13979 research outputs found
Sort by
Leitlinien für den Betrieb des Publikationsservice der TU Darmstadt – TUprints
Get PDFDer Publikationsservice der Technischen Universität Darmstadt - TUprints - bietet die organisatorischen
und technischen Rahmenbedingungen zur elektronischen Publikation wissenschaftlicher Dokumente im
Sinne des offenen und freien Zugangs zu Wissenschaft, Lehre und Forschung im Internet
Single-cell Multi-omics Dissection of CAR T cells Reveals Inhibitors of Gene Delivery Mediated by Lentiviral Vectors
Get PDFCAR T cell therapy has shown promising results in clinical trials and has been approved for use in certain types of cancer, specifically B-cell lymphoblastic leukemia, and large B-cell lymphoma. However, as a relatively new treatment, there are still some challenges to overcome. These include improving efficacy, managing side effects, and making the treatment more widely available and affordable. Lentiviral vectors (LVs) are considered the gold standard tool in the generation of CAR T cells due to their ability to efficiently deliver genetic material into T cells. Several CAR T cell products currently on the market use LVs for CAR T cell generation. During the manufacturing process, T cells are collected from the patient and genetically modified to express the CAR. The modified T cells are then expanded before being infused back into the patient. The amount of CAR T cells in the final product can be influenced by factors such as the purity and quality of the starting T cells, the efficiency of the gene transfer process, and the expansion protocol. Furthermore, the complex and multifactorial manufacturing process of CAR T cells results in a heterogeneous population of which only a fraction are genetically modified T cells.
Among LVs conventional VSV-LV is most frequently used. Besides, T cell targeted vectors using CD3, CD4, and CD8 as entry receptors have been developed. These targeted LVs enable the selective transduction of T cell subtypes, including CD3, CD4, and CD8 T cells. Employing these targeted vectors during in vitro transduction reduces the complexity of manufacturing processes, omits the risk of off target modifications, and holds potential for use in direct in vivo gene transfer applications. Limited information is available about the molecular consequences of vector exposure to T lymphocytes, and the characteristics of CAR T cells generated from the listed types of lentiviral vectors. Therefore, additional evaluation is needed to compare and characterize the phenotype and transcriptome of CAR T cells produced with these targeted vectors compared to those generated with VSV-LV.
T cell transduction is initiated by receptor binding of the LVs followed by endocytosis for VSV-LV and fusion at the cell membrane for the targeted vectors. Whether using conventional lentiviral vectors or targeted vectors, it was unclear whether all cells that bind to LVs would be transduced, or whether there would be cell populations that attach to LVs but do not express the CAR. To address this question, vector binding and CAR expression on T cells were monitored at various time points following the addition of LVs, by using specific antibodies against CAR and LVs. Analysis of vector binding and CAR expression on T cells at different time points after the addition of vector particles revealed that both VSV-LV and targeted vectors attachment rates to T cells were consistently above 60% at 4 hours after vector addition. In contrast, CAR expression 72 hours after adding the vectors was significantly lower than the percentage of vector binding, and it varied between donors and LVs. This thesis aimed at understanding the status and gene expression profiles of cells negative for CAR expression compared to successfully transduced cells that convert into CAR T cells.
For these purposes, single-cell multi-omics analysis was established. CAR T cells generated from two healthy donors were analyzed at early time points after exposure to CD3-LV, CD4-LV, CD8-LV, and VSV-LV. Unlike other single-cell studies that have analyzed CAR T cells cultured for extended periods before infusion, this thesis focused on early time points after vector exposure, analyzing the cells 3 days after encountering lentiviral vectors.
