Refine
Has Fulltext
- yes (15)
Is part of the Bibliography
- yes (15)
Document Type
- Doctoral Thesis (15)
Language
- English (15) (remove)
Keywords
- Immunologie (3)
- T-Lymphozyt (3)
- Transplantat-Wirt-Reaktion (3)
- Darm (2)
- GvHD (2)
- Peyer's patch (2)
- Zellmigration (2)
- Akute myeloische Leukämie (1)
- Arteriosklerose (1)
- Aspergillus fumigatus (1)
Institute
- Graduate School of Life Sciences (15)
- Fakultät für Biologie (3)
- Medizinische Fakultät (2)
- Medizinische Klinik und Poliklinik II (2)
- Abteilung für Molekulare Innere Medizin (in der Medizinischen Klinik und Poliklinik II) (1)
- Deutsches Zentrum für Herzinsuffizienz (DZHI) (1)
- Institut für Virologie und Immunbiologie (1)
- Rudolf-Virchow-Zentrum (1)
Humans are continuously exposed to airborne spores of the saprophytic fungus Aspergillus fumigatus. In healthy individuals, local pulmonary host defence mechanisms can efficiently eliminate the fungus without any overt symptoms. In contrast, A. fumigatus causes devastating infections in immunocompromised patients. However, local host immune responses against A. fumigatus lung infections in immunocompromised conditions have remained largely elusive.
Given the dynamic changes in immune cell subsets within tissues upon immunosuppressive therapy, we dissected the spatiotemporal pulmonary immune response after A. fumigatus infection to reveal basic immunological events that fail to effectively control the invasive fungal disease. In different immunocompromised murine models, myeloid but not lymphoid cells were strongly recruited upon infection. Notably, neutrophils and macrophages were recruited to infected lungs in different immunosuppressed regimens. Other myeloid cells, particularly dendritic cells and monocytes were only recruited in the corticosteroid model after infection. Lymphoid cells, particularly CD4+ or CD8+ T-cells and NK cells were highly reduced upon immunosuppression and were not recruited after A. fumigatus infection. Importantly, adoptive CD11b+ myeloid cell transfer rescued immunosuppressed mice from lethal A. fumigatus infection. These findings illustrate that CD11b+ myeloid cells are critical for anti-A. fumigatus defence under immunocompromised conditions.
Despite improved antifungal agents, invasive A. fumigatus lung infections cause a high rate morbidity and mortality in neutropenic patients. Granulocyte transfusions have been tested as an alternative therapy for the management of high-risk neutropenic patients with invasive A. fumigatus infections. To increase the granulocyte yield for transfusion, donors are treated with corticosteroids. Yet, the efficacy of granulocyte transfusion and the functional defence mechanisms of granulocytes collected from corticosteroid treated donors remain largely elusive.
We aimed to assess the efficacy of granulocyte transfusion and functional defence mechanisms of corticosteroid treated granulocytes using mouse models.
In this thesis, we show that transfusion of granulocytes from corticosteroid treated mice did not protect cyclophosphamide immunosuppressed mice against lethal A. fumigatus infection in contrast to granulocytes from untreated mice. Upon infection, increased levels of inflammatory cytokines helped to recruit granulocytes to the lungs without any recruitment defects in corticosteroid treated and infected mice or in cyclophosphamide immunosuppressed and infected mice that have received the granulocytes from corticosteroid treated mice. However, corticosteroid treated human or mouse neutrophils failed to form neutrophil extracellular traps (NETs) in in vitro and in vivo conditions. Further, corticosteroid treated granulocytes exhibited impaired ROS production against A. fumigatus. Notably, corticosteroids impaired the β-glucan receptor Dectin-1 (CLEC7A) on mouse and human granulocytes to efficiently recognize and phagocytize A. fumigatus, which markedly impaired fungal killing. We conclude that corticosteroid treatment of granulocyte donors for increasing neutrophil yields or patients with ongoing corticosteroid treatment could result in deleterious effects on granulocyte antifungal functions, thereby limiting the benefit of granulocyte transfusion therapies against invasive fungal infections.
