Bone represents a common site of metastases for several solid tumors. However, the ability of neuroendocrine neoplasms (NENs) to localize to bone has always been considered a rare and late event. Thanks to the improvement of therapeutic options, which results in longer survival, and of imaging techniques, particularly after the introduction of positron emission tomography (PET) with gallium peptides, the diagnosis of bone metastases (BMs) in NENs is increasing. The onset of BMs can be associated with severe skeletal complications that impair the patient's quality of life. Moreover, BMs negatively affect the prognosis of NEN patients, bringing out the lack of curative treatment options for advanced NENs. The current knowledge on BMs in gastro-entero-pancreatic (GEP) and bronchopulmonary (BP) NENs is still scant and is derived from a few retrospective studies and case reports. This review aims to perform a critical analysis of the evidence regarding the role of BMs in GEP- and BP-NENs, focusing on the molecular mechanisms underlining the development of BMs, as well as clinical presentation, diagnosis, and treatment of BMs, in an attempt to provide suggestions that can be used in clinical practice.

Among the defense strategies developed in microbes over millions of years, the innate adaptive CRISPR-Cas immune systems have spread across most of bacteria and archaea. The flexibility, simplicity, and specificity of CRISPR-Cas systems have laid the foundation for CRISPR-based genetic tools. Yet, the efficient administration of CRISPR-based tools demands rational designs to maximize the on-target efficiency and off-target specificity. Specifically, the selection of guide RNAs (gRNAs), which play a crucial role in the target recognition of CRISPR-Cas systems, is non-trivial. Despite the fact that the emerging machine learning techniques provide a solution to aid in gRNA design with prediction algorithms, design rules for many CRISPR-Cas systems are ill-defined, hindering their broader applications. CRISPR interference (CRISPRi), an alternative gene silencing technique using a catalytically dead Cas protein to interfere with transcription, is a leading technique in bacteria for functional interrogation, pathway manipulation, and genome-wide screens. Although the application is promising, it also is hindered by under-investigated design rules. Therefore, in this work, I develop a state-of-art predictive machine learning model for guide silencing efficiency in bacteria leveraging the advantages of feature engineering, data integration, interpretable AI, and automated machine learning. I first systematically investigate the influential factors that attribute to the extent of depletion in multiple CRISPRi genome-wide essentiality screens in Escherichia coli and demonstrate the surprising dominant contribution of gene-specific effects, such as gene expression level. These observations allowed me to segregate the confounding gene-specific effects using a mixed-effect random forest (MERF) model to provide a better estimate of guide efficiency, together with the improvement led by integrating multiple screens. The MERF model outperformed existing tools in an independent high-throughput saturating screen. I next interpret the predictive model to extract the design rules for robust gene silencing, such as the preference for cytosine and disfavoring for guanine and thymine within and around the protospacer adjacent motif (PAM) sequence. I further incorporated the MERF model in a web-based tool that is freely accessible at www.ciao.helmholtz-hiri.de. When comparing the MERF model with existing tools, the performance of the alternative gRNA design tool optimized for CRISPRi in eukaryotes when applied to bacteria was far from satisfying, questioning the robustness of prediction algorithms across organisms. In addition, the CRISPR-Cas systems exhibit diverse mechanisms albeit with some similarities. The captured predictive patterns from one dataset thereby are at risk of poor generalization when applied across organisms and CRISPR-Cas techniques. To fill the gap, the machine learning approach I present here for CRISPRi could serve as a blueprint for the effective development of prediction algorithms for specific organisms or CRISPR-Cas systems of interest. The explicit workflow includes three principle steps: 1) accommodating the feature set for the CRISPR-Cas system or technique; 2) optimizing a machine learning model using automated machine learning; 3) explaining the model using interpretable AI. To illustrate the applicability of the workflow and diversity of results when applied across different bacteria and CRISPR-Cas systems, I have applied this workflow to analyze three distinct CRISPR-Cas genome-wide screens. From the CRISPR base editor essentiality screen in E. coli, I have determined the PAM preference and sequence context in the editing window for efficient editing, such as A at the 2nd position of PAM, A/TT/TG downstream of PAM, and TC at the 4th to 5th position of gRNAs. From the CRISPR-Cas13a screen in E. coli, in addition to the strong correlation with the guide depletion, the target expression level is the strongest predictor in the model, supporting it as a main determinant of the activation of Cas13-induced immunity and better characterizing the CRISPR-Cas13 system. From the CRISPR-Cas12a screen in Klebsiella pneumoniae, I have extracted the design rules for robust antimicrobial activity across K. pneumoniae strains and provided a predictive algorithm for gRNA design, facilitating CRISPR-Cas12a as an alternative technique to tackle antibiotic resistance. Overall, this thesis presents an accurate prediction algorithm for CRISPRi guide efficiency in bacteria, providing insights into the determinants of efficient silencing and guide designs. The systematic exploration has led to a robust machine learning approach for effective model development in other bacteria and CRISPR-Cas systems. Applying the approach in the analysis of independent CRISPR-Cas screens not only sheds light on the design rules but also the mechanisms of the CRISPR-Cas systems. Together, I demonstrate that applied machine learning paves the way to a deeper understanding and a broader application of CRISPR-Cas systems.