For single-cell analysis, a T cell targeted gene panel, together with primers for the detection of CAR mRNA, was utilized. This method provided sensitive detection of transcripts and allowed for high-resolution differential gene expression analysis, while requiring less sequencing depth. Besides detecting mRNA, this study included barcoded antibodies (AbSeqs) to detect LV binding, CAR, and 30 immune markers in single-cell analysis. AbSeqs were effectively detected and utilized to differentiate various populations of CD4 and CD8 T cells, as well as for the detection of vector bound and CARpos T cells. All AbSeqs showed a bimodal distribution, enabling the separation of cells into positive and negative populations within single-cell data. The percentage of positive cells, as defined by the AbSeqs, was well in line with protein expression data obtained through flow cytometry. This comparison confirmed the validity of the AbSeq data. Based on the AbSeq and transcriptome data, different cell clusters were identified across all samples. Specifically, the CD4-LV sample contained a high proportion of CD4+ cytotoxic cells, and CD8-LV sample predominantly CD8+ naive T cells. The CD3-LV group included the highest percentage of CD4+ central memory T cells, and the VSV-LV sample was notable for the highest frequency of CD4+ effector memory T cells.
When comparing the transcriptome of CAR T cells generated with the targeted LVs to those generated with VSV-LV, distinct gene expression profiles were observed in both CD4, and CD8 CAR T cells derived from each vector type. Specifically, LEF1 and FAM65B, genes that promote a naive phenotype in T cells, were universally upregulated in CD4 and CD8 CAR T cells from targeted vectors. Despite the expression of naive T cell markers both CD4 and CD8 CAR T cells from targeted vectors showed upregulation of cytotoxic markers. Notably, CD4 CAR T cells derived from CD4-LV displayed characteristics of cytotoxic CD4 T cells, marked by high expression of GZMA. These observations indicated that CAR T cells, generated from targeted vectors, had a less differentiated phenotype, while maintaining cytotoxic activity.
Focusing on the heterogeneous population of T cells present during CAR T cell generation, specifically in terms of vector binding and CAR expression, an analysis was conducted on the differentially expressed genes between CARneg T cells with vector binding signal compared to CARpos T cell. In this analysis, a group of antiviral genes stimulated by type I interferon were detected to be upregulated in CARneg T cells. These genes included IFITM2, IFITM3, STAT1, SAMHD1, MX1, IFI6, ADAR, OAS2, TRIM22, TRIM25, and APOEC3G. To discern whether the expression of these genes was induced in T cells after exposure to the LVs, or if it was high in T cells in advance, the expression patterns among different cell subsets were investigated. The expression of these genes, specifically for CD4-LV and CD8-LV treated samples, followed a general pattern: it was low in cells that were negative for both CAR and vector binding (indicating cells that had not encountered the vector particles), while it was high in cells that were CARneg but positive for vector binding, and low in CARpos T cells. This pattern of expression suggested that the expression of interferon stimulated genes was upregulated in T cells after encountering viral vectors.
The examination of the genes commonly upregulated in CARpos T cells, irrespective of the vector used, revealed a consistent upregulation of CSF2, CTLA4, MYB, LAG3, JUNB, and CD2, which are known for their role in the activation, differentiation, and exhaustion of T cells. Additionally, other genes with less understood functions in CAR T cells were also upregulated, including DUSP2, DUSP4, RPDM1, and ZBED2.
To validate the observations made in two healthy donors, single-cell data from anti-CD19 CAR T cell infusion products of 24 patients diagnosed with LBCL were analyzed. The differential expression of genes between CARpos and CARneg T cells confirmed the observations made in healthy donors. Specifically, upregulation of certain interferon-stimulated genes, including IFITM1, IFITM2, and ISG20, in CARneg T cells, as well as an upregulation of DUSP2 and DUSP4 genes in CARpos T cells was detected, despite several days longer cultivation of these cells before analysis.
Considering the upregulation of interferon-stimulated genes in CARneg T cells in both healthy donors and patient’s samples, it was hypothesized that their antiviral activity might be involved in inhibiting LV-mediated gene transfer into T cells. Furthermore, the upregulation of DUSP2 and DUSP4 in CARpos T cells was suggested to play a positive role in T cell transduction.
Based on these observations, various approaches were investigated to either downmodulate interferon signaling or directly inhibit interferon-induced transmembrane proteins (IFITMs).
First, downmodulation of interferon signaling by deucravacitinib (BMS-986165), an inhibitor of TYK2, was investigated. TYK2 is a key kinase at the start of interferon signaling. Its activation leads to the expression of interferon stimulated genes. deucravacitinib (BMS-986165) treatment resulted in increasing CAR expression with up to 3-fold increase in transduction mediated by all targeted vectors and VSV-LV without compromising T cell viability. Although different degrees of enhancement were observed among different LVs and donors.
