Biology subjects

Koester, J.

Publications and source records attributed to Koester, J..

8 recordsLinked to original sources

bioRxiv2026

Structural dynamics insights into principles underlying the fitness of new broadly potent AAVs

Adeno-associated virus (AAV) is a leading platform for gene therapy, but current clinical-stage vectors require high doses associated with adverse events. Engineering of AAVs has produced more efficient vectors, although the mechanism underlying these improvements often remains poorly understood, limiting further development and raising potential safety concerns. Here, we leveraged a new workflow for AAV engineering with single-cell resolution, called scAAVengr-Hunt, to create best-in-class AAVs for gene delivery. ATX002, the top-performing vector, demonstrates broad potency across species, including nonhuman primate, mouse, and human, as well as across retina and brain. To understand the mechanism underlying this broad potency, we performed molecular dynamics simulations comparing AAV variants spanning a range of fitness levels. Structural dynamics analysis revealed a bifunctional molecular mechanism that confers potency through increased affinity of the capsid to the AAV receptor and regulation of heparan sulfate binding. This work provides critical insights relating structural mechanism to the fitness of engineered AAVs and establishes rich new avenues for AAV engineering through the integration of sequence-level analysis with computational biophysics.

bioengineering↗

bioRxiv2025

Redirecting resistance evolution in BRAFV600 melanoma by inhibition of the peroxiredoxin-thioredoxin system

Drug-tolerant persister cells (DTPs) exhibit remarkable cell state heterogeneity and phenotypic evolvability. However, the central question of how DTPs epigenetically coordinate their metabolic flexibility to adapt early to therapeutic stress remains unanswered. We have recently shown that the histone demethylase KDM5B, which is intrinsically expressed in differentiated melanoma DTPs, reprograms the metabolic cell landscape. However, the exact mechanism by which KDM5B affects underlying metabolic enzymes, and whether this reveals new druggable vulnerabilities, remain unknown. By transcriptional and epigenetic profiling of BRAFV600 melanoma cells following KDM5B gene silencing, we discovered a direct molecular axis between the epigenetic regulator KDM5B and the PRDX/TXN ROS detoxification system. This metabolic axis is regulated independently of KDM5Bs demethylase activity. Furthermore, RNAi approaches and the pharmacological inhibition of the PRDX/TXN system led to ROS-induced cell death in differentiated melanoma DTPs and a delay of resistance development to MAPK inhibition. This process was independent of lipid-ROS-driven ferroptosis. Additionally, single-cell transcriptome analyses from pre-clinical melanoma models under continuous MAPK inhibitory treatment demonstrated altered cellular differentiation dynamics, with a reduction in the early evolution into the mesenchymal DTP state under concomitant PRDX inhibition. Interestingly, the degree of melanoma cell state differentiation at the onset of treatment was a major determinant for the transition towards the neural crest-like DTP state. Our study identified a high degree of epigenetic-metabolic connectivity and flexibility within the melanoma DTP pool and urges caution with single redox pathway-targeted strategies for tumor elimination in the future. Prospectively, our results point towards a new resistance targeting strategy for BRAFV600 melanoma patients based on pharmacological re-direction of the evolution of melanoma cell states already at therapy onset. HighlightsO_LITranscriptional and epigenetic profiling identified KDM5B as a regulator of the PRDX/TXN-ROS detoxification system C_LIO_LIPRDX inhibition increases the vulnerability of KDM5Bhigh DTPs to ROS, independent of ferroptosis C_LIO_LIPRDX inhibition delays resistance to MAPK inhibition in BRAFV600 melanoma cells C_LIO_LILongitudinal single-cell transcriptome analysis reveals that PRDX inhibition re-directs early DTP evolution C_LIO_LIPRDX/TXN gene expression is predictive for melanoma patient survival C_LI

cancer biology↗

bioRxiv2025

Diverse ssRNA viruses associated with Karenia brevis harmful algal blooms in Southwest Florida

