22455 research outputs found
Sort by
Exploring the interactions of engineered antibodies with FcRn and TRIM21 for new avenues in antibody design, analysis, and gene therapy applications
Get PDFMonoclonal antibodies (mAbs) have emerged as pivotal therapeutic agents, with their effectiveness hinging on complex pharmacokinetic properties and interactions with immune receptors. This thesis investigates the nuanced interplay between mAbs and two key receptors: the neonatal Fc receptor (FcRn) and the tripartite motif-containing protein 21 (TRIM21). FcRn is known to affect the serum half-life of mAbs, while TRIM21 is involved in antibody-dependent intracellular neutralization (ADIN). Research has often focused on the IgG-FcRn affinity, neglecting the combined impact of both affinity and avidity, as well as the potential role of TRIM21 in antiviral therapy through Fc engineering. The research presented in this thesis aims to deepen our understanding of mAb interactions with FcRn and TRIM21, focusing on the mechanisms that govern these interactions. It employs advanced methodologies to elucidate the relationship between affinity and avidity. Specifically, it investigates the impact of Fc modifications that alter interactions with FcRn, affecting serum half-life, and with TRIM21, which is involved in viral neutralization within cells. The findings aim to guide the development of more effective therapeutic antibodies, with broad implications for the treatment of diseases.
In the publication titled 'Insight into the Avidity-Affinity Relationship of the Bivalent, pH-Dependent Interaction Between IgG and FcRn,' we explore the intricate binding dynamics between IgG and FcRn, advancing beyond traditional analyses that consider only single-affinity interactions. Utilizing switchSENSE technology, which closely mimics the membrane orientation of FcRn, we conducted a comprehensive examination of both affinity and avidity across the broad endosomal pH spectrum (pH 5.8–7.4). Our findings reveal that the engineered IgG1-YTE (M252Y/S254T/T256E) variant demonstrates a critical affinity shift at pH 7.2, indicative of its enhanced design for FcRn interaction. It also exhibits a marked avidity switch at pH 6.2, which is absent at pH 7.4. This dual engagement capability distinguishes IgG1-YTE from the wild-type, demonstrating the impact of Fc engineering on binding properties. Our research emphasizes the importance of avidity in IgG recycling, which is dictated by the variable expression of FcRn and its higher density in endosomes, necessitating a 2:1 stoichiometry for an extended serum half-life. The switchSENSE platform emerges as a powerful analytical tool, superior to traditional surface plasmon resonance (SPR), capturing a full range of kinetic parameters and accurately differentiating between monovalent and bivalent binding modes. This methodological advancement is crucial for understanding the dynamics of IgG binding in physiological contexts. The superior binding characteristics of the YTE variant suggest improved pharmacokinetics, potentially leading to increased therapeutic efficacy through an optimized recycling mechanism. The variant's higher affinity and significant contribution to avidity, especially during endosomal acidification, result in more stable FcRn complex formation, a desirable feature for antibodies engineered for extended serum half-life. The findings confirm the importance of pH-dependent binding in antibody design and have significant implications for the development of antibodies with improved recycling and extended half-life. Future research will utilize switchSENSE to further explore molecular interactions in various antibody mutants and formats, aiming to refine FcRn-mediated recycling for next-generation antibody therapies.