Wilms tumor protein 1 (WT1) is a suitable target to develop an immunotherapeutic approach against high risk acute myeloid leukemia (AML), particularly their relapse after allogeneic hematopoietic stem cell transplantation (HSCT). As an intracellular protein traversing between nucleus and cytoplasm, recombinant expression of WT1 is difficult. Therefore, an induction of WT1-specific T-cell responses is mostly based on peptide vaccination as well as dendritic cell (DC) electroporation with mRNA encoding full-length protein to mount WT1-derived peptide variations presented to T cells. Alternatively, the WT1 peptide presentation could be broadened by forcing receptor-mediated endocytosis of DCs.
In this study, antibody fusion proteins consisting of an antibody specific to the human DEC205 endocytic receptor and various fragments of WT1 (anti-hDEC205-WT1) were generated for a potential DC-targeted recombinant WT1 vaccine. Anti-hDEC205-WT1 antibody fusion proteins containing full-length or major parts of WT1 were not efficiently expressed and secreted due to their poor solubility and secretory capacity. However, small fragment-containing variants: anti-hDEC205-WT110-35, anti-hDEC205-WT191-138, anti-hDEC205-WT1223-273, and anti-hDEC205-WT1324-371 were obtained in good yields.
Since three of these fusion proteins contain the most of the known immunogenic epitopes in their sequences, the anti-hDEC205-WT191-138, anti-hDEC205-WT1223-273, and anti-hDEC205-WT1324-371 were tested for their T-cell stimulatory capacities. Mature monocyte-derived DCs loaded with anti-hDEC205-WT191-138 could induce ex vivo T-cell responses in 12 of 16 blood samples collected from either healthy or HSC transplanted individuals compared to included controls (P < 0.01). Furthermore, these T cells could kill WT1-overexpressing THP-1 leukemia cells in vitro after expansion.
In conclusion, alongside proving the difficulty in expression and purification of intracellular WT1 as a vaccine protein, our results from this work introduce an alternative therapeutic vaccine approach to improve an anti-leukemia immune response in the context of allogeneic HSCT and potentially beyond.
The thesis provides insights in reconstruction and analysis pipelines for processing of
three-dimensional cell and vessel images of megakaryopoiesis in intact murine bone.
The images were captured in a Light Sheet Fluorescence Microscope. The work
presented here is part of Collaborative Research Centre (CRC) 688 (project B07) of
the University of Würzburg, performed at the Rudolf-Virchow Center. Despite ongoing
research within the field of megakaryopoiesis, its spatio-temporal pattern of
megakaryopoiesis is largely unknown. Deeper insight to this field is highly desirable to
promote development of new therapeutic strategies for conditions related to
thrombocytopathy as well as thrombocytopenia. The current concept of
megakaryopoiesis is largely based on data from cryosectioning or in vitro studies
indicating the existence of spatial niches within the bone marrow where specific stages
of megakaryopoiesis take place. Since classic imaging of bone sections is typically
limited to selective two-dimensional views and prone to cutting artefacts, imaging of
intact murine bone is highly desired. However, this has its own challenges to meet,
particularly in image reconstruction. Here, I worked on processing pipelines to account
for irregular specimen staining or attenuation as well as the extreme heterogeneity of
megakaryocyte morphology. Specific challenges for imaging and image reconstruction
are tackled and solution strategies as well as remaining limitations are presented and
discussed. Fortunately, modern image processing and segmentation strongly benefits
from continuous advances in hardware as well as software-development. This thesis
exemplifies how a combined effort in biomedicine, computer vision, data processing
and image technology leads to deeper understanding of megakaryopoiesis. Tailored
imaging pipelines significantly helped elucidating that the large megakaryocytes are
broadly distributed throughout the bone marrow facing a surprisingly dense vessel
network. No evidence was found for spatial niches in the bone marrow, eventually
resulting in a revised model of megakaryopoiesis.