The immune system is responsible for the preservation of homeostasis whenever a given organism is exposed to distinct kinds of perturbations. Given the complexity of certain organisms like mammals, and the diverse types of challenges that they encounter (e.g. infection or disease), the immune system evolved to harbor a great variety of distinct immune cell populations with specialized functions. For instance, the family of T cells is sub-divided into conventional (Tconv) and unconventional T cells (UTCs). Tconv form part of the adaptive arm of the immune system and are comprised of αβ CD4+ or CD8+ cells that differentiate from naïve to effector and memory populations upon activation and are essential during infection and cancer. Furthermore, UTCs, which include γδ T cells, NKT and MAIT, are involved in innate and adaptive immune responses, due to their dual mode of activation, through cytokines (innate-like) or TCR (adaptive), and function. Despite our understanding of the basic functions of T cells in several contexts, a great number of open questions related to their basic biology remain. For instance, the mechanism behind the differentiation of naïve CD4+ and CD8+ T cells into effector and memory populations is not fully understood. Moreover, the exact function and relevance of distinct UTC subpopulations in a physiological context have not been fully clarified. Here, we investigated the factors mediating naïve CD8+ T cell differentiation into effector and memory cells. By using flow cytometry, mass spectrometry, enzymatic assays, and transgenic mouse models, we found that the membrane bound enzyme sphingomyelin-phosphodiesterase acid-like 3b (Smpdl3b) is crucial for the maintenance of memory CD8+ T cells. Our data show that the absence of Smpdl3b leads to diminished CD8+ T cell memory, and a loss of stem-like memory populations due to an aggravated contraction. Our scRNA-seq data suggest that Smpdl3b could be involved in clathrinmediated endocytosis through modulation of Huntingtin interacting protein 1 (Hip1) levels, likely regulating TCR-independent signaling events. Furthermore, in this study we explored the role of UTCs in lymph node-specific immune responses. By using transgenic mouse models for photolabeling, lymph node transplantation models, infection models and flow cytometry, we demonstrate that S1P regulates the migration of tissue-derived UTC from tissues to draining lymph nodes, resulting in heterogeneous immune responses mounted by lymph nodes draining different tissues. Moreover, our unbiased scRNAseq and single lineage-deficient mouse models analysis revealed that all UTC lineages (γδ T cells, NKT and MAIT) are organized in functional units, based on transcriptional homogeneity, shared microanatomical location and migratory behavior, and numerical and functional redundancy. Taken together, our studies describe additional cell intrinsic (Smpdl3b) and extrinsic (S1Pmediated migration) functions of sphingolipid metabolism modulating T cell biology. We propose the S1P/S1PR1/5 signaling axis as the potential survival pathway for Smpdl3b+ memory CD8+ T cells and UTCs, mainly in lymph nodes. Possibly, Smpdl3b regulates S1P/S1PR signaling by balancing ligandreceptor endocytosis, while UTCs migrate to lymph nodes during homeostasis to be exposed to specific levels of S1P that assure their maintenance. Our results are clinically relevant, since several drugs modulating the S1P/S1PR signaling axis or the levels of Smpdl3b are currently used to treat human diseases, such as multiple sclerosis and B cell-mediated diseases. We hope that our discoveries will inspire future studies focusing on sphingolipid metabolism in immune cell biology.

