeclampsia) (Jiang et al., and paternal-fetal antigens since fetal resorption and in utero 2014; Santner-Nanan et al., 2009; Sasaki et al., 2004), the pro- bacterial invasion each rebounded when third-party males tective benefits of NIMA-specific tolerance on fetal wastage trig- bearing irrelevant MHC haplotype alleles (e.g., H-2k CBA/J gered by partial depletion of bulk maternal FOXP3+ Tregs during mice) were used to sire allogeneic pregnancy in NIMA H-2d allogeneic pregnancy were investigated. Diphtheria toxin admin- female mice (Figures 5A and S3). Thus, protection against pre- istration to female mice heterozygous for co-expression of the natal infection conferred by non-inherited antigenic overlap high-affinity human diphtheria toxin receptor (DTR) with FOXP3 between maternal grandmother and the developing fetus high- during pregnancy causes partial transient depletion of bulk lights profound cross-generational benefits of persistent NIMA- maternal Tregs to levels comparable to virgin control mice with specific tolerance. disruptions in fetal tolerance and ensuing fetal wastage (Kim Given the requirement for postnatal maternal microchimerism et al., 2007; Rowe et al., 2011; Rowe et al., 2012b). Therefore, to maintain expanded NIMA-specific tolerance (Figure 2), we our breeding strategy was modified to transform MHC haplotype next addressed the necessity for microchimeric maternal cells alleles (e.g., H-2d, H-2k) along with 2W1S-OVA into surrogate to protect against fetal wastage after prenatal L. monocytogenes NIMA in genetically identical H-2b FOXP3DTR/WT female mice infection. Here, cross-reactivity between anti-OVA antibody (Figure S4). We found fetal resorption triggered by partial deple- used to deplete H-2d-2W1S-OVA+ microchimeric maternal cells tion of bulk maternal Tregs during allogeneic pregnancy in naive

Cell 162, 1–11, July 30, 2015 ª2015 Elsevier Inc. 5 Please cite this article in press as: Kinder et al., Cross-Generational Reproductive Fitness Enforced by Microchimeric Maternal Cells, Cell (2015), http://dx.doi.org/10.1016/j.cell.2015.07.006

NIMA-2W1S + + + persistent postnatal NIMA-speciﬁc tolerance encompasses Pregnant + + + both infectious and non-infectious perturbations in fetal toler- 2W1S+ + + ance during next-generation pregnancies. DISCUSSION

Reproductive success in female placental mammals requires CD44 sustained maternal tolerance to immunologically foreign I-Ab:2W1S paternal antigens expressed by the developing fetus (Erle- 14 19 56 80 21 bacher, 2013; Munoz-Suano et al., 2011; Samstein et al., 2012). Reciprocally, disruptions in fetal tolerance are increasingly recognized in spontaneous abortion, preeclampsia, and prematurity that occur commonly in human pregnancy (Jiang FOXP3 CD25 et al., 2014; Santner-Nanan et al., 2009; Wilcox et al., 1988; Blen- cowe et al., 2012; Duley, 2009). Given this sustained backdrop of

: *** b 80 reﬁning selection that likely occurs across all outbred mammalian species, conservation of phenotypic traits that improve

T cells T 60 reproductive ﬁtness is a biological imperative. +

among I-A Herein, this reasoning was applied to investigate the ontolog- + 40 ical conservation of NIMA-speciﬁc tolerance and maternal cell 20 microchimerism across placental mammalian species (Andrassy

2W1S CD4 et al., 2003; Bakkour et al., 2014; Dutta and Burlingham, 2011; % FOXP3 0 Gammill et al., 2015). Using mice with deﬁned MHC haplotype al- 103 7.2x leles in multi-generational breeding that transforms MHC haplotype alleles into surrogate NIMA, we show sharply increased 3.4x 102 resiliency against fetal wastage in the presence of overlap between NIMA and paternal-fetal antigen encountered during

1 next-generation pregnancies. Cross-generational reproductive :2W1S Tregs :2W1S Tregs

