Natural killer (NK) cells represent a crucial component of the innate immune system and are primarily regulated by the interactions of their activation and inhibitory receptors with ligands available on target cells. The genetically linked Ly49 and NKR-P1 family of receptors constitute two of the major regulatory receptor systems used by NK cells and have been shown to bind different ligands. Whereas the Ly49 receptors survey MHC-I ligands on target cells, the NKR-Pl receptor family members bind to various members of the C-type lectin-related (Clr) family. Interestingly, NKR-P1 and Clr haplotypes possess a stable genomic polymorphism across multiple mouse strains, suggesting that this inhibitory receptor:ligand relationship has an important role in the maintenance of host cellular cognate specificities. The NKR-P1 and Clr receptor-ligand pairs identified in mice include the NKR-P1B:Clr-b and the NKR-P1G:Clr-f interacting pairs. Previous RT-PCR and in situ RNA hybridization data generated by our laboratory determined that kidney tubular epithelium as well as the small and large intestinal epithelial cells specifically and highly expresses the Clr-f transcripts. Contrarily, the Clr-b transcripts were only detected on hematopoietic cells of various lymphoid organs and kidneys. Moreover, foregoing studies revealed that the loss of Clr-b following viral or chemical induced stress mediates NK cell killing of the target cell, suggesting a tissue-specific immune-surveillance mechanism in parallel with the global MHC-I-dependent missing-self model. However, the role of the NKR-P1B:Clr-b recognition-system have never been examined in the intestine. Additionally, the role of Clr-f in the kidney and intestines, where they are highly expressed, has not been investigated. For these reasons, I aimed in my thesis to provide a better understanding of the functional aspect of the NKR-P1B:Clr-b and NKR-P1G:Clr-f recognition systems in mediating gut mucosal and renal homeostasis, respectively. First, in order to determine the role of NKR-P1B and Clrb receptor:ligand pair as a “missing-self” immunosurveillance system in the gut, I started by identifying the expression pattern of both the receptor and ligand on various intestinal cells. My results demonstrate that NK cells do not represent the major NKR-P1B-expressing cells in the gut lamina propria. Instead, ILC3 subsets constituted the predominant cell population expressing the receptor, whereas γδT cells composed a small fraction of NKR-P1B+ lymphocytes. In addition, the NKR-P1B expression on myeloid cells was exclusive to colon macrophages and DC subsets. Interestingly, the highest percentage of NKR-P1B+ immune cells was found in the gut, which suggests the dominant role of NKR-P1B in regulating immune functions at the level of intestinal mucosa. As expected, the expression of the NKR-P1B ligand, Clr-b, appeared on all innate immune cell types in the gut. Next, using oral infection models of Salmonela typhimurium and Citrobacter rodentium, I showed that NKR-P1B-deficient NK cells, ILC3 and γδ T cells are hyporesponsive compared to their WT counterparts. In particular, gut NKR-P1B-deficient NK cells and γδT cells secreted low levels of IFNγ cytokine while infected with S.typhimurium. Importantly, the decreased IFNγ secretion by NK and γδT cells was associated with an increased dissemination of the bacterium into the knockout spleens at day 5 post-infection. Likewise, I detected a significant decrease in IL-22 cytokine production by NKR-P1B-deficient ILC3 compared to their WT counterparts at both steady state and following C.rodentium infection. Next, I address the potential role of Clr-f in the kidney. Renal tubular epithelial cells have been shown to express high levels of Clr-f transcripts. Epithelial cells constitute the major cellular component of kidney tubules and are well known to mediate metabolic waste excretion, reabsorption of essential molecules as well as other physiological functions, such as ions exchange and water retention. To determine the role of Clr-f in renal epithelial cells, I generated a Clr-f-deficient mouse with the help of two of my previous lab colleagues. Importantly, chemical analysis on urine and serum samples from knockout and WT littermates indicated that Clr-f-deficient kidneys display a decreased filtration capacity. In particular, higher creatinine levels were detected in the Clr-f deficient serum. In addition, Clr-f-deficient mice appeared to have a lower fractional excretion of sodium (FENa) in their urine filtrates in comparison to WT excreted urine. Blood pressure measurements on the same mice at 12 and 24 weeks of age revealed a hypotensive phenotype in the Clr-f-deficient mice. Furthermore, pathological assessment of Clr-f-deficient kidneys exhibited moderate and aggravated lesions of the tubular epithelium along with marked glomerular mesangiolysis. Lastly, flow cytometry analysis on isolated lymphocytes from Clr-f-deficient and WT mice demonstrated comparable immune infiltrates between the two mouse genotypes. Altogether, our data shows that the absence of Clr-f results in the development of glomerular and tubular lesions in an immune-independent manner leading to an abnormal kidney function. Additionally, the disruption of NKR-P1B:Clr-b recognition system results in abnormal innate immune cell number and function in the mouse intestine. These novel findings sheds light on the important role of Clr-f in maintaining healthy kidney morphology and function, as well as the crucial role for NKR-P1B:Clr-b interactions in mediating intestinal homeostasis at steady and infected states.