The outcome of the innate immune response to biomaterials mainly determines whether the material will be incorporated in the body to fulfill its desired function or, when it gets encapsulated, will be rejected in the worst case. Macrophages are key players in this process, and their polarization state with either pro- (M1), anti-inflammatory (M2), or intermediate characteristics is crucial for deciding on the biomaterial’s fate. While a transient initial pro-inflammatory state is helpful, a prolonged inflammation deteriorates the proper healing and subsequent regeneration. Therefore, biomaterial-based polarization may aid in driving macrophages in the desired direction. However, the in vivo process is highly complex, and a mono-culture of macrophages in vitro displays only one part of the cellular system, but, to this date, there is a lack of established co-cultures to assess the immune response to biomaterials. Thus, this thesis aimed to establish a functional co-culture system of human macrophages and human mesenchymal stromal cells (hMSCs) to improve the assessment of the immune response to biomaterials in vitro. Together with macrophages, hMSCs are involved in tissue regeneration and inflammatory reactions and can modulate the immune response. In particular, endogenously derived hMSCs considerably contribute to the successful engrafting of biomaterials. This thesis focused on poly(ε-caprolactone) (PCL) fiber-based scaffolds produced by the technique of melt electrowriting (MEW) as biomaterial constructs. Via this fabrication technique, uniform, precisely ordered scaffolds varying in geometry and pore size have been created in-house.
To determine the impact of scaffold geometries and pore sizes on macrophages, mono-cultures incubated on scaffolds were conducted. As a pre-requisite to achieve a functional co-culture system on scaffolds, setups for direct and indirect systems in 2D have initially been established. These setups were analyzed for the capability of cell-cell communication. In parallel, a co-culture medium suitable for both cell types was defined, prior to the establishment of a step-by-step procedure for the co-cultivation of human macrophages and hMSCs on fiber-based scaffolds.
Regarding the scaffold morphologies tested within this thesis to improve M2-like polarization, box-shaped scaffolds outperformed triangular-, round- or disordered-shaped ones. Upon further investigation of scaffolds with box-shaped pores and precise inter-fiber spacing from 100 µm down to only 40 µm, decreasing pore sizes facilitated primary human macrophage elongation accompanied by their differentiation towards the M2 type, which was most pronounced for the smallest pore size of 40 µm. To the best of my knowledge, this was the first time that the elongation of human macrophages in a 3D environment has been correlated to their M2-like polarization. Thus, these results may set the stage for the design, the assessment, and the selection of new biomaterials, which can positively affect the tissue regeneration.
The cell communication of both cell types, detected via mitochondria exchange in direct and indirect co-cultures systems, took place in both directions, i.e., from hMSCs to macrophages and vice versa. Thereby, in direct co-culture, tunneling nanotubes enabled the transfer from one cell type to the respective other, while in indirect co-culture, a non-directional transfer through extracellular vesicles (EVs) released into the medium seemed likely. Moreover, the phagocytic activity of macrophages after 2D co-cultivation and hence immunomodulation by hMSCs increased with the highest phagocytic rate after 48 h being most pronounced in direct co-cultivation.
As the commonly used serum supplements for macrophages and hMSCs, i.e., human serum (hS) and fetal calf serum (FCS), respectively, failed to support the respective other cell type during prolonged cultivation, these sera were replaced by human platelet lysate (hPL), which has been proven to be the optimal supplement for the co-cultivation of human macrophages with hMSCs within this thesis. Thereby, the phenotype of both cell types, the distribution of both cell populations, the phagocytic activity of macrophages, and the gene expression profiles were maintained and comparable to the respective standard mono-culture conditions. This was even true when hPL was applied without the anticoagulant heparin in all cultures with macrophages, and therefore, heparin was omitted for further experiments comprising hPL and macrophages.