per mouse 10 b benefits conferred by NIMA-specific tolerance shown here for I-A mice are consistent with pioneering observations of reduced 0 10 erythrocyte Rh antigen sensitization among Rh-negative women + + + NIMA-2W1S born to Rh-positive mothers (Owen et al., 1954). However, while Pregnant + + + developmental exposure to this single minor alloantigen does not prevent hemolytic disease of the newborn (Booth et al., 1953; Owen et al., 1954), we find broader non-inherited antigenic Figure 4. NIMA-Specific Treg Expansion Accelerated during Preg- overlap between maternal grandmother and offspring that en- nancy with NIMA-Matched Fetal Antigen Stimulation compasses MHC haplotype alleles efficiently protects against b + Representative plots and composite data showing I-A :2W1S-specific CD4 fetal wastage (Figure 7). By establishing clear reproductive ben- + b + T cells (top), FOXP3 Tregs among I-A :2W1S-specific CD4 T cells (middle), efits for NIMA-specific tolerance applicable to all placental and composite data for percent and number of FOXP3+ CD4+ T cells with I- Ab:2W1S specificity in virgin and midgestation (E11.5) naive female (blue) mammalian species, these results highlight broad evolutionary compared with NIMA-2W1S (red) female mice after mating with 2W1S-OVA+ advantages for persistent postnatal NIMA-specific tolerance transgenic male mice or non-transgenic controls. Each point represents the beyond averting anti-maternal immunity for human and other result from an individual mouse, these data are representative of at least three species with comparatively more developed fetal adaptive im- separate experiments each with similar results. Bars, mean ± 95% confidence mune components at the time of birth (Mold and McCune, interval. ***p < 0.001. 2012; Mold et al., 2008). Dissecting the mechanistic relationship between NIMA-spe- control female mice was reversed to near completion by overlap cific tolerance and microchimeric maternal cells that both persist between NIMA and fetal expressed MHC haplotype antigens in offspring through adulthood requires strategies for precisely (NIMA H-2d females mated with H-2d Balb/c male mice or identifying NIMA-specific immune components along with NIMA H-2k females mated with H-2k CBA/J male mice) (Figures manipulation of microchimeric maternal cells. Prior limitations 6A and 6B). Protection against fetal wastage induced by partial restricting these analyses were simultaneously bypassed by maternal Treg depletion was paternal-antigen specific and transforming defined model antigens into surrogate NIMA using required ongoing stimulation by microchimeric maternal cells female mice heterozygous for a transgene encoding constitutive since fetal resorption rebounded during pregnancies sired by expression of model antigens for breeding with non-transgenic third-party males that express irrelevant MHC haplotype alleles males (Figure S1). Using antigen-specific tools, endogenous or if 2W1S-OVA+ microchimeric maternal cells were depleted CD4+ T cells with surrogate NIMA specificity were shown to be prior to mating (Figures 6A and 6B). Taken together, these find- highly enriched for expression of the Treg lineage defining tran- ings demonstrate resiliency against fetal wastage conferred by scriptional regulator, FOXP3 (Fontenot et al., 2003; Hori et al.,

6 Cell 162, 1–11, July 30, 2015 ª2015 Elsevier Inc. Please cite this article in press as: Kinder et al., Cross-Generational Reproductive Fitness Enforced by Microchimeric Maternal Cells, Cell (2015), http://dx.doi.org/10.1016/j.cell.2015.07.006

A *** B *** Figure 5. Overlap between NIMA and Paternal-Fetal *** *** Alloantigen Protects against Fetal Resorption and In 100 100 Utero Bacterial Invasion following Prenatal Infection (A) Percent fetal resorption (top) and average recoverable 75 75 bacterial CFUs from each concepti per litter (bottom) ﬁve days following L. monocytogenes intravenous maternal infection 50 50 initiated midgestation (E11.5) for naive female (blue) compared with NIMA-H-2d-(2W1S-OVA) (red) female mice 25 25 during allogeneic pregnancy sired by H-2d or third party H-2k % fetal resorption % fetal resorption males, or depletion of microchimeric 2W1S-OVA+ maternal 0 0 cells with anti-OVA antibody prior to mating. *** *** (B) Percent fetal resorption (top) and average recoverable 7 7 bacterial CFUs from each concepti per litter (bottom) ﬁve days ) )

10 6 6 L. monocytogenes *** 10 *** following intravenous maternal infection 5 5 initiated midgestation (E11.5) for naive female (blue) k 4 4 compared with NIMA-H-2 -(2W1S-OVA) (red) female mice k d 3 3 during allogeneic pregnancy sired by H-2 or third party H-2 males, or depletion of microchimeric 2W1S-OVA+ maternal concepti (log concepti (log Lm CFUs among 2 2 L.O.D. Lm CFUs among L.O.D. cells with anti-OVA antibody prior to mating. Each point rep- 1 1 resents the result from an individual mouse, and these data NIMA H-2d NIMA H-2k + + + + + + are representative of at least three separate experiments each (2W1S-OVA) (2W1S-OVA) with similar results. Bars, mean ± 95% conﬁdence interval. H2d H2d H2k H2d H2k H2k H2d H2k ***p < 0.001. L.O.D., limit of detection. See also Figure S3. -OVA + -OVA + antibody antibody