Next, considering that IFITM genes were commonly upregulated in CARneg T cells of healthy donors and patient’s samples, the direct inhibition of IFITMs with caraphenol A was investigated. Caraphenol A is a resveratrol trimer that has been shown to transiently reduce IFITM protein expression. In this thesis, caraphenol A treatment enhanced gene transfer by targeted LVs up to 2-fold with a less pronounced effect on VSV-LV.
As a final approach to target interferon signaling, ERK inhibitors were utilized to mimic the activity of DUSP2 and DUSP4 proteins in T cells. Upregulation of DUSP2 and DUSP4 were observed in CARpos T cells. This led to the hypothesis that these genes might play a role in the downregulation of interferon-stimulated gene expression by dephosphorylating MAPK1, thereby promoting T cell transduction. Treating T cells with four different ERK inhibitors resulted in increased CAR expression in T cells. The response to various doses of each type of ERK inhibitor varied based on the vector type used for T cell transduction,
but CD4-LV, CD8-LV, and VSV-LV showed an up to a 2-fold increase, while CD3-LV demonstrated an up to a 3-fold increase in response to ERK inhibitor treatment.
In conclusion, by employing single-cell multi-omics, this thesis highlighted the impact of different LVs, on the transcriptomic profiles and the phenotypes of the generated CAR T cells. Furthermore, the heterogeneity observed in CAR T cell products were found to be modulated by interferon-stimulated genes that are induced in T cells by LVs. Some of these genes were identified as inhibitors of CAR T cell generation and could be observed in both healthy donors and patient samples. Moreover, a novel role for the DUSP2 gene in promoting T cell transduction was proposed. The use of different ERK inhibitors, which mimic DUPS2 function, improved transduction with both targeted and VSV-LV vectors, providing the first evidence for this hypothesis and could be used to optimize CAR-T cell production
On interactivity in probabilistic pragmatics: yet another rational analysis of scalar implicatures
Get PDFProbabilistic pragmatics follows the recent trend in cognitive science to understand cognition as Bayesian inference and rational decision making under uncertainty (Chater, Tenenbaum, and Yuille 2006; Tenenbaum et al. 2011). As the insights, tools, and ideas that drive this trend may also be useful for linguists, Franke and Jäger (this volume), henceforce F&J, draw attention to these developments and give a timely introduction to probabilistic pragmatics by showcasing some substantive examples (see also Goodman and Lassiter 2015)
Information Leakage Attacks and Defenses in the IoT Domain
Get PDFThe Internet of Things (IoT) market is experiencing rapid growth as the market is projected to double between 2020 and 2025. The surge in the adoption of IoT technologies into devices of daily life, from smartphones to satellite communication, highlights significant concerns about their security. Even with heightened security measures, vulnerabilities persist due to oversight of potential attack surfaces and scenarios, leaving entire systems vulnerable to the risk of information leakage. Therefore, numerous challenges within this field must still be addressed. For that reason, it is crucial for manufacturers and users to understand emerging attack surfaces resulting from the IoT field. IoT devices can leak information through adversarial exploitation of novel human-computer interfaces, such as voice or touch, which are often incorporated into smart speakers and smartphones.
The analysis of such attack vectors plays a pivotal role in uncovering vulnerabilities across the field of IoT devices, thereby paving the way for mitigations to increase security levels to prevent information leakage. This goal can be achieved by strengthening User Interfaces (UIs) or protecting private user data by making it inaccessible to the adversary.
In this dissertation, we design, implement and evaluate novel approaches to exploit the attack surfaces of different IoT devices and to offer mitigations of exploits for upcoming technologies in the area of IoT devices. Specifically, we propose 1) novel attacks on smart speaker virtual assistant’s interaction model using ultrasonic audio, 2) improvements for virtual assistant’s wake-word detectors and means to detect malicious wake-word activated devices, 3) a novel attack on smart devices’ capacitive touch screens using airborne electro-magnetic-interference and 4) protection against location leakage of satellite internet users using IoT long-range radio network technology
The impact of land use on the acoustic behaviour of cicadas in the Chocó lowland tropical forest of Ecuador
Get PDF1. The biodiversity of tropical rainforests is under extreme pressure due to the expansion of agricultural land. Beyond the immediate risk of species extinction, the intensification of land use can alter species' behaviour with consequences for the entire ecosystem.