Besides their central role in haemostasis and thrombosis, platelets are increasingly recognised as versatile effector cells in inflammation, the innate and adaptive immune response, extracellular matrix reorganisation and fibrosis, maintenance of barrier and organ integrity, and host response to pathogens. These platelet functions, referred to as thrombo-inflammation and immunothrombosis, have gained major attention in the COVID-19 pandemic, where patients develop an inflammatory disease state with severe and life-threatening thromboembolic complications. In the CRC/TR 240, a highly interdisciplinary team of basic, translational and clinical scientists explored these emerging roles of platelets with the aim to develop novel treatment concepts for cardiovascular disorders and beyond. We have i) unravelled mechanisms leading to life-threatening thromboembolic complica-tions following vaccination against SARS-CoV-2 with adenoviral vector-based vaccines, ii) identified unrecognised functions of platelet receptors and their regulation, offering new potential targets for pharmacological intervention and iii) developed new methodology to study the biology of megakar-yocytes (MKs), the precursor cells of platelets in the bone marrow, which lay the foundation for the modulation of platelet biogenesis and function. The projects of the CRC/TR 240 built on the unique expertise of our research network and focussed on the following complementary fields: (A) Cell bi-ology of megakaryocytes and platelets and (B) Platelets as regulators and effectors in disease. To achieve this aim, we followed a comprehensive approach starting out from in vitro systems and animal models to clinical research with large prospective patient cohorts and data-/biobanking. Despite the comparably short funding period the CRC/TR 240 discovered basic new mechanisms of platelet biogenesis, signal transduction and effector function and identified potential MK/platelet-specific molecular targets for diagnosis and therapy of thrombotic, haemorrhagic and thrombo-inflammatory disease states.

Bulk RNA sequencing technologies have provided invaluable insights into host and bacterial gene expression and associated regulatory networks. Nevertheless, the majority of these approaches report average expression across cell populations, hiding the true underlying expression patterns that are often heterogeneous in nature. Due to technical advances, single-cell transcriptomics in bacteria has recently become reality, allowing exploration of these heterogeneous populations, which are often the result of environmental changes and stressors. In this work, we have improved our previously published bacterial single-cell RNA sequencing (scRNA-seq) protocol that is based on multiple annealing and deoxycytidine (dC) tailing-based quantitative scRNA-seq (MATQ-seq), achieving a higher throughput through the integration of automation. We also selected a more efficient reverse transcriptase, which led to reduced cell loss and higher workflow robustness. Moreover, we successfully implemented a Cas9-based rRNA depletion protocol into the MATQ-seq workflow. Applying our improved protocol on a large set of single Salmonella cells sampled over different growth conditions revealed improved gene coverage and a higher gene detection limit compared to our original protocol and allowed us to detect the expression of small regulatory RNAs, such as GcvB or CsrB at a single-cell level. In addition, we confirmed previously described phenotypic heterogeneity in Salmonella in regard to expression of pathogenicity-associated genes. Overall, the low percentage of cell loss and high gene detection limit makes the improved MATQ-seq protocol particularly well suited for studies with limited input material, such as analysis of small bacterial populations in host niches or intracellular bacteria. IMPORTANCE: Gene expression heterogeneity among isogenic bacteria is linked to clinically relevant scenarios, like biofilm formation and antibiotic tolerance. The recent development of bacterial single-cell RNA sequencing (scRNA-seq) enables the study of cell-to-cell variability in bacterial populations and the mechanisms underlying these phenomena. Here, we report a scRNA-seq workflow based on MATQ-seq with increased robustness, reduced cell loss, and improved transcript capture rate and gene coverage. Use of a more efficient reverse transcriptase and the integration of an rRNA depletion step, which can be adapted to other bacterial single-cell workflows, was instrumental for these improvements. Applying the protocol to the foodborne pathogen Salmonella, we confirmed transcriptional heterogeneity across and within different growth phases and demonstrated that our workflow captures small regulatory RNAs at a single-cell level. Due to low cell loss and high transcript capture rates, this protocol is uniquely suited for experimental settings in which the starting material is limited, such as infected tissues.