Publikationer från Uppsala Universitet
Not a member yet
127603 research outputs found
Sort by
Computational Studies of Enzyme Thermodynamics and Protein Stability
No full textEnzymes adapt to varying environmental temperatures through a process often involving a fine balance between molecular flexibility and rigidity to maintain optimal catalytic activity. This thesis investigates the intricate mechanisms underlying enzyme temperature adaptation and protein stability using advanced computational methods. In Paper I, the aim was to investigate the general notion that psychrophilic (cold-adapted) enzymes exhibit enhanced catalytic activity at lower temperatures. Computational simulations of lactate dehydrogenases from fish species with differing thermal adaptations revealed that their increased low-temperature activity stems from a redistribution of action enthalpy and entropy favourable in low temperature environments. This advantageous thermodynamic profile was shown to be directly linked to enhanced conformational flexibility in specific surface regions of the enzyme. Paper II then explored whether enzyme oligomerization could serve as an additional strategy for temperature adaptation. Investigating a cold-adapted (R)-3-hydroxybutyrate dehydrogenase, empirical valence bond (EVB) simulations demonstrated that while oligomerization affects the mobility of intersubunit surface residues, it does not significantly alter the enzyme's thermodynamic activation parameters. This suggests that, for this system, the assembly of subunits does not impede cold adaptation. Moving beyond catalytic rates, Paper III addressed how temperature influences the diffusive process of substrate binding, focusing on heat capacity changes. Computational analysis of human thrombin inhibitors indicated that the observed negative heat capacity changes upon ligand binding are primarily due to the suppression of conformational equilibria present in the free inhibitors in solution. This occurred as opposed to being solely from changes in enzyme flexibility upon binding and highlights a crucial factor influencing the temperature dependence of binding processes. Finally, Paper IV focuses on developing a robust and efficient computational approach for predicting protein mutational effects on stability and ligand binding. This work presents QresFEP-2, a novel hybrid-topology free energy perturbation (FEP) protocol. Through rigorous benchmarking, QresFEP-2 demonstrated state-of-the-art accuracy compared to experimental data, while significantly improving computational efficiency. This tool provides a highly efficient and competitive open-source alternative for high-throughput in silico mutagenesis studies, relevant for protein engineering and drug design.
Thigh Muscles and Metabolic Dysfunction-Associated Steatotic Liver Disease : Findings From the SCAPIS/IGT-Microbiota Study
No full textBackground and Aims The relationship between skeletal muscle features and hepatic fat infiltration remains understudied. To address this knowledge gap, a cross-sectional observational study was conducted using data from two ancillary studies of the SCAPIS cohort. Method The study aimed to examine the relationship between skeletal thigh muscle radiodensity (Hounsfield Units, HU) and area (cm2), and metabolic dysfunction-associated steatotic liver disease (MASLD) and hepatic radiodensity (HU). Multivariable linear regression analyses were applied to data from 4620 participants (52% women) with a mean age of 57.9 years. Adjustments for confounders were computed in four theoretical models. Results Results showed a positive significant association between thigh muscle area and MASLD (OR = 1.29, 95% CI 1.02, 1.62, p = 0.033), and a negative association with hepatic radiodensity (B = -0.76, 95% CI -1.19, -0.34, p = 0.001), independent of muscle radiodensity. Additionally, a significant association was observed between muscle radiodensity and hepatic radiodensity (B = 0.37, 95% 0.09, 0.64, p = 0.008). Finally, sex differences were notable in the association between thigh muscle area and MASLD (F-test = 0.10). Specifically, we observed statistically significant associations between thigh muscle features and liver density/MASLD in men, but not in women. Conclusion Conclusively, increased thigh muscle volume was associated with greater odds of MASLD and hepatic steatosis, independent of muscle radiodensity. Yet, greater thigh muscle radiodensity was associated with decreased odds of hepatic steatosis, regardless of the muscle volume. Furthermore, a sex difference was observed in our study, underscoring the importance of considering sex-specific factors on the development of MASLD