2. In this study we investigated the impact of land use on the acoustic behaviour of cicadas in the highly biodiverse Ecuadorian Chocó region. We used passive acoustic monitoring (PAM) for the collection of audio data, from which we identified and analysed the sound activity of cicadas and the structure of daily patterns along a chronosequence of forest recovery. At landscape scale we studied the impact of a surrounding either dominated by agricultural land use or forests on the acoustic behaviour of cicadas.
3. Cicada sound activity was significantly lower in active agriculture compared to undisturbed old‐growth forest and increased along the forest recovery gradient. The diurnal pattern changed from simple in active agriculture to more complex and highly synchronized along the recovery gradient towards old‐growth forests. A surrounding dominated by agricultural land use additionally reduced the sound activity of cicadas and simplified the diurnal pattern in old‐growth forests.
4. Taken together, agricultural land use at local and landscape scales affects overall activity, diurnal patterns and synchrony of vocalizing song cicadas. This is a concerning trend considering the direct link between chorusing and fitness for cicadas. However, mature restoration forests embedded in forest dominated landscape surroundings showed restored cicada song behaviours similar to those of old‐growth forests, which underlines the conservation value of advanced secondary forests and the importance to support forest regeneration in the tropics
Flexible development and evaluation of machine‐learning‐supported optimal control and estimation methods via HILO‐MPC
Get PDFModel‐based optimization approaches for monitoring and control, such as model predictive control and optimal state and parameter estimation, have been used successfully for decades in many engineering applications. Models describing the dynamics, constraints, and desired performance criteria are fundamental to model‐based approaches. Thanks to recent technological advancements in digitalization, machine‐learning methods such as deep learning, and computing power, there has been an increasing interest in using machine learning methods alongside model‐based approaches for control and estimation. The number of new methods and theoretical findings using machine learning for model‐based control and optimization is increasing rapidly. However, there are no easy‐to‐use, flexible, and freely available open‐source tools that support the development and straightforward solution to these problems. This article outlines the basic ideas and principles behind an easy‐to‐use Python toolbox that allows to solve machine‐learning‐supported optimization, model predictive control, and estimation problems quickly and efficiently. The toolbox leverages state‐of‐the‐art machine learning libraries to train components used to define the problem. Machine learning can be used for a broad spectrum of problems, ranging from model predictive control for stabilization, set point tracking, path following, and trajectory tracking to moving horizon estimation and Kalman filtering. For linear systems, it enables quick generation of code for embedded model predictive control applications. HILO‐MPC is flexible and adaptable, making it especially suitable for research and fundamental development tasks. Due to its simplicity and numerous already implemented examples, it is also a powerful teaching tool. The usability is underlined, presenting a series of application examples
Breaking it down, to build it back up: Attacks and Defenses for RPKI
Get PDFThe Border Gateway Protocol (BGP) is the glue that holds the Internet together and enables packets to reach their destinations. However, BGP is not secure by design. It is vulnerable to hijacking attacks and route leaks, and the community has tried for decades to find a solution for this design error.
The Resource Public Key Infrastructure (RPKI) has emerged as the only currently feasible solution to BGP's woes. It is an intuitive, flexible infrastructure that allows any BGP security protocol that relies on distributed, cryptographically verifiable data, to get incorporated and effectively deployed across BGP routers. RPKI already covers over 50% of network prefixes and is deployed by at least 27% of networks in the world. It has already proven its benefits over the past few years due to many BGP hijacks, which went unnoticed by those deploying RPKI, but caused severe consequences for those who didn't. RPKI has proven itself so successful, that the Federal Communications Commission (FCC) published a recommendation on routing security, where they suggested mandating the use of RPKI for all major ISP providers in the US. However, not all that glitters is gold. While RPKI is an excellent approach to solving the security issues of BGP, it is not perfect.