In our publication titled 'TRIM21 and Fc-Engineered Antibodies: Decoding its Complex Antibody Binding Mode with Implications for Viral Neutralization, we explore the complex role of TRIM21 within the immune system, focusing on its interaction with Fc-engineered antibodies and the subsequent impact on viral neutralization. Utilizing a combination of biosensor assays, mass photometry, electron microscopy, and structural predictions, our study dissects the intricate binding dynamics between TRIM21 and various antibody Fc variants, revealing a novel binding mechanism that is pivotal for developing effective viral neutralization strategies. Our investigation employs optimized SPR assays to establish precise affinities and avidities, underscoring the importance of assay conditions in accurately analyzing interactions. We demonstrate that TRIM21 PRYSPRY domains (monomers) bind symmetrically to a single IgG Fc homodimer in a non-cooperative manner, adhering to a 2:1 stoichiometry. This symmetry is consistent with crystallographic evidence of TRIM21 PRYSPRY-IgG interactions. Significantly, we identify that Fc mutations, such as YTE and HH (T307H, N434H), reduce TRIM21 binding while enhancing interaction with FcRn in a pH-dependent manner. This dual effect underlines the complexity of antibody engineering, where mutations can differentially influence receptor interactions, which is crucial for optimizing antibody recycling and immune defense. The mutation Y436A within the Fc CH2-CH3 domain notably decreases the affinity to both FcRn and TRIM21, with the latter by 180-fold. This mutation also demonstrates a pronounced shift from micromolar affinity to nanomolar avidity and the critical role of bivalent engagement. Our structural analysis indicates that TRIM21 undergoes a dynamic rearrangement upon Fc binding, likely influencing its immune function. We propose a novel two-step binding mechanism whereby TRIM21's initial attachment to an Fc site facilitates a conformational change, enhancing its interaction with a second Fc site through increased PRYSPRY domain mobility. This process significantly boosts avidity, which is essential for effective antibody function. Using adeno-associated virus (AAV) as a model, our findings suggest that antibody clustering on the virus is possible and could activate TRIM21's E3 ligase activity, demonstrated as a crucial step in virus neutralization. These insights are vital for advancing antibody engineering and understanding TRIM21’s role in immune responses, offering significant implications for therapeutic applications
Kognitive Bias bei Depressionen im Kindes- und Jugendalter und bei Kindern depressiver Eltern
Get PDFUntersuchung des Einflusses von Antibiotic Stewardship Interventionen auf das Verordnungsverhalten von systemischen Antibiotika in der Chirurgie
Get PDFStatistical analyses of combinatorial effects in high-throughput biological data
Get PDFDas Aufkommen von großen Mengen an biologischen Zähldaten durch Hochdurchsatz-Technologien hat die Entwicklung geeigneter statistischer Methoden zu einer wichtigen Herausforderung moderner interdisziplinärer Forschung gemacht. Diese Daten weisen oft eine Vielzahl von Kovariablen auf, sind jedoch durch geringe Beobachtungsgrößen und experimentelles Rauschen limitiert. Eine zentrale Forschungsfrage in datengetriebenen Untersuchungen ist, wie biologische Kovariablen eine relevante Zielvariable beeinflussen.
Meist sind nur einige der Kovariablen von Bedeutung. Diese können jedoch auf komplexe Art und Weise miteinander interagieren. Eine Hauptaufgabe besteht daher darin, die relevanten Effekte aus einer Vielzahl an möglichen Kombinationen zu identifizieren.
In dieser Arbeit habe ich Methoden entwickelt, die robuste Schätzungen von Interaktionseffekten durch quadratische Regressionsmodelle ermöglichen. Diese Methoden sind sowohl für Beobachtungs- als auch für experimentelle Daten geeignet, unabhängig davon, ob die experimentellen Designs vollständig sind. Die entwickelten Modelle berücksichtigen verschiedene Arten von biologischen Zähldaten: (i) quantitative Zähldaten, (ii) binäre Daten und (iii) relative Zähldaten, auch bekannt als kompositionelle Daten. Um in Szenarien mit mehr Kovariablen als Beobachtungen sowie in niedrigdimensionalen Szenarien interpretierbare Modelle zu entwickeln, habe ich in meinen Ansätzen Penalisierung verwendet.
Durch die Integration von Konzepten der hierarchisch Interaktionsmodellierung und der stabilitätsbasierten Modellselektion wird die Interpretierbarkeit gewährleistet. Zur Reduktion ungewollter, auf technisches und biologisches Rauschen zurückzuführender Effekte sind Ansätze entwickelt worden, die weniger anfällig für Ausreißer sind. Dies ist von besonderer Bedeutung, wenn nur wenige und inkonsistente Replikate vorliegen.
In meinem ersten Projekt habe ich Daten der Affinitätsreinigung von Nukleosomen mit quantitativer Proteomik und hierarchischer Interaktionsmodellierung kombiniert. Ziel war es, die kombinatorischen Effekte bestimmter Chromatinmodifikationen auf die Proteinrekrutierung in einem unvollständigen experimentellen Design zu schätzen. Der hierfür entwickelte Workflow, asteRIa, ermöglicht eine stabile Schätzung robuster Interaktionen zwischen Chromatinmodifikationen und hat mehrere Proteine als epigenetische Leser-Kandidaten identifiziert.