Angeborene lymphoide Zellen (ILCs) sind sich im Gewebe befindliche Zellen, die eine wichtige Rolle bei der Aufrechterhaltung der Gewebehomöostase spielen. ILCs wurden in verschiedenen Organen untersucht. Ob ILCs im zentralen Nervensystem (ZNS) vorhanden sind und wenn ja, welchen ​​Phänotyp und welche funktionellen Eigenschaften sie in diesem Organ aufweisen, sind Fragen, die bisher unbeantwortet blieben. NK-Zellen sind die seit langem bekannten ILC-Mitglieder, die viele Merkmale mit ILC1s teilen. Im Zusammenhang mit der Autoimmunität wurde gezeigt, dass NK-Zellen eine immunmodulatorische Rollen spielen. Anhand des Tiermodells von Multiple Sklerose, der experimentellen autoimmunen Enzephalomyelitis, zeigte unsere Gruppe, dass reife NK-Zellen auf CX3CR1-abhängige Weise in das ZNS rekrutiert werden. Auf der Grundlage dieser Beobachtungen will ich in meinem PhD Projekt die Chemokinrezeptoren definieren, die die Rekrutierung der unreifen NK-Zellen in das entzündete ZNS vermitteln. Des Weiteren will ich herausfinden, ob die phänotypisch definierten NK-Zellen (CD3-NK1.1 + -Zellen) die im gesunden ZNS vorhanden sind, echte NK-Zellen sind, oder sie zu den ILC1s gezählt werden können. Die Ergebnisse der vorliegenden Arbeit zeigen, dass die im gesunden ZNS vorhandenen CD3-NK1.1+ -Zellen verschiedene Typ-1-ILC-Subsets umfassen: NK-Zellen, ILC1s, Intermediat-ILC1s und Ex-ILC3s. CXCR3 wurde auf ILC1s und einer Fraktion von unreifen NK-Zellen exprimiert, trug jedoch nicht zur Rekrutierung von NK-Zellen in das ZNS im EAE-Modell bei. Die Expression von CD49a, CD69, CXCR6, DNAM-1high, TRAIL und CD200R und das Fehlen von Eomes unterschieden die ILC1 von den NK-Zellen im ZNS. Zusätzlich ILC1s sezernierten mehr TNF-α als NK-Zellen. ILC1s waren die dominante Typ-1-ILC Subgruppe im Plexus choroideus und im Gehirnparenchym und waren auch in den Meningen vorhanden. Zusammenfassend bietet die vorliegende Arbeit zum ersten Mal eine umfassende Charakterisierung von ILCs des Typs I im ZNS. / Innate lymphoid cells (ILCs) are tissue resident cells that play important roles in the maintenance of tissue homeostasis. ILCs have been characterized in diverse organs like the gut and liver. However, whether ILCs are present in the central nervous system (CNS) and if so, what are their phenotype and function in this organ are questions that remain unanswered. NK cells are the longer-known ILC members that share many phenotypical and functional features with ILC1s. Using the animal model of MS, the experimental autoimmune encephalomyelitis (EAE), our group showed that protective mature NK cells are recruited to the CNS on an CX3CR1-dependent manner. Based on that observations, my PhD project aimed to 1) define the chemokine receptors that mediate the recruitment of the immature NK cells into the inflamed CNS and to 2) determine whether the phenotypically defined NK cells (CD3-NK1.1+ cells) that are present in the CNS during steady state constitute bona fide NK cells or constitute also other group 1 ILC subsets. The results of this work indicate that the CD3-NK1.1+ cells present in the healthy CNS comprise diverse group 1 ILC subsets that include conventional NK cells, ILC1s, intermediate-ILC1s and ex-ILC3s. CXCR3 was expressed on ILC1s and a fraction of immature NK cells, but did not contributed to the recruitment of NK cells into the CNS in the EAE model. In addition, the phenotypic and functional characterization of the newly identified CNS-ILC1s is described. The exclusive expression of CD49a, CD69, CXCR6, DNAM-1high, TRAIL and CD200R, and lack of Eomes distinguished the ILC1s from the NK cells in the CNS. IILC1s secreted IFN-γ and more TNF-α than NK cells upon stimulation in the healthy and EAE mice and were the dominant group 1 ILC subset in the choroid plexus and brain parenchyma and were also present in the meninges. In sum, the present work provides for the first time a comprehensive characterization of group 1 ILCs in the CNS.