In this work, the author evaluates the security of the RPKI ecosystem as a whole, and that of all RPKI software components individually. The author discovers a range of attacks that lead to the silent downgrade of RPKI protection, or the Denial-of-Service (DoS) of RPKI components, and evaluates current RPKI deployment practices only to discover trends that are concerning when extrapolated to full RPKI deployment. Finally, this work also provides the first attempt to mitigate all above mentioned RPKI issues through a distributed infrastructure that enhances RPKI component security and efficiency, and is backwards compatible with the current RPKI environment.
This thesis is based on work published in 6 full papers and 2 posters in international academic conferences. This work resulted in the discovery of 18 vulnerabilities in RPKI code, and the issuance of 5 Common Vulnerabilities and Exposures (CVEs)
Calorimetry of extracellular vesicles fusion to single phospholipid membrane
No full textExtracellular vesicles (EVs)-mediated communication relies not only on the delivery of complex molecular cargoes as lipids, proteins, genetic material, and metabolites to their target cells but also on the modification of the cell surface local properties induced by the eventual fusion of EVs’ membranes with the cells’ plasma membrane. Here we applied scanning calorimetry to study the phase transition of single phospholipid (DMPC) monolamellar vesicles, investigating the thermodynamical effects caused by the fusion of doping amounts of mesenchymal stem cells-derived EVs. Specifically, we studied EVs-induced consequences on the lipids distributed in the differently curved membrane leaflets, having different density and order. The effect of EV components was found to be not homogeneous in the two leaflets, the inner (more disordered one) being mainly affected. Fusion resulted in phospholipid membrane flattening associated with lipid ordering, while the transition cooperativity, linked to membrane domains’ coexistence during the transition process, was decreased. Our results open new horizons for the investigation of the peculiar effects of EVs of different origins on target cell membrane properties and functionality
Elucidating the Reaction Mechanism and Deactivation of CO₂-Assisted Propane ODH over VOₓ/TiO₂ Catalysts: A Multiple Operando Spectroscopic Study
No full textThe CO₂-assisted oxidative dehydrogenation (ODH) of propane is of great technical importance and enables the use (and thus removal from the atmosphere) of CO₂, a greenhouse gas, in a value-adding process. Supported vanadium oxide (VOₓ) catalysts are a promising alternative to more active but toxic chromium oxide catalysts. Despite its common use, TiO₂ has not been investigated as a support material for VOₓ in the CO₂–ODH of propane. In this study, we elucidate the interaction between titania (P25) and vanadia in the reaction mechanism by analyzing the reaction network and investigating the catalyst using X-ray diffraction (XRD), multiwavelength Raman, UV–vis and diffuse reflectance IR Fourier transform (DRIFT) spectroscopy. Besides direct and indirect ODH reaction pathways, propane dry reforming (PDR) is identified as a side reaction, which is more prominent on bare titania. The presence of VOₓ enhances the stability and the selectivity toward propylene by participating in the redox cycle, activating CO₂ and leading to a higher rate of regeneration. Additionally, VOₓ catalyzes the conversion of anatase to rutile, which facilitates CO₂ activation, thereby leading to an encapsulation of vanadium. At higher loadings, reducible VOₓ oligomers are present on the surface, facilitating some PDR, but less than on bare P25. As the main deactivation mechanisms of the catalyst system, we propose the reduction of the titania lattice and the consumption of vanadium, while carbon formation appears to be less relevant. Our results highlight the importance of analyzing the CO₂–ODH reaction network and applying a multispectroscopic approach to obtain a detailed mechanistic understanding of CO₂-assisted propane ODH over supported VOₓ catalysts
Investigation on THz response of dielectric substrates for integration and packaging of direct THz detectors
Get PDFWithin this work, we investigated the response of dielectric materials used in high-frequency laminates, such as Rogers laminates and quartz substrates. A free-space, frequency-domain, non-destructive technique using a continuous-wave THz source was employed in these experiments to provide insights into the material parameters from the experimental results with a focus on the losses induced by these materials in the THz domain (0.02 to 2.5 THz). We extracted the material parameters from the experimental results. The results on the broadband response of materials provide valuable insights for designing novel frequency-selective passive components, which can be used for packaging and integration of THz detectors