Accordingly, a step-by-step operating procedure for the co-cultivation on fiber-based scaffolds has been established comprising the setup for 3D cultivation as well as the description of methods for the analysis of phenotypical and molecular changes upon contact with the biomaterial. The evaluation of the macrophage response depending on the cultivation with or without hMSCs and either on scaffolds or on plastic surfaces has been successfully achieved and confirmed the functionality of the suggested procedures.
In conclusion, the functional co-culture system of human macrophages and hMSCs established here can now be employed to assess biomaterials in terms of the immune response in a more in vivo-related way. Moreover, specifically designed scaffolds used within the present thesis showed auspicious design criteria positively influencing the macrophage polarization towards the anti-inflammatory, pro-healing type and might be adaptable to other biomaterials in future approaches.
Hence, follow-up experiments should focus on the evaluation of the co-culture outcome on promising scaffolds, and the suggested operating procedures should be adjusted to further kinds of biomaterials, such as cements or hydrogels.
The role of host dendritic cells during the effector phase of intestinal graft-versus-host disease
(2014)
Monocytes can be functionally divided in two subsets, both capable to differentiate into dendritic cells (DCs): CX3CR1loCCR2+ classical monocytes, actively recruited to the sites of inflammation and direct precursors of inflammatory DCs; and CX3CR1hiCCR2− non-classical monocytes, characterized by CX3CR1-dependent recruitment to non-inflamed tissues. Yet, the function of non-classical monocyte-derived DCs (nc-mo-DCs), and the factors, which trigger their recruitment and DC differentiation, have not been clearly defined to date. Here we show that in situ differentiated nc-moDCs mediate immunosuppression in the context of intestinal graft-versus-host disease (GVHD).
Employing multi-color confocal microscopy we observed a dramatic loss of steady state host-type CD103+ DC subset immediately after transplantation, followed by an enrichment of immune-regulatory CD11b+ nc-moDCs. Parabiosis experiments revealed that tissue-resident non-classical CX3CR1+ monocytes differentiated in situ into intestinal CD11b+ nc-moDCs after allogeneic hematopoietic cell transplantation (allo-HCT). Differentiation of this intestinal DC subset depended on CSF-1 but not on Flt3L, thus defining the precursors as monocytes and not pre-DCs. Importantly, CX3CR1 but not CCR2 was required for this DC subset differentiation, hence defining the precursors as non-classical monocytes. In addition, we identify PD-L1 expression by CX3CR1+ nc-moDCs as the major mechanism they employ to suppress alloreactive T cells during acute intestinal GVHD. All together, we demonstrate that host nc-moDCs surprisingly mediate immunosuppression in the context of murine intestinal GVHD – as opposed to classical “inflammatory” monocyte-derived dendritic cells (mo-DCs) – via coinhibitory signaling. This thorough study unravels for the first time a biological function of a - so far only in vitro and phenotypically described - DC subset. Our identification of this beneficial immunoregulatory DC subset points towards alternate future strategies in underpinning molecular pathways to foster their function. We describe an unexpected mechanism of nc-moDCs in allo-HCT and intestinal GVHD, which might also be important for autoimmune disorders or infections of the gastrointestinal tract.
Cardiovascular disease and the acute consequence of myocardial infarc- tion remain one of the most important causes of morbidity and mortality in all western societies. While much progress has been made in mitigating the acute, life-threatening ischemia caused by infarction, heart failure of the damaged my- ocardium remains prevalent. There is mounting evidence for the role of T cells in the healing process after myocardial infarction, but relevant autoantigens, which might trigger and regulate adaptive immune involvement have not been discov- ered in patients.
In this work, we discovered an autoantigenic epitope in the adrenergic receptor beta 1, which is highly expressed in the heart. This autoantigenic epitope causes a pro-inflammatory immune reaction in T cells isolated from pa- tients after myocardial infarction (MI) but not in control patients. This immune reaction was only observed in a subset of MI patients, which carry at least one allele of the HLA-DRB1*13 family. Interestingly, HLA-DRB1*13 was more com- monly expressed in patients in the MI group than in the control group.
