Biology subjects

Kucera, T.

Publications and source records attributed to Kucera, T..

4 recordsLinked to original sources

bioRxiv2024

Morphing Cholinesterase Inhibitor Amiridine into Multipotent Drugs for the Treatment of Alzheimer's Disease

The search for novel drugs to address the medical needs of Alzheimers disease (AD) is an ongoing process relying on the discovery of disease-modifying agents. Given the complexity of the disease, such an aim can be pursued by developing so-called multi-target directed ligands (MTDLs) that will impact the disease pathophysiology more comprehensively. Herewith, we contemplated the therapeutic efficacy of an amiridine drug acting as a cholinesterase inhibitor by converting it into a novel class of novel MTDLs. Applying the linking approach, we have paired amiridine as a core building block with memantine/adamantylamine, trolox, and substituted benzothiazole moieties to generate novel MTDLs endowed with additional properties like N-methyl-D-aspartate (NMDA) receptor affinity, antioxidant capacity, and anti-amyloid properties, respectively. The top-ranked amiridine-based compound 5d was also inspected by in silico to reveal the butyrylcholinesterase binding differences with its close structural analogue 5b. Our study provides insight into the discovery of novel amiridine-based drugs by broadening their target-engaged profile from cholinesterase inhibitors towards MTDLs with potential implications in AD therapy.

neuroscience↗

bioRxiv2024

Independent regulation of early trafficking of NMDA receptors by ligand-binding domains of the GluN1 and GluN2A subunits

The essential role of N-methyl-D-aspartate receptors (NMDARs) in excitatory neurotransmission is underscored by numerous pathogenic variants in the GluN subunits, including those identified in their ligand-binding domains (LBDs). The prevailing hypothesis postulates that the endoplasmic reticulum (ER) quality control machinery verifies the agonist occupancy of NMDARs; however, whether it controls the structure of LBDs or the functionality of NMDARs is unknown. Using alanine substitutions combined with microscopy and electrophysiology, we found that surface expression of GluN1/GluN2A receptors, the primary NMDAR subtype in the adult forebrain, strongly correlates with EC50 values for glycine and L-glutamate. Interestingly, co-expression of both GluN1 and GluN2A subunits with alanine substitutions led to an additive reduction in the surface number of GluN1/GluN2A receptors, as did co-expression of both GluN1 and GluN2A subunits containing closed cleft conformation of LBDs. The synchronized ER release confirmed the altered regulation of early trafficking of GluN1/GluN2A receptors bearing alanine substitutions in the LBDs. Furthermore, the human versions of GluN1/GluN2A receptors containing pathogenic GluN1-S688Y, GluN1-S688P, GluN1-D732E, GluN2A-S511L, and GluN2A-T690M variants exhibited distinct surface expression compared to the corresponding alanine substitutions. Mutant cycles of GluN1-S688, GluN1-D732, GluN2A-S511, and GluN2A-T690 residues revealed, in most cases, a weak correlation between surface expression of the mutant GluN1/GluN2A receptors and their EC50 values for glycine or L-glutamate. Consistent with our experimental data, molecular modeling and dynamics showed that the ER quality control machinery likely perceives structural changes of the LBDs but not the functionality of GluN1/GluN2A receptors. Significant statementOur study showed that structural changes in LBDs independently regulate the early trafficking of GluN1/GluN2A receptors and that the surface numbers of mutant GluN1/GluN2A receptors do not necessarily correlate with their agonist sensitivity. In addition, we validated a novel system of synchronized release of GluN1/GluN2A receptors from the ER. Together, our experimental and in silico findings support the urgency of further detailed research on the regulation of early trafficking of NMDARs, as it may open the avenue to targeted therapeutic intervention of CNS disorders associated with pathogenic variants in GluN subunits.

neuroscience↗

bioRxiv2021

Conditional Generative Modeling for De Novo Protein Design with Hierarchical Functions

MotivationProtein design has become increasingly important for medical and biotechnological applications. Because of the complex mechanisms underlying protein formation, the creation of a novel protein requires tedious and time-consuming computational or experimental protocols. At the same time, machine learning has enabled the solving of complex problems by leveraging large amounts of available data, more recently with great improvements on the domain of generative modeling. Yet, generative models have mainly been applied to specific sub-problems of protein design. ResultsHere we approach the problem of general purpose protein design conditioned on functional labels of the hierarchical Gene Ontology. Since a canonical way to evaluate generative models in this domain is missing, we devise an evaluation scheme of several biologically and statistically inspired metrics. We then develop the conditional generative adversarial network ProteoGAN and show that it outperforms several classic and more recent deep learning baselines for protein sequence generation. We further give insights into the model by analysing hyperparameters and ablation baselines. Lastly, we hypothesize that a functionally conditional model could generate proteins with novel functions by combining labels and provide first steps into this direction of research. AvailabilityCode and data is available at https://github.com/timkucera/proteogan Contacttim.kucera@bsse.ethz.ch, mt@visium.ch, lpapaxanthos@google.com

bioinformatics↗

bioRxiv2021

Heterodimers of functionally divergent ARF-GEF paralogues prevented by self-interacting dimerisation domain

Functionally divergent paralogs of homomeric proteins do not form potentially deleterious heteromers, which requires distinction between self and non-self (Hochberg et al., 2018; Marchant et al, 2019; Marsh and Teichmann, 2015). In Arabidopsis, two ARF guanine-nucleotide exchange factors (ARF-GEFs) related to mammalian GBF1, named GNOM and GNL1, can mediate coatomer complex (COPI)-coated vesicle formation in retrograde Golgi-endoplasmic reticulum (ER) traffic (Geldner et al., 2003; Richter et al., 2007; Teh and Moore, 2007). Unlike GNL1, however, GNOM is also required for polar recycling of endocytosed auxin efflux regulator PIN1 from endosomes to the plasma membrane. Here we show that these paralogues form homodimers constitutively but no heterodimers. We also address why and how GNOM and GNL1 might be kept separate. These paralogues share a common domain organisation and each N-terminal dimerisation (DCB) domain can interact with the complementary fragment ({Delta}DCB) of its own and the other protein. However, unlike self-interacting DCBGNOM (Grebe et al., 2000; Anders et al., 2008), DCBGNL1 did not interact with itself nor DCBGNOM. DCBGNOM removal or replacement with DCBGNL1, but not disruption of cysteine bridges that stabilise DCB-DCB interaction, resulted in GNOM-GNL1 heterodimers which impaired developmental processes such as lateral root formation. We propose precocious self-interaction of the DCBGNOM domain as a mechanism to preclude formation of fitness-reducing GNOM-GNL1 heterodimers.

plant biology↗