In meinem zweiten Projekt habe ich ein generisches quadratisches Interaktionsmodell entwickelt, um Umwelt- oder Wirtsbedingungen aus Daten über die mikrobielle Abundanz vorherzusagen. Dieses Modell unterstützt verschiedene Datenmodalitäten und hat einen breiten Anwendbarkeitsbereich. Diesen habe ich auf unterschiedlichen Daten demonstriert und robuste Interaktionseffekte zwischen mikrobiellen Taxa aufgedeckt.
In meinem dritten Projekt habe ich Wechselwirkungen von Medikamenten in Hochdurchsatz-Screening-Verfahren für einzelne Zellen analysiert. Dabei habe ich hierarchische Interaktionsmodellierung mit einer Optimierungstechnik kombiniert, die robust gegenüber Ausreißern ist, und somit einen generischen und reproduzierbaren Workflow erstellt.
Insgesamt habe ich statistische Methoden zur Schätzung robuster Interaktionseffekte in biologischen Daten entwickelt. Die Modelle ermöglichen präzise Analysen verschiedener Datentypen und identifizieren Interaktionseffekte, die Hypothesen für weiterführende funktionelle Untersuchungen darstellen.The advent of large-scale biological count data from high-throughput technologies has made the development of suitable statistical techniques a cornerstone of modern inter-disciplinary research. These data often contain many features but limited sample size, and are accompanied by experimental noise. A common research question in data-driven observational studies is to determine how such biological features impact a readout of interest. Typically, only a subset of features is relevant, and they may interact in a con- certed fashion. Thus, a major concern is to identify these relevant effects from a large number of possible combinations of features.
In this thesis, I developed and evaluated ways to estimate stable main and interaction effects via quadratic regression models in both observational and experimental data with complete or incomplete designs. The models developed are applicable to different data modalities in which biological count data typically appear: (i) quantitative count data, (ii) presence-absence data, and (iii) relative count data, also known as compositional data. To derive parsimonious models in underdetermined regimes, as well as in low- and moderate-dimensional settings, I implemented the models under penalization. To facilitate interpretability, I included the concept of hierarchy in interaction modeling and stability-based model selection. In order to account for technical and biological noise in the data, I introduced ways to be less sensitive towards outliers, especially when few and inconsistent replicates are available.
In my first project, I integrated nucleosome affinity purification data with high-throughput quantitative proteomics and hierarchical interaction modeling to estimate combinatorial effects of the presence or absence of certain chromatin modifications on protein recruit- ment within an incomplete experimental design study. This is facilitated by the computa- tional workflow asteRIa which combines hierarchical interaction modeling, stability-based model selection, and replicate consistency checks for a stable estimation of robust inter- actions among chromatin modifications. asteRIa identifies several epigenetic “reader” candidate proteins responding to specific interactions between chromatin modifications. In my second project, I developed a generic quadratic interaction model for the prediction of environmental or host-related conditions from observational and experimental micro- bial abundance data. The interaction model covers common data modalities of microbial data, ranging from quantitative microbiome and presence-absence information to com- positional microbiome data. I demonstrated the broad applicability of our framework across various ecosystems and showcased how quadratic models improve predictive accu- racy while uncovering stable interaction effects between microbial taxa when integrated with hierarchical interaction modeling and stability-based model selection.
In my third project, I analyzed drug interaction effects in high-content screening (HCS) cell studies. Here, I combined hierarchical interaction modeling with an optimization that is less sensitive to outliers within a generally applicable and reproducible computational workflow to analyze combinatorial effects in HCS data.
In summary, I have developed statistical approaches for the stable estimation of interaction effects, with a particular emphasis on high-throughput biological data. The workflows and statistical models I developed enable the precise analysis of various data types to reveal highly stable interaction effects, facilitating further functional analyses
Herausforderungen in der Elimination und Therapie der Hepatitis C Virus-Infektion im Zeitalter direkt antiviraler Substanzen (DAAs)
Get PDF
Evidenzbasierte Belastungseinschätzung und Refinement: Haltungsformen von Labormäusen und perioperative Analgesie für murine Kraniotomien
Get PDF