Taken together, our data suggests antigen-specific priming of T cells in MI patients, which leads to a pro-inflammatory phenotype. The primed T cells react to a cardiac derived autoantigen ex vivo and are likely to exhibit a similar phenotype in vivo. This immune phenotype was only observed in a certain sub- set of patients sharing a common HLA-allele, which was more commonly ex- pressed in MI patients, suggesting a possible role as a risk factor for cardiovas- cular disease.
While our results are observational and do not have enough power to show strong clinical associations, our discoveries provide an essential tool to further our understanding of involvement of the immune system in cardiovascu- lar disease. We describe the first cardiac autoantigen in the clinical context of MI and provide an important basis for further translational and clinical research in cardiac autoimmunity.
Regulatory T cells (Tregs) are the masters of immune regulation controlling inflammation and tolerance, tissue repair and homeostasis. Multiple immunological diseases result from altered Treg frequencies and Treg dysfunction. We hypothesized that augmenting Treg function and numbers would prevent inflammatory disease whereas inhibiting or depleting Tregs would improve cancer immunotherapy.
In the first part of this thesis, we explored whether in vivo activation and expansion of Tregs would impair acute graft-versus-host disease (aGvHD). In this inflammatory disease, Tregs are highly pathophysiological relevant and their adoptive transfer proved beneficial on disease outcome in preclinical models and clinical studies. IL-2 has been recognized as a key cytokine for Treg function. Yet, attempts in translating Treg expansion via IL-2 have remained challenging, due to IL-2s extremely broad action on other cell types including effector T cells, NK cells, eosinophils and vascular leakage syndrome, and importantly, due to poor pharmacokinetics in vivo. We addressed the latter issue using an IL-2-IgG-fusion protein (irrIgG-IL-2) with improved serum retention and demonstrated profound Treg expansion in vivo in FoxP3-luciferase reporter mice. Further, we augmented Treg numbers and function via the selective-TNF based agonists of TNFR2 (STAR2). Subsequently, we tested a next-generation TNFR2 agonist, termed NewSTAR, which proved even more effective. TNFR2 stimulation augmented Treg numbers and function and was as good as or even superior to the IL-2 strategy. Finally, in a mouse model of aGvHD we proved the clinical relevance of Treg expansion and activation with irrIgG-IL-2, STAR2 and NewSTAR. Notably, the TNFR2 stimulating constructs were outstanding as we observed not the IL-2 prototypic effects on other cell populations and no severe side effects.
In the second part of this thesis, we explored Tregs in pancreatic ductal adenocarcinoma (PDAC) and developed targeting strategies. Among several tumor entities in which Tregs impact survival, preclinical and clinical data demonstrated their negative role on PDAC. In our studies we employed the orthotopic syngeneic Panc02 model in immunocompetent mice. Based on flow cytometric analysis of the tumor microenvironment we propose TIGIT and TNFRSF members as novel therapeutic targets. Surprisingly, we found that blocking TNFR2 did not interfere with intratumoral Treg accumulation. However, we decreased the highly abundant intratumoral Tregs when we disrupted the tumor extracellular matrix. In PDAC, Treg manipulation alone did not lead to tumor regression and we propose that an additional immune boost may be necessary for efficient tumor immune surveillance and cancer clearance. This contrasts with aGvHD, in which Treg manipulation alone was sufficient to improve disease outcome.
Conclusively, we demonstrated the enormous medical benefit of Treg manipulation. Our promising data obtained with our newly developed powerful tools highlight the potential to translate our findings into clinical practice to therapeutically target human Tregs in patients. With novel TNFR2 agonists (STAR2, NewSTAR) we augmented Treg numbers and function as (or even more) effectively than with IL-2, without causing adverse side effects. Importantly, exogenous in vivo Treg expansion protected mice from aGvHD. For the therapy of PDAC, we identified novel targets on Tregs, notably TIGIT and members of the TNFRSF. We demonstrated that altering the extracellular tumor matrix can efficiently disrupt the Treg abundance in tumors. These novel targeting strategies appear as attractive new treatment options and they may benefit patients suffering from inflammatory disease and cancer in the future.