2003). By establishing NIMA specificity for this essential immune the causative relationship between microchimeric maternal cells regulatory CD4+ T cell subset, these results extend previously and postnatal persistence of NIMA-specific tolerance. Similar to described reversal of NIMA-specific tolerance by depleting of the necessity of low-level exposure to cognate antigen in numer- bulk CD4+ T cells or Tregs (Akiyama et al., 2011; Matsuoka ical maintenance of ‘‘memory’’ effector CD4+ T cells with foreign et al., 2006; Mold et al., 2008; Molitor-Dart et al., 2007). In turn, microbial specificity (Belkaid et al., 2002; Nelson et al., 2013; protection against fetal wastage conferred by expanded Tregs Uzonna et al., 2001), sustained expansion of NIMA-specific with shared NIMA plus fetal specificity also reinforce beneficial FOXP3+ CD4+ T cells also requires postnatal exposure to properties of expanded maternal Tregs with pre-exiting fetal cognate maternal antigen expressed by microchimeric maternal specificity retained after prior pregnancy in partner-specific pro- cells. Reciprocally, in vivo depletion of microchimeric maternal tection against complications in subsequent pregnancy (Camp- cells efficiently overturned both expanded NIMA-specific Treg bell et al., 1985; Rowe et al., 2012b; Trupin et al., 1996). accumulation and protection against fetal wastage. Thus, verti- More importantly, the concurrent ability to deplete maternal cally transferred maternal cells that establish microchimerism cells retained in offspring allowed us to definitively establish in offspring promote cross-generational reproductive fitness by

A B * Figure 6. Overlap between NIMA and Paternal-Fetal ** Antigen Protects against Fetal Wastage Triggered by 100 *** 100 + * *** Partial Depletion of Maternal FOXP3 Regulatory T Cells 75 *** 75 (A) Percent fetal resorption for naive (blue) FOXP3WT/WT and DTR/WT 50 50 FOXP3 female mice compared with each group of NIMA-H-2d FOXP3DTR/WT (red) female mice ﬁve days after 25 25 initiating diphtheria toxin during allogeneic pregnancy sired by

% fetal resorption d k % fetal resorption H-2 or third party H-2 males, or depletion of microchimeric 0 0 2W1S-OVA+ maternal cells with anti-OVA antibody prior to FOXP3DTR/WT ++++ FOXP3DTR/WT ++++ mating. (B) Percent fetal resorption for naive (blue) FOXP3WT/WT and d k NIMA H-2 +++ NIMA H-2 +++ FOXP3DTR/WT female mice compared with each group of (2W1S-OVA) (2W1S-OVA) NIMA-H-2k FOXP3DTR/WT (red) female mice ﬁve days after H2d H2d H2d H2k H2d H2k H2k H2k H2d H2k initiating diphtheria toxin during allogeneic pregnancy sired by H-2k or third party H-2d males, or depletion of micro- -OVA + -OVA + antibody antibody chimeric 2W1S-OVA+ maternal cells with anti-OVA antibody prior to mating. Each point represents the result from an individual mouse, these data are representative of at least three separate experiments each with similar results. Bars, mean ± 95% conﬁdence interval. *p < 0.05, **p < 0.01, ***p < 0.001. See also Figure S4.

Cell 162, 1–11, July 30, 2015 ª2015 Elsevier Inc. 7 Please cite this article in press as: Kinder et al., Cross-Generational Reproductive Fitness Enforced by Microchimeric Maternal Cells, Cell (2015), http://dx.doi.org/10.1016/j.cell.2015.07.006

Traditional Mendelian genetics of immune tolerance from the intriguing perspective of constitutive chimerism beyond engrained pillars of binary ‘‘self’’ versus Two unique MHC haplotypes alleles ‘‘non-self’’ antigen distinction deﬁned using genetically homoge- Mother (red, green), transmitted separately to nous inbred mice that artiﬁcially eliminates cross-generational individual offspring tolerance (Jenkins et al., 2010; Nelson, 2012).

Common EXPERIMENTAL PROCEDURES susceptibility to Female fetal wastage Mice offspring regardless of C57BL/6 (H-2b; CD45.2+), Balb/c (H-2d), CBA/J (H-2k), B6.C-H2d/bByJ paternal-fetal (H-2d), B6.Ak-H2k/J (H-2k), and B6.SJL-PtprcaPepcb/BoyJ (H-2b; CD45.1+) MHC haplotype mice were purchased from The Jackson Laboratory. 2W1S-OVA+ transgenic