Harmful algal blooms (HABs) caused by the dinoflagellate Karenia brevis frequently occur in the eastern Gulf of Mexico, where they negatively impact the environment, human health, and economy. Very little is known about viruses associated with K. brevis blooms, although viral infection of other HAB-forming phytoplankton species can play an important role in bloom dynamics. We used viral metagenomics to identify viruses in 11 pooled seawater samples collected from Southwest Florida, USA in 2021 during a severe, spatiotemporally dynamic K. brevis bloom. Assembled viral genomes were similar to published genomes from the order Picornavirales, family Marnaviridae, and genera Sogarnavirus, Bacillarnavirus, and Marnarnavirus. Viruses from these groups infect bloom-forming diatoms (Chaetoceros sp. and Rhizosolenia setigera) and the raphidophyte Heterosigma akashiwo. We also recovered unclassified Riboviria genomes related to a Symbiodinium positive-sense ssRNA virus sequenced from coral dinoflagellate symbionts. Reverse-transcriptase PCR assays were performed to monitor the occurrence of seven representative virus genomes in these 2021 samples and 43 seawater samples collected during a subsequent, typical bloom between November 2022 and May 2023. Over half of the samples contained multiple viruses, and at least one viral genome was detected in 44 of 54 samples, collected across seasons and years, highlighting the ubiquity of these viruses in this region. Alpha diversity was highest in the summer months and positively correlated with K. brevis cell counts. Multiple regression revealed month and the presence of unclassified Riboviria sequences most similar to dinoflagellate viruses as significant predictors of K. brevis cellular abundance. ImportanceHarmful algal blooms caused by the dinoflagellate Karenia brevis negatively impact the tourism, fisheries, and public health sectors. Anticipated impacts of climate change, nutrient pollution, and ocean acidification may sustain and/or exacerbate K. brevis blooms in the future, underscoring the need for proactive monitoring, communication, and mitigation strategies. This study represents a pioneering effort in monitoring viruses associated with K. brevis blooms. The findings lay the groundwork for studying the effects of environmental drivers on K. brevis blooms and their associated viruses, as well as for exploring the roles of viruses in bloom dynamics and potential applications of viruses as biocontrol agents for K. brevis blooms. Furthermore, the comparison of viral dynamics relative to local and regional bloom dynamics in this study helps inform future monitoring and modeling needs.

microbiology↗

bioRxiv2022

Epigenetic plasticity via adaptive DNA hypermethylation and clonal expansion underlie resistance to oncogenic pathway inhibition in pancreatic cancer

Pancreatic ductal adenocarcinoma (PDAC) is an aggressive cancer with poor prognosis. Drug resistance is the major cause for therapeutic failure in PDAC patients with progressive disease. The mechanisms underlying resistance formation are complex and remain poorly understood. To gain insights into molecular changes during the formation of resistance to oncogenic MAPK pathway inhibition we utilized short-term passaged primary tumor cells from ten PDACs of genetically engineered mice. We followed gain and loss of resistance upon MEKi exposure and withdrawal by longitudinal integrative analysis of whole genome sequencing, whole genome bisulfite sequencing, RNA-sequencing and mass spectrometry data. We found that resistant cell populations under increasing MEKi treatment evolved by the expansion of a single clone but were not a direct consequence of known resistance-conferring mutations. Rather, resistant cells showed adaptive DNA hypermethylation of 209 and hypomethylation of 8 genomic sites, most of which overlap with regulatory elements known to be active in murine PDAC cells. Both DNA methylation changes and MEKi resistance were transient and reversible upon drug withdrawal. The effector caspase CASP3 is one of the 114 genes for which transcriptional downregulation inversely correlated with the methylation status of the associated DNA region. CASP3 inactivation in resistant cells led to attenuation of drug-induced apoptosis which could be reversed by DNA methyltransferase inhibition with remarkable sensitivity exclusively in the resistant cells. Overall, our data provide a context for characterization and targeting of epigenetically mediated resistance mechanisms in PDAC.

cancer biology↗

bioRxiv2021

Microphaser - a small-scale phasing approach for improved personalized neopeptidome creation

MotivationHaplotype phasing approaches have been shown to improve accuracy of the search space of neoantigen prediction by determining if adjacent variants are located on the same chromosomal copy. However, the aneuploid nature of cancer cells as well as the admixture of different clones in tumor bulk sequencing data are challenging the current diploid based phasing algorithms. We present microphaser, a small-scale phasing approach to improve haplotyping variants in cancer samples. Microphaser aims to create a more accurate neopeptidome for downstream neoantigen prediction. ResultsMicrophaser achieves large concordance with state-of-the-art phasing-aware neoantigen prediction pipeline neoepiscope, with differences in edge cases and an improved filtering step. AvailabilityMicrophaser is written in the Rust programming language. It is made available via Github (https://github.com/koesterlab/microphaser)and Bioconda. The corresponding prediction pipeline (https://github.com/snakemake-workflows/dna-seq-neoantigen-prediction) has been written within the Snakemake workflow management system and can be deployed as part of the snakemake-workflows project.

bioinformatics↗

bioRxiv2020

A novel single-domain Na+-selective voltage-gated channel in photosynthetic eukaryotes