In the initiation phase of acute graft-versus-host disease (aGvHD), CD4+ T cells are activated by hematopoietic antigen presenting cells in secondary lymphoid organs whereas in effector phase by non-hematopoietic cells in the small intestine. We hypothesized that alloreactive CD4+ T cells primarily home to the secondary lymphoid organs subsequent to allogeneic hematopoietic cell transplantation in the initiation phase of aGvHD and are activated by the non-hematopoietic lymph node stromal cells via MHC class II. To test this hypothesis, we employed CD4+ T cell-dependent major mismatch aGvHD mouse model to study this correlation.
Upon analyzing the early events following allo-HCT with bioluminescence imaging, flow cytometry and whole-mount light sheet fluorescence microscopy, we found that allogeneic T cells exclusively home to the spleen, lymph nodes and the Peyer’s patches and not to the intestinal lamina propria in the initiation phase of aGvHD. Utilizing mice devoid of partial or complete hematopoietic antigen presentation we could show allogeneic CD4+ T cells activation in the lymphoid organs of MHCIIΔCD11c and MHCIIΔ BM chimeric mice early after allo-HCT. MHCIIΔ BM chimeras failure of thymic negative selection and developing tissue wasting disease upon syn-HCT deemed them unsuitable to study non-hematopoietic antigen presentation in aGvHD. To overcome this challenge, we generated MHCIIΔVav1 mice that lack MHC class II expression on all hematopoietic cells. MHCIIΔVav1 mice were susceptible to aGvHD and LNSCs from these animals activated allogeneic CD4+ T cells in mixed lymphocyte reaction. Likewise, mesenteric lymph nodes from CD11c.DTR mice surgically transplanted into a MHCIIΔ mouse could activate CD4+ T cells in vivo, clearly demonstrating LNSCs as non-hematopoietic APCs of the lymphoid organs.
We specifically target lymph node stromal cell subsets via the Cre/loxP system, we employed single cell RNA sequencing and selected Ccl19 and VE-Cadherin to specifically target the fibroblastic reticular cells and endothelial cells of the lymph nodes respectively. In MHCIIΔCcl19 mice, alloreactive CD4+ T cells activation was discreetly reduced in the initiation phase of aGvHD whereas absence of MHCII on fibroblastic reticular cells resulted in hyper-activation of allogeneic CD4+ T cells leading to poor survival. This phenotype was modulated by the regulatory T cells that were able to rescue H2-Ab1fl mice but not the MHCIIΔCcl19 subsequent to GvHD.
Knock-out of MHCII on endothelial cells MHCIIΔVE Cadherin, resulted only in modest reduction of CD4+ T cells activation in the initiation phase of GvHD, conversely MHCIIΔVE Cadherin mice showed a protective phenotype compared against littermates H2-Ab1fl mice in long-term survival. Furthermore, to pin-point endothelial cells MHCII antigen presentation we generated MHCIIΔVE Cadherin ΔVav1 animals devoid of antigen presentation in both endothelial and hematopoietic compartments. LNSCs from MHCIIΔVE Cadherin ΔVav1 were unable to activate alloreactive CD4+ T cells in mixed lymphocyte reaction.
Altogether, we demonstrate for the first time that MHC class II on the lymph node stromal cells plays a crucial role in the modulation of allogeneic CD4+ T cells in the initiation and later in the effector phase of graft-versus-host-disease.