mice that constitutively express recombinant 2W1S55-68-OVA protein behind the b-actin promoter, and FOXP3DTR/DTR mice where FOXP3+ cells are sus- Cross-generational reproductive fitness (enforced ceptible to diphtheria toxin induced ablation have each been described (Kim by tolerance to non-inherited maternal antigen) et al., 2007; Moon et al., 2011). 2W1S-OVA+ mice were maintained on either the C57BL/6 or Balb/c strain backgrounds after backcrossing for >10 generations. For cross-fostering, pregnant mice were checked twice daily for Mother birth timing, and newborn offspring introduced to lactating foster mothers Increased resiliency within 12 hr after birth, with weaning 21 days thereafter and analysis at against fetal wastage in 8 weeks of age. For partial transient maternal Treg depletion, FOXP3DTR/WT pregnancies sired by pregnant females were administered purified diphtheria toxin daily (Sigma- males with shared NIMA Aldrich, USA) (0.5 mg first dose, followed by 0.1 mg/dose) beginning midges- specificity Female tation (E11.5) for 5 consecutive days, and the frequency of fetal resorption offspring Background evaluated E16.5. All experiments were performed using sex and age- susceptibility to matched controls under Cincinnati Children’s Hospital IACUC approved fetal wastage in protocols. pregnancies sired by males without Tetramer Enrichment and Flow Cytometry Microchimeric shared NIMA Cell surface staining with phycoerythrin (PE)-conjugated MHC class II specificity b maternal cells I-A :2W1S55-68 tetramer followed by enrichment using anti-PE-conjugated magnetic beads (Miltenyi Biotec) have been described (Moon et al., 2007; Moon et al., 2009; Rowe et al., 2012b). To identify CD4+ T cells with Figure 7. Cross-Generational Reproductive Fitness Enforced by b I-A :2W1S specificity, cells in secondary lymphoid tissue (spleen plus axillary, Vertically Transferred Microchimeric Maternal Cells in Eutherian brachial, cervical, inguinal, mesenteric, pancreatic, para-aortic/uterine lymph Placental Mammals nodes) of each mouse were combined, enriched with PE conjugated In traditional Mendelian genetics (top), pregnancies among female offspring b I-A :2W1S55-68 tetramer, and stained for cell-surface CD4 (GK1.5), CD8a are equally susceptible to fetal wastage or other complications stemming from (53-7.3), CD25 (PC61), CD44 (IM7), CD11b (M1/70), CD11c (N418), B220 disruptions in fetal tolerance regardless of paternal MHC haplotype specificity. (RA3-B62), F4/80 (BM8), along with intranuclear FOXP3 (FJK-16 s) or Comparatively, persistent postnatal maintenance of tolerogenic micro- Helios (22F6) expression using commercially available antibodies and cell chimeric maternal cells in female offspring promotes cross-generational permeabilization reagents (BD PharMingen or eBioscience). For cell surface reproductive fitness (bottom) by selectively protecting against fetal wastage ovalbumin expression, cells were stained initially with polyclonal rabbit during next-generation pregnancies sired by males with shared overlapping a-OVA (EMP Millipore) or IgG isotype antibodies followed by secondary stain- NIMA specificity. ing with PE conjugated anti-rabbit IgG (eBioscience) antibody. Cells stained with fluorochrome-conjugated tetramer and/or antibody were acquired using preserving tolerance to NIMA along with non-inherited genetic a FACSCanto cytometer (Becton Dickinson) and analyzed using FlowJo (TreeStar) software. alleles within a population (Figure 7). In the broader context, these results indicate genetic fitness, Bacteria canonically thought to be restricted to transmitting only half For infection, Listeria monocytogenes (wild-type strain 10403s) was grown of homologous chromosomes through Mendelian inheritance, is to early log phase (OD600 0.1) in brain heart infusion media at 37 C, washed, enhanced in female placental mammals to also promote conser- and diluted with sterile saline, and inoculated intravenously via the lateral 4 vation of non-inherited antigens by vertical transmission of tail vein (10 CFUs) at midgestation (E11.5) as described (Chaturvedi tolerogenic maternal cells that establish microchimerism in et al., 2015; Rowe et al., 2011). The inoculum for each experiment was confirmed by spreading diluted aliquots onto agar plates. Five days thereafter, offspring. However, in nature, this engrained drive for genetic fetal resorption and in utero bacteria invasion was evaluated by sterilely dis- fitness in each individual is likely counterbalanced by pathogen- secting each concepti, homogenization in sterile saline containing 0.05% mediated selection for MHC diversity across the entire population Triton X-100 to release intracellular bacteria, plating serial dilutions of each (Spurgin and Richardson, 2010). Nonetheless, our findings sug- concepti homogenate onto agar plates, and enumeration after incubation at gest more extended cross-generational analysis will illuminate 37 C for 24 hr. the ongoing controversy regarding how MHC haplotype similarity DNA Extraction and Quantitative PCR impacts mate selection and pregnancy outcomes (Chaix et al., The heart, liver, uterus, or prostate was sterilely dissected, and DNA extracted 2008; Israeli et al., 2014; Ober et al., 1997). Finally, reproductive from each tissue using the QIAamp DNA extraction kit (QIAGEN). Thereafter, advantages actively maintained by tolerogenic microchimeric PCR for enumerating 2W1S-OVA+ DNA was performed in 20 separate wells maternal cells underscore the need for renewed consideration per tissue each containing 333 ng genomic DNA (3.33 3 105 cells) in 20 ml