Lacking the confidence of one's convictions : Gender differences in energy tariff literacy
No full textThe adoption of demand response programs, where end-users adjust their energy consumption to system constraints, is important for achieving a sustainable energy transition. While prior research has examined households' adoption and responses to price signals and tariffs that are used to elicit demand response, less attention has been given to how information on tariffs is differently understood between genders. This study investigates gender differences in both understanding of energy tariffs (energy tariff literacy) and confidence in understanding such tariffs, and examines whether differences in confidence are attributable to differences in understanding. Using secondary data from two studies (N = 1367 and N = 783), findings show a small gender gap in energy tariff literacy, with men scoring slightly higher, in one dataset, and no differences in the other. The confidence gap was substantially larger in both datasets; men reported greater confidence than women, a difference largely unexplained by differences in energy tariff literacy. These results extend gender confidence effects observed in other domains to energy literacy and understanding. Practically, the confidence gap may affect household decisions relating to energy consumption patterns, and risk that behaviour change or investment decisions are made by those with the strongest perceived knowledge, rather than actual knowledge. These findings emphasize the need for a gender-sensitive approach to demand response program design
Novel Synthetic Strategies for the Development of Peptides and Peptidomimetics in Alzheimer's Disease and Prostate Cancer
No full textPeptides are attractive therapeutic agents due to their high target affinity and specificity. However, their clinical application is often limited by poor metabolic stability, rapid renal clearance and low bioavailability, necessitating chemical modifications to generate viable drug candidates. This thesis addresses these challenges through the development of safer synthetic methods and the design of modified peptides targeting two clinically relevant proteins: the insulin-regulated aminopeptidase (IRAP) in Alzheimer’s disease and the gastrin-releasing peptide receptor (GRPR) in prostate cancer. The first part focuses on method development for the safer generation of diazomethane, a key reagent in peptide modification, particularly for the synthesis of β3-amino acids via α-diazoketones in the Arndt-Eistert homologation. However, diazomethane is explosive and carcinogenic, limiting its practical use. In Paper I, we report a method for the in situ generation of diazomethane through a click and release reaction between an enamine and a sulfonyl azide, thereby eliminating the hazardous handling of diazomethane. This strategy enabled the synthesis of α-diazoketones and methyl esters of carboxylic acids under mild, base-free conditions. The second part investigates structure–activity relationships (SAR) of modified peptides targeting IRAP and GRPR. In Paper II, twelve novel analogues of the macrocyclic IRAP inhibitor HA08 were synthesised via a new divergent route focusing on C-terminal modifications. IRAP inhibition has shown cognitive benefits in preclinical models of Alzheimer’s disease. Thus, inhibitors like HA08 are potential cognitive enhancers. However, this peptidomimetic suffers from poor metabolic stability, necessitating further structural modifications. The SAR study revealed that aromaticity and correct spatial orientation at the C-terminus were crucial for IRAP inhibition. Papers III and IV focus on the development of GRPR antagonists for imaging and therapy in early-stage prostate cancer. GRPR is overexpressed in prostate tumour cells but minimally expressed in healthy tissues, making it a valuable molecular target for theranostic application. In Paper III, the influence of rigidity and lipophilicity of aromatic linkers in 68Ga-labelled radiopeptides for PET imaging was explored. Moderate lipophilicity improved tumour uptake and imaging contrast, suggesting that fine-tuning linker properties can optimise radioligand pharmacokinetics. Finally, therapeutic application of GRPR antagonists requires enhancement of circulatory half-life to prolong tumour irradiation. In paper IV, albumin-binding moieties (ABMs) were incorporated into 117Lu-labelled radiopeptides, which extended circulatory half-life for all conjugates and increased GRPR-mediated uptake. The position and lipophilicity of the ABM strongly influenced biodistribution profile of the radioconjugates, with [177Lu]Lu-CA6356 and [177Lu]Lu-CA6357 showing the most favourable profile.