MDSCs are suppressive immune cells with a high relevance in various pathologies including cancer, autoimmunity, and chronic infections. Surface marker expression of MDSCs resembles monocytes and neutrophils which have immunostimulatory functions instead of suppressing T cells. Therefore, finding specific surface markers for MDSCs is important for MDSC research and therapeutic MDSC manipulation. In this study, we analyzed if the integrin VLA-1 has the potential as a novel MDSC marker. VLA-1 was expressed by M-MDSCs but not by G-MDSCs as well as by Teff cells. VLA-1 deficiency did not impact iNOS expression, the distribution of M-MDSC and G-MDSC subsets, and the suppressive capacity of MDSCs towards naïve and Teff cells in vitro. In mice, VLA-1 had no effect on the homing capability of MDSCs to the spleen, which is a major reservoir for MDSCs. Since the splenic red pulp contains collagen IV and VLA-1 binds collagen IV with a high affinity, we found MDSCs and Teff cells in this area as expected. We showed that T cell suppression in the spleen, indicated by reduced T cell recovery and proliferation as well as increased apoptosis and cell death, partially depended on VLA-1 expression by the MDSCs. In a mouse model of multiple sclerosis, MDSC injection prior to disease onset led to a decrease of the disease score, and this effect was significantly reduced when MDSCs were VLA-1 deficient. The expression of Sema7A by Teff cells, a ligand for VLA-1 which is implicated in negative T cell regulation, resulted in a slightly stronger Teff cell suppression by MDSCs compared to Sema7A deficient T cells. Live cell imaging and intravital 2-photon microscopy showed that the interaction time of MDSCs and Teff cells was shorter when MDSCs lacked VLA 1 expression, however VLA-1 expression had no impact on MDSC mobility. Therefore, the VLA-1-dependent interaction of MDSC and Teff cells on collagen IV in the splenic red pulp is implicated MDSC-mediated Teff cell suppression.
T cell infiltration into the intestine occurs after priming and activation in the mesenteric lymph nodes and Peyer’s patches and subsequent trafficking via the blood circulation. We hypothesized that additionally to the vascular trafficking route, a fraction of T cells in the Peyer’s patches directly migrate into the adjacent lamina propria of the small intestine. To test this hypothesis, we employed a mouse model of acute Graft-versus-Host Disease to study the direct T cell migration from the Peyer’s patches to the adjacent lamina propria.
First, we analyzed the border of Peyer’s patches on histological sections and found that the Peyer’s patch is not enclosed by a capsule or basement membrane. Thus, the tissue architecture allows for direct access to the surrounding tissue. With whole-mount light sheet fluorescence microscopy we quantified a three-dimensional gradient of T cells around Peyer’s patches on day 2.5 and day 3 after transplantation. This gradient evened out at day 4 and day 6 when high numbers of T cells started to evenly infiltrate the intestine from the blood circulation. We confirmed that gradient-forming T cells around Peyer’s patches resided within the tissue parenchyma of the lamina propria and not inside lymphatic vessels.
To positively prove that the recently activated donor T cells around Peyer’s patches have egressed directly from that patch, we established a protocol for intravital photoconversion of T cells inside Peyer’s patches. 12 h after photoconversion inside a single Peyer’s patch, photoconverted T cells resided only around this particular Peyer’s patch and not elsewhere in the small intestine. This indicated that the T cells did not infiltrate via the blood but migrated to the adjacent lamina propria of the small intestine. Dynamic intravital two-photon microscopy revealed that these T cells next to the Peyer’s patch migrated in a random pattern. This suggested that these cells did not follow a positive chemoattractive gradient once they had reached the lamina propria. Laser-capture microdissection combined with RNA sequencing of the mucosa near the Peyer’s patch identified a wide range of migration-promoting factors. These included chemokines, co-stimulatory receptors and migration-associated intracellular molecules, which are candidates to promote this direct migration from Peyer’s patches.
Altogether, we demonstrate for the first time that additionally to the vascular trafficking route, a fraction of T cells migrates directly from the Peyer’s patch to the surrounding mucosa. This mechanism implies so far unrecognized regional specification of Peyer’s-patch-primed T cells. Our findings may impact treatment strategies to avoid intestinal inflammation or foster immunity after oral vaccination.