8 Cell 162, 1–11, July 30, 2015 ª2015 Elsevier Inc. Please cite this article in press as: Kinder et al., Cross-Generational Reproductive Fitness Enforced by Microchimeric Maternal Cells, Cell (2015), http://dx.doi.org/10.1016/j.cell.2015.07.006

total volume supplemented with 10 ml Taqman Gene Expression Master Mix lingham, W.J. (2003). Tolerance to noninherited maternal MHC antigens in and 1 ml ovalbumin Taqman assay (Applied Biosystems) for a detection limit mice. J. Immunol. 171, 5554–5561. 3 6 of 1 in 6.66 10 cells per tissue. Amplification was performed using the Araki, M., Hirayama, M., Azuma, E., Kumamoto, T., Iwamoto, S., Toyoda, H., 7500 Fast Real-Time PCR System (Life Technologies) under the following pro- Ito, M., Amano, K., and Komada, Y. (2010). Prediction of reactivity to nonin- gram: 95 C for 10 min, followed by 40 cycles of 95 C for 15 s and 60 C for herited maternal antigen in MHC-mismatched, minor histocompatibility anti- + 1 min. For generating standard curve for 2W1S-OVA DNA, DNA from gen-matched stem cell transplantation in a mouse model. J. Immunol. 185, + 2W1S-OVA splenocytes or C57BL/6 control mice were isolated, and com- 7739–7745. bined with six serial 10-fold dilutions (10À1 to 10À6) of 2W1S-OVA+ DNA into Bakkour, S., Baker, C.A., Tarantal, A.F., Wen, L., Busch, M.P., Lee, T.H., and C57BL/6 control DNA so that the DNA concentration remained identical in McCune, J.M. (2014). Analysis of maternal microchimerism in rhesus monkeys each well (333 ng total DNA in 20 ml). The resulting linear regression equation (Macaca mulatta) using real-time quantitative PCR amplification of MHC poly- y=À1.137ln(x) + 38.443 (R2 = 0.986) was used to calculate the amount of morphisms. Chimerism 5, 6–15. 2W1S-OVA+ DNA in each tissue sample. Belkaid, Y., Piccirillo, C.A., Mendez, S., Shevach, E.M., and Sacks, D.L. (2002). Depletion of Microchimeric 2W1S-OVA+ Maternal Cells CD4+CD25+ regulatory T cells control Leishmania major persistence and immunity. Nature 420, 502–507. To deplete 2W1S-OVA+ cells, 2W1S-NIMA mice were administered 650 mg purified rabbit a-OVA antibody (EMP Millipore) or IgG isotype antibody Billingham, R.E., Brent, L., and Medawar, P.B. (1953). Actively acquired toler- 172 (Sigma-Aldrich) by intraperitoneal injection, followed 10 days later by a second ance of foreign cells. Nature , 603–606. treatment with 325 mg of the same antibody. Two days after the second anti- Blencowe, H., Cousens, S., Oestergaard, M.Z., Chou, D., Moller, A.B., Narwal, body inoculation, the level of 2W1S-OVA+ cells in each tissue was analyzed R., Adler, A., Vera Garcia, C., Rohde, S., Say, L., and Lawn, J.E. (2012). by quantitative real-time PCR, antigen-specific CD4+ T cells investigated using National, regional, and worldwide estimates of preterm birth rates in the year b d k I-A :2W1S55-68 tetramer staining or used for mating with H-2 Balb/c or H-2 2010 with time trends since 1990 for selected countries: a systematic analysis CBA/J males to investigate pregnancy outcomes. and implications. Lancet 379, 2162–2172. Booth, P.B., Dunsford, I., Grant, J., and Murray, S. (1953). Haemolytic disease Statistical Analysis in first-born infants. Br. Med. J 2, 41–42. Where applicable, NIMA mice in each group were randomized for either Burlingham, W.J., Grailer, A.P., Heisey, D.M., Claas, F.H., Norman, D., Moha- administration of anti-OVA or isotype antibody, or for breeding with either nakumar, T., Brennan, D.C., de Fijter, H., van Gelder, T., Pirsch, J.D., et al. NIMA-matched or NIMA-discordant MHC haplotype males. Considering (1998). The effect of tolerance to noninherited maternal HLA antigens on the data sets did not consistently show a normal distribution, differences between survival of renal transplants from sibling donors. N. Engl. J. Med. 339, 1657– groups were analyzed using the Mann-Whitney non-parametric test (Prism, 1664. GraphPad); and p < 0.05 was taken as statistical significance. Campbell, D.A., Jr., Lorber, M.I., Sweeton, J.C., Turcotte, J.G., Niederhuber, J.E., and Beer, A.E. (1984). Breast feeding and maternal-donor renal allo- SUPPLEMENTAL INFORMATION grafts. Possibly the original donor-specific transfusion. Transplantation 37, 340–344. Supplemental Information includes four figures and can be found with this Campbell, D.M., MacGillivray, I., and Carr-Hill, R. (1985). Pre-eclampsia in sec- article online at http://dx.doi.org/10.1016/j.cell.2015.07.006. ond pregnancy. Br. J. Obstet. Gynaecol. 92, 131–140. Chaix, R., Cao, C., and Donnelly, P. (2008). Is mate choice in humans MHC- AUTHOR CONTRIBUTIONS dependent? PLoS Genet. 4, e1000184.