Self-Antigens Select B Cells : A New Perspective on B Cell Selection and Function
No full textThe adaptive immune system is shaped by self-recognition, creating a paradox where autoreactivity is essential for immune regulation, yet implicated in autoimmune diseases. Traditionally, B cell selection in the bone marrow (BM) has been viewed through the lens of negative selection, eliminating potentially harmful clones. Emerging evidence challenges this perspective, revealing a subset of B suppressor cells (Bsup) that actively regulate immune homeostasis. Unlike conventional negative selection, C1-specific Bsup cells, which recognize collagen type II (Col2), engage Col2-specific regulatory T cells (Tregs) to suppress inflammation in healthy individuals. This suggests that Bsup play a role in both peripheral and central tolerance, akin to Tregs. However, the molecular mechanisms governing Bsup selection, differentiation, and function remain unknown. Understanding how Bsup distinguish homeostatic from pathogenic autoreactivity could transform autoimmune disease treatment, shifting the focus from eliminating autoreactive B cells to harnessing their regulatory potential for precision immunotherapy
Enriching Retrieval-AugmentedGeneration with Non-TextualInformation to Support ScientificWriting
No full textRetrieval-Augmented Generation (RAG) systems have primarily focused on leveraging textualinformation, often overlooking the insight offered by non-textual elements such as figures, tables,equations, and code snippets. Furthermore, research on this topic remains sparse, withfew comprehensive approaches to tackling the integration of such elements into RAG pipelines.This thesis addresses this gap by proposing a method to enrich RAG systems with non-textualinformation, with a particular focus on research papers. The objective of this work is to enhance the capabilities of an academic research assistant RAGsystem by integrating non-textual content extracted from scientific documents. The proposedmethod involves a multi-step approach: extraction of non-textual components from diverse researchpapers using tools such as GROBID and Marker, processing and contextualization ofsaid components through classification and the extraction of related information from the papersand ultimately converting them into textual information using multimodal Large Language Models(LLMs) and iterative prompt engineering. This methodology is tested across multiple modelsprovided by OpenAI and Gemini to demonstrate its adaptability. The core result of this thesis is a complete, modular pipeline that transforms non-textual elementsinto meaningful textual descriptions, enabling their use within existing RAG systems, with a focuson academic writing. The framework supports customization, with each subsystem (extraction,processing, contextualization, and summarization) designed to be independently upgradable.Additionally, the thesis provides validated starting points for each subsystem: a reliable methodfor extracting non-textual elements, an effective approach to processing them accordingly throughsegmentation and classification, contextualizing them using related textual information, and a setof tested prompts to guide LLMs during summarization. In conclusion, this work attempts to lay the groundwork for future research into more efficientintegration of non-textual information into RAG systems of any type. It does so by offering asimple but scalable pipeline and proposing starting points for each step
Divergence of Leptin Receptor and Interleukin-6 Receptor Subunit b in Early Vertebrate Evolution and Physiological Insights from the Sea Lamprey
No full textCurrent knowledge of class-I cytokine receptors comes primarily from studies in jawed vertebrates (gnathostomes), and their origin and evolution remain unresolved. In this study, we identified a leptin receptor-like sequence (LepRL) and three interleukin-6 receptor subunit b-like sequences (IL6RBL) from a jawless vertebrate (cyclostome), the sea lamprey (Petromyzon marinus). Based on structural, phylogenetic, and syntenic analyses, we deduced that these lamprey receptors are likely distinct ohnologs to gnathostome LepR and IL6RB-related receptors, respectively, that arose in the two rounds of vertebrate whole-genome duplication (1R and 2R). Notably, lamprey LepRL likely originated from a different 1R progenitor than the one giving rise to gnathostome LepR during cyclostome hexaploidization. Differential patterns in mRNA expression of LepRL and IL6RBLs were observed among adult tissues, during larval metamorphosis, and in response to juvenile feeding. Feeding stimulated hepatic expression of LepRL and IL6RBL (namely, IL6RBL1) mRNAs in correlation with upregulation of insulin-like growth factor mRNA, whereas brain LepRL and IL6RBL1 mRNA expression was correlated positively with neuropeptide Y but inversely with intestinal content in fed juveniles. Notably, these observations along with immunolocalization of LepRL in the hypothalamus suggest a role of leptin signaling in regulating energy balance that is conserved among vertebrates. Additionally, seawater exposure stimulated branchial LepRL expression coincident with increased expression of ion transporters in ionocytes, indicating a role of leptin signaling in osmoregulation. These findings provide new insight into the early evolution of class-I cytokine receptors and reveal diverse functions of the leptin signaling system in jawless vertebrate