Uterine NK (uNK) cells are a distinct subset of NK cells in the decidua of humans and rodents during pregnancy, which are essential for remodelling of the spiral arteries supplying the feto-placental unit. Similarly to peripheral NK cells, uNK cells express Natural Killer receptors (NKRs) that engage MHC class I molecules. Evidence from human genetic association studies suggests that, in the presence of allogeneic cognate paternal MHC class I ligands, inhibitory uterine NKRs are associated with disorders of pregnancy arising from impaired decidual vascular remodelling. Conversely, enhancement of human uNK cell activity through activating NKRs is associated with high birth weight. Evidence from mouse models corroborates that uNK cell activity is modulated by interactions between NKRs and MHC class I, but has largely focussed on the effect of paternal MHC. In this study, the contribution of maternal immunogenetic regulation of NK cell function to reproductive outcome was assessed independently of parental MHC disparity in mice. To evaluate the role of NKR genes in isolation, I used congenic B6.BALB-TC1 (TC1) mice that differ from C57BL/6 (B6) mice only within the region of chromosome six encoding NKRs that recognise MHC class I. Absence of a major inhibitory NKR for self-MHC, Ly49I, in TC1 mice causes a compensatory shift in the NKR repertoire expressed and preserves a majority subpopulation of educated NK cells. B6 and TC1 splenic and uterine NK cells are similarly functionally reactive and mature, and no significant differences could be detected in spiral arterial remodelling or fetal growth between these strains in MHC-syngeneic matings. This supports data from human immunogenetic studies showing that maternal uterine NKRs are not associated with differences in pregnancy outcome in the absence of novel paternal MHC class I ligands, and highlights the importance of maternal and paternal co-regulation of uNK cell activity during pregnancy. No mouse models of uNK cell activation are currently available with which to corroborate human immunogenetic associations between activating uterine NKRs and high birth weight. Male m157-transgenic (m157-Tg) mice, which ubiquitously express viral m157 glycoprotein ligands for the activating NKR Ly49H, were mated with B6 females. Exclusive expression of m157 glycoprotein by trophoblast improved placental efficiency, but did not enhance fetal growth. Some fertility clinics surmise that uNK cell activation initiates the pathogenesis of spontaneous abortion. It has been suggested that this may occur due to reduced expression by human uNK cells of miR-483-3p, which stimulates endogenous insulin-like growth factor (IGF)-1 production and uNK cell cytotoxicity in vitro. It is demonstrated here that neither miR-483-3p nor IGF-1 regulate murine NK cell development, maturation or function. No discernible reproductive phenotype is evident in miR-483 deficient females. It can be inferred that post-transcriptional control by miR-483 is not biologically relevant to murine NK cell function. Although m157-Tg mice may provide an interesting model to further study uNK cell-mediated placental adaptations, it remains important to identify a murine model of enhanced uNK cell function to corroborate human immunogenetic associations with high birth weight and to challenge the supposition that uNK cell activation is harmful to pregnancy.