J.M.K., T.T.J., J.M.E., L.X., and B.S.S. performed the experiments. All authors Chaturvedi, V., Ertelt, J.M., Jiang, T.T., Kinder, J.M., Xin, L., Owens, K.J., participated in the experimental design and data analysis. J.M.K. and S.S.W. Jones, H.N., and Way, S.S. (2015). CXCR3 blockade protects against Listeria 125 wrote the manuscript with editorial input from all the authors. monocytogenes infection-induced fetal wastage. J. Clin. Invest. , 1713– 1725.

ACKNOWLEDGMENTS Claas, F.H., Gijbels, Y., van der Velden-de Munck, J., and van Rood, J.J. (1988). Induction of B cell unresponsiveness to noninherited maternal HLA antigens during fetal life. Science 241, 1815–1817. We thank Dr. Marc Jenkins for providing 2W1S-OVA transgenic mice and Drs. James Moon, Louis Muglia, Joseph Qualls, Anne Stevens, Kevin Urdahl Duley, L. (2009). The global impact of pre-eclampsia and eclampsia. Semin. for helpful discussions. This work as supported by the NIH-NIAID through Perinatol. 33, 130–137. awards R01-AI100934 and R21-AI112186 (to S.S.W.) and the NIH-NHLBI Dutta, P., and Burlingham, W.J. (2011). Microchimerism: tolerance vs. sensiti- through award R01-HL103745 (to A.F.S.). S.S.W. holds an Investigator in the zation. Curr. Opin. Organ Transplant. 16, 359–365. Pathogenesis of Infectious Disease award from the Burroughs Wellcome Dutta, P., Molitor-Dart, M., Bobadilla, J.L., Roenneburg, D.A., Yan, Z., Fund. Torrealba, J.R., and Burlingham, W.J. (2009). Microchimerism is strongly correlated with tolerance to noninherited maternal antigens in mice. Blood Received: April 13, 2015 114, 3578–3587. Revised: May 18, 2015 Accepted: May 27, 2015 Eikmans, M., van Halteren, A.G., van Besien, K., van Rood, J.J., Drabbels, J.J., Published: July 23, 2015 and Claas, F.H. (2014). Naturally acquired microchimerism: implications for transplantation outcome and novel methodologies for detection. Chimerism 5, 24–39. REFERENCES Erlebacher, A. (2013). Mechanisms of T cell tolerance towards the allogeneic 13 Akiyama, Y., Caucheteux, S.M., Vernochet, C., Iwamoto, Y., Tanaka, K., fetus. Nat. Rev. Immunol. , 23–33. Kanellopoulos-Langevin, C., and Benichou, G. (2011). Transplantation toler- Fontenot, J.D., Gavin, M.A., and Rudensky, A.Y. (2003). Foxp3 programs the ance to a single noninherited MHC class I maternal alloantigen studied in a development and function of CD4+CD25+ regulatory T cells. Nat. Immunol. TCR-transgenic mouse model. J. Immunol. 186, 1442–1449. 4, 330–336. Andrassy, J., Kusaka, S., Jankowska-Gan, E., Torrealba, J.R., Haynes, L.D., Gammill, H.S., Stephenson, M.D., Aydelotte, T.M., and Nelson, J.L. (2015). Marthaler, B.R., Tam, R.C., Illigens, B.M., Anosova, N., Benichou, G., and Bur- Microchimerism in women with recurrent miscarriage. Chimerism 16, 1–3.

Cell 162, 1–11, July 30, 2015 ª2015 Elsevier Inc. 9 Please cite this article in press as: Kinder et al., Cross-Generational Reproductive Fitness Enforced by Microchimeric Maternal Cells, Cell (2015), http://dx.doi.org/10.1016/j.cell.2015.07.006