The evolution of Na+-selective four-domain voltage-gated channels (4D-Navs) in animals allowed rapid Na+-dependent electrical excitability, and enabled the development of sophisticated systems for rapid and long-range signalling. Whilst bacteria encode single-domain Na+-selective voltage-gated channels (BacNav), they typically exhibit much slower kinetics than 4D-Navs, and are not thought to have crossed the prokaryote-eukaryote boundary. As such, the capacity for rapid Na+-selective signalling is considered to be confined to certain animal taxa, and absent from photosynthetic eukaryotes. Certainly, in land plants, such as the Venus Flytrap where fast electrical excitability has been described, this is most likely based on fast anion channels. Here, we report a unique class of eukaryotic Na+-selective single-domain channels (EukCatBs) that are present primarily in haptophyte algae, including the ecologically important calcifying coccolithophores. The EukCatB channels exhibit very rapid voltage-dependent activation and inactivation kinetics, and sensitivity to the highly selective 4D-Nav blocker tetrodotoxin. The results demonstrate that the capacity for rapid Na+-based signalling in eukaryotes is not restricted to animals or to the presence of 4D-Navs. The EukCatB channels therefore represent an independent evolution of fast Na+-based electrical signalling in eukaryotes that likely contribute to sophisticated cellular control mechanisms operating on very short time scales in unicellular algae. One Sentence SummaryThe capacity for rapid Na+-based signalling has evolved in ecologically important coccolithophore species via a novel class of voltage-gated Na+ channels, EukCatBs.

microbiology↗

bioRxiv2018

Full-length de novo viral quasispecies assembly through variation graph construction

MotivationViruses populate their hosts as a viral quasispecies: a collection of genetically related mutant strains. Viral quasispecies assembly refers to reconstructing the strain-specific haplotypes from read data, and predicting their relative abundances within the mix of strains, an important step for various treatment-related reasons. Reference-genome-independent (\"de novo\") approaches have yielded benefits over reference-guided approaches, because reference-induced biases can become overwhelming when dealing with divergent strains. While being very accurate, extant de novo methods only yield rather short contigs. It remains to reconstruct full-length haplotypes together with their abundances from such contigs.\n\nMethodWe first construct a variation graph, a recently popular, suitable structure for arranging and integrating several related genomes, from the short input contigs, without making use of a reference genome. To obtain paths through the variation graph that reflect the original haplotypes, we solve a minimization problem that yields a selection of maximal-length paths that is optimal in terms of being compatible with the read coverages computed for the nodes of the variation graph. We output the resulting selection of maximal length paths as the haplotypes, together with their abundances.\n\nResultsBenchmarking experiments on challenging simulated data sets show significant improvements in assembly contiguity compared to the input contigs, while preserving low error rates. As a consequence, our method outperforms all state-of-the-art viral quasispecies assemblers that aim at the construction of full-length haplotypes, in terms of various relevant assembly measures. Our tool, Virus-VG, is publicly available at https://bitbucket.org/jbaaijens/virus-vg.

bioinformatics↗

bioRxiv2017

Modeling and simulating networks of interdependent protein interactions

Protein interactions are fundamental building blocks of biochemical reaction systems underlying cellular functions. The complexity and functionality of these systems emerge not only from the protein interactions themselves but also from the dependencies between these interactions, e.g., allosteric effects, mutual exclusion or steric hindrance. Therefore, formal models for integrating and using information about such dependencies are of high interest. We present an approach for endowing protein networks with interaction dependencies using propositional logic, thereby obtaining constrained protein interaction networks (\"constrained networks\"). The construction of these networks is based on public interaction databases and known as well as text-mined interaction dependencies. We present an efficient data structure and algorithm to simulate protein complex formation in constrained networks. The efficiency of the model allows a fast simulation and enables the analysis of many proteins in large networks. Therefore, we are able to simulate perturbation effects (knockout and overexpression of single or multiple proteins, changes of protein concentrations). We illustrate how our model can be used to analyze a partially constrained human adhesome network. Comparing complex formation under known dependencies against without dependencies, we find that interaction dependencies limit the resulting complex sizes. Further we demonstrate that our model enables us to investigate how the interplay of network topology and interaction dependencies influences the propagation of perturbation effects. Our simulation software CPINSim (for Constrained Protein Interaction Network Simulator) is available under the MIT license at http://github.com/BiancaStoecker/cpinsimandviaBioconda (https://bioconda.github.io).\n\nAuthor summaryProteins are the main molecular tools of cells. They do not act individually, but rather collectively in order to peform complex cellular actions. Recent years have led to a relatively good understanding about which proteins may interact, both in general and in specific conditions, leading to the definition of protein interaction networks. However, the reality is more complex, and protein interactions are not independent of each other. Instead, several potential interaction partners of a specific protein may compete for the same binding domain, making all of these interactions mutually exclusive. Additionally, a binding of a protein to another one can enable or prevent their interactions with other proteins, even if those interactions are mediated by different domains. Hence, understanding how the dependencies (or constraints) of protein interactions affect the behaviour of the system is an important and timely goal, as data is now becoming available. Here we present a mathematical framework to formalize such interaction constraints and incorporate them into the simulation of protein complex formation. With our framework, we are able to better understand how perturbations of single proteins (knockout or overexpression) impact other proteins in the network.

systems biology↗