Smooth Calabi-Yau structures and the noncommutative Legendre transform
No full textWe elucidate the relation between smooth Calabi-Yau structures and pre-Calabi-Yau structures. We show that, from a smooth Calabi-Yau structure on an Ate-category A, one can produce a pre-Calabi-Yau structure on A; as defined in our previous work, this is a shifted noncommutative version of an integrable polyvector field. We explain how this relation is an analog of the Legendre transform, and how it defines a one-to-one mapping, in a certain homological sense. For concreteness, we apply this formalism to chains on based loop spaces of (possibly non-simply connected) Poincar & eacute; duality spaces and fully calculate the case of the circle
Organic Polymer and Molecular Nanoparticles for Photocatalytic Hydrogen Production : Insights into Excited-State Dynamics
No full textThe growing need for sustainable energy solutions has increased research efforts in artificial photosynthesis, particularly photocatalytic systems that convert solar energy into chemical fuels. Among these, organic nanoparticle (NP) photocatalysts have shown to be promising candidates for hydrogen evolution due to their earth-abundant composition, structural tunability, and ability to absorb visible light. However, their performance can only be improved through a more detailed understanding of the mechanisms involved. This thesis investigates how exciton dynamics, excited-state lifetimes, and molecular architecture influence the photocatalytic performance of organic NP systems. Using steady-state and time-resolved spectroscopic techniques across femtosecond to second timescales, this study provides new insights into the fundamental photophysical processes governing hydrogen evolution. Paper I demonstrates that the exciton diffusion length in the commonly used polymer NP photocatalyst PFBT significantly exceeds previous estimates, indicating that excitons can reach the NP-water interface before recombination with high probability. This finding is important because the accessibility of excitons at the interface determines whether charges can separate and participate in photocatalytic reactions. Paper II shows that molecular design strategies, such as incorporating strong electron-withdrawing units, adopting star-shaped geometries, and having favourable NP morphologies, synergistically enhance hydrogen evolution performance. These improvements are additionally linked to the formation of long-lived triplet charge-transfer states. Papers III and IV explore NP systems based on non-fullerene acceptors, which are widely used in organic photovoltaics, but underexplored in photocatalysis. Paper III reveals that Y5 NPs exhibit higher hydrogen evolution rates than Y6 NPs due to thermally activated delayed fluorescence and more favourable reduction potentials, contrasting with trends observed in photovoltaics. Paper IV introduces a newly designed A–D–A molecule, IT-PMI, which promotes intramolecular charge transfer and intermolecular interactions. This design enables charge hopping between chromophores, leading to improved photocatalytic activity and enhanced photostability compared to Y-series NPs.
Decentralized Voltage Regulation in Low-Voltage Grids Using Projection-Constrained Multi-Agent Deep Reinforcement Learning : A Case Study in E.ON’s Swedish Distribution Grid
No full textThis thesis presents a decentralized inverter-based voltage control strategy for low-voltage (LV) power grids with high photovoltaic (PV) penetration, using Multi-Agent Deep Reinforcement Learning (MADRL). Inspired by recent advancements in safe reinforcement learning, the pro-posed method integrates a projection layer into a MADRL framework to ensure that smart inverter actions respect physical and regulatory constraints. A real-world LV grid operated by E.ON in southern Sweden serves as the case study. The model is implemented in Python us-ing open-source libraries and trained on real consumption and simulated PV data. Simulation results show that the method decreases voltage deviations from the reference-voltage, but it increases the power losses in the grid, and outperforms simpler voltage control methods, evenunder high R/X-ratio conditions and varied PV-penetration levels. The study also discusses the technical, legal, and economic feasibility of deploying such AI-driven control in practice