Hirayama, M., Azuma, E., and Komada, Y. (2012). Tolerogenic effect of non- Foxp3+ and Foxp3- subsets of self-peptide/MHC class II-specific CD4+ inherited maternal antigens in hematopoietic stem cell transplantation. Front. T cells. Proc. Natl. Acad. Sci. USA 108, 14602–14607. 3 Immunol. , 135. Munoz-Suano, A., Hamilton, A.B., and Betz, A.G. (2011). Gimme shelter: the Hori, S., Nomura, T., and Sakaguchi, S. (2003). Control of regulatory immune system during pregnancy. Immunol. Rev. 241, 20–38. 299 T cell development by the transcription factor Foxp3. Science , Mylonakis, E., Paliou, M., Hohmann, E.L., Calderwood, S.B., and Wing, E.J. 1057–1061. (2002). Listeriosis during pregnancy: a case series and review of 222 cases. Ichinohe, T., Uchiyama, T., Shimazaki, C., Matsuo, K., Tamaki, S., Hino, M., Medicine (Baltimore) 81, 260–269. Watanabe, A., Hamaguchi, M., Adachi, S., Gondo, H., et al.; Japanese Collab- Nelson, J.L. (2012). The otherness of self: microchimerism in health and dis- orative Study Group for NIMA-Complementary Haploidentical Stem Cell ease. Trends Immunol. 33, 421–427. Transplantation (2004). Feasibility of HLA-haploidentical hematopoietic stem cell transplantation between noninherited maternal antigen (NIMA)-mis- Nelson, R.W., McLachlan, J.B., Kurtz, J.R., and Jenkins, M.K. (2013). matched family members linked with long-term fetomaternal microchimerism. CD4+ T cell persistence and function after infection are maintained 190 Blood 104, 3821–3828. by low-level peptide:MHC class II presentation. J. Immunol. , 2828– 2834. Israeli, M., Kristt, D., Nardi, Y., and Klein, T. (2014). Genetic considerations in human sex-mate selection: partners share human leukocyte antigen but Ober, C., Weitkamp, L.R., Cox, N., Dytch, H., Kostyu, D., and Elias, S. (1997). 61 not short-tandem-repeat identity markers. Am. J. Reprod. Immunol. 71, HLA and mate choice in humans. Am. J. Hum. Genet. , 497–504. 467–471. Owen, R.D. (1945). Immunogenetic Consequences of Vascular Anastomoses 102 Jenkins, M.K., Chu, H.H., McLachlan, J.B., and Moon, J.J. (2010). between Bovine Twins. Science , 400–401. On the composition of the preimmune repertoire of T cells specific for Owen, R.D., Wood, H.R., Foord, A.G., Sturgeon, P., and Baldwin, L.G. (1954). Peptide-major histocompatibility complex ligands. Annu. Rev. Immunol. Evidence for Actively Acquired Tolerance to Rh Antigens. Proc. Natl. Acad. 28, 275–294. Sci. USA 40, 420–424. Jiang, T.T., Chaturvedi, V., Ertelt, J.M., Kinder, J.M., Clark, D.R., Valent, A.M., Picus, J., Aldrich, W.R., and Letvin, N.L. (1985). A naturally occurring bone- Xin, L., and Way, S.S. (2014). Regulatory T cells: new keys for further unlocking marrow-chimeric primate. I. Integrity of its immune system. Transplantation the enigma of fetal tolerance and pregnancy complications. J. Immunol. 192, 39, 297–303. 4949–4956. Rees, W., Bender, J., Teague, T.K., Kedl, R.M., Crawford, F., Marrack, P., and Kim, J.M., Rasmussen, J.P., and Rudensky, A.Y. (2007). Regulatory T cells Kappler, J. (1999). An inverse relationship between T cell receptor affinity and prevent catastrophic autoimmunity throughout the lifespan of mice. Nat. antigen dose during CD4(+) T cell responses in vivo and in vitro. Proc. Natl. Immunol. 8, 191–197. Acad. Sci. USA 96, 9781–9786. Loubie` re, L.S., Lambert, N.C., Flinn, L.J., Erickson, T.D., Yan, Z., Guthrie, K.A., Rowe, J.H., Ertelt, J.M., Aguilera, M.N., Farrar, M.A., and Way, S.S. (2011). Vickers, K.T., and Nelson, J.L. (2006). Maternal microchimerism in healthy Foxp3(+) regulatory T cell expansion required for sustaining pregnancy adults in lymphocytes, monocyte/macrophages and NK cells. Lab. Invest. compromises host defense against prenatal bacterial pathogens. Cell Host 86, 1185–1192. Microbe 10, 54–64. Maloney, S., Smith, A., Furst, D.E., Myerson, D., Rupert, K., Evans, P.C., and Rowe, J.H., Ertelt, J.M., Xin, L., and Way, S.S. (2012a). Listeria monocyto- Nelson, J.L. (1999). Microchimerism of maternal origin persists into adult life. genes cytoplasmic entry induces fetal wastage by disrupting maternal J. Clin. Invest. 104, 41–47. Foxp3+ regulatory T cell-sustained fetal tolerance. PLoS Pathog. 8, Matsuoka, K., Ichinohe, T., Hashimoto, D., Asakura, S., Tanimoto, M., and e1002873. Teshima, T. (2006). Fetal tolerance to maternal antigens improves the outcome Rowe, J.H., Ertelt, J.M., Xin, L., and Way, S.S. (2012b). Pregnancy imprints of allogeneic bone marrow transplantation by a CD4+ CD25+ T-cell-depen- regulatory memory that sustains anergy to fetal antigen. Nature 490, dent mechanism. Blood 107, 404–409. 102–106. Mold, J.E., and McCune, J.M. (2012). Immunological tolerance during fetal Samstein, R.M., Josefowicz, S.Z., Arvey, A., Treuting, P.M., and Rudensky, development: from mouse to man. Adv. Immunol. 115, 73–111. A.Y. (2012). Extrathymic generation of regulatory T cells in placental mammals 150 Mold, J.E., Michae¨ lsson, J., Burt, T.D., Muench, M.O., Beckerman, K.P., mitigates maternal-fetal conflict. Cell , 29–38. Busch, M.P., Lee, T.H., Nixon, D.F., and McCune, J.M. (2008). Maternal allo- Santner-Nanan, B., Peek, M.J., Khanam, R., Richarts, L., Zhu, E., Fazekas de antigens promote the development of tolerogenic fetal regulatory T cells in St Groth, B., and Nanan, R. (2009). Systemic increase in the ratio between utero. Science 322, 1562–1565. Foxp3+ and IL-17-producing CD4+ T cells in healthy pregnancy but not in pre- 183 Molitor, M.L., Haynes, L.D., Jankowska-Gan, E., Mulder, A., and Burlingham, eclampsia. J. Immunol. , 7023–7030. W.J. (2004). HLA class I noninherited maternal antigens in cord blood and Sasaki, Y., Sakai, M., Miyazaki, S., Higuma, S., Shiozaki, A., and Saito, S. breast milk. Hum. Immunol. 65, 231–239. (2004). Decidual and peripheral blood CD4+CD25+ regulatory T cells in early Molitor-Dart, M.L., Andrassy, J., Kwun, J., Kayaoglu, H.A., Roenneburg, D.A., pregnancy subjects and spontaneous abortion cases. Mol. Hum. Reprod. 10 Haynes, L.D., Torrealba, J.R., Bobadilla, J.L., Sollinger, H.W., Knechtle, S.J., , 347–353. and Burlingham, W.J. (2007). Developmental exposure to noninherited Spurgin, L.G., and Richardson, D.S. (2010). How pathogens drive genetic di- maternal antigens induces CD4+ T regulatory cells: relevance to mechanism versity: MHC, mechanisms and misunderstandings. Proc. Biol. Sci. 277, of heart allograft tolerance. J. Immunol. 179, 6749–6761. 979–988. Moon, J.J., Chu, H.H., Pepper, M., McSorley, S.J., Jameson, S.C., Kedl, R.M., Thornton, A.M., Korty, P.E., Tran, D.Q., Wohlfert, E.A., Murray, P.E., Belkaid, and Jenkins, M.K. (2007). Naive CD4(+) T cell frequency varies for different epi- Y., and Shevach, E.M. (2010). Expression of Helios, an Ikaros transcription fac- topes and predicts repertoire diversity and response magnitude. Immunity 27, tor family member, differentiates thymic-derived from peripherally induced 203–213. Foxp3+ T regulatory cells. J. Immunol. 184, 3433–3441. Moon, J.J., Chu, H.H., Hataye, J., Paga´ n, A.J., Pepper, M., McLachlan, J.B., Trupin, L.S., Simon, L.P., and Eskenazi, B. (1996). Change in paternity: a risk Zell, T., and Jenkins, M.K. (2009). Tracking epitope-specific T cells. Nat. Pro- factor for preeclampsia in multiparas. Epidemiology 7, 240–244. 4 toc. , 565–581. Uzonna, J.E., Wei, G., Yurkowski, D., and Bretscher, P. (2001). Immune elim- Moon, J.J., Dash, P., Oguin, T.H., 3rd, McClaren, J.L., Chu, H.H., Thomas, ination of Leishmania major in mice: implications for immune memory, vacci- P.G., and Jenkins, M.K. (2011). Quantitative impact of thymic selection on nation, and reactivation disease. J. Immunol. 167, 6967–6974.

10 Cell 162, 1–11, July 30, 2015 ª2015 Elsevier Inc. Please cite this article in press as: Kinder et al., Cross-Generational Reproductive Fitness Enforced by Microchimeric Maternal Cells, Cell (2015), http://dx.doi.org/10.1016/j.cell.2015.07.006

van Rood, J.J., Loberiza, F.R., Jr., Zhang, M.J., Oudshoorn, M., Claas, F., Zhou, L., Yoshimura, Y., Huang, Y., Suzuki, R., Yokoyama, M., Okabe, M., and Cairo, M.S., Champlin, R.E., Gale, R.P., Ringde´ n, O., Hows, J.M., and Horo- Shimamura, M. (2000). Two independent pathways of maternal cell transmis- witz, M.H. (2002). Effect of tolerance to noninherited maternal antigens on sion to offspring: through placenta during pregnancy and by breast-feeding the occurrence of graft-versus-host disease after bone marrow transplantation after birth. Immunology 101, 570–580. from a parent or an HLA-haploidentical sibling. Blood 99, 1572–1577. Zinkernagel, R.M., and Doherty, P.C. (1979). MHC-restricted cytotoxic T cells: Wilcox, A.J., Weinberg, C.R., O’Connor, J.F., Baird, D.D., Schlatterer, J.P., studies on the biological role of polymorphic major transplantation antigens Canﬁeld, R.E., Armstrong, E.G., and Nisula, B.C. (1988). Incidence of early determining T-cell restriction-speciﬁcity, function, and responsiveness. Adv. loss of pregnancy. N. Engl. J. Med. 319, 189–194. Immunol. 27, 51–177.

Cell 162, 1–11, July 30, 2015 ª2015 Elsevier Inc. 11