Faculty Bibliography

The risk of zoonotic coronavirus spillover into the human population, as highlighted by the SARS-CoV-2 pandemic, demands the development of pan-coronavirus antivirals. The efficacy of existing antiviral ribonucleoside/ribonucleotide analogs, such as remdesivir, is decreased by the viral proofreading exonuclease NSP14-NSP10 complex. Here, using a novel assay and in silico modeling and screening, we identified NSP14-NSP10 inhibitors that increase remdesivir's potency. A model compound, sofalcone, both inhibits the exonuclease activity of SARS-CoV-2, SARS-CoV, and MERS-CoV in vitro, and synergistically enhances the antiviral effect of remdesivir, suppressing the replication of SARS-CoV-2 and the related human coronavirus OC43. The validation of top hits from our primary screenings using cellular systems provides proof-of-concept for the NSP14 complex as a therapeutic target.

MOTIVATION/BACKGROUND:Cell polarity refers to the asymmetric organization of cellular components in various cells. Epithelial cells are the best-known examples of polarized cells, featuring apical and basolateral membrane domains. Mounting evidence suggests that short linear motifs play a major role in protein trafficking to these domains, although the exact rules governing them are still elusive. RESULTS:In this study we prepared neural networks that capture recurrent patterns to classify transmembrane proteins localizing into apical and basolateral membranes. Asymmetric expression of drug transporters results in vectorial drug transport, governing the pharmacokinetics of numerous substances, yet the data on how proteins are sorted in epithelial cells is very scattered. The provided method may offer help to experimentalists to identify or better characterize molecular networks regulating the distribution of transporters or surface receptors (including viral entry receptors like that of COVID-19). AVAILABILITY/BACKGROUND:The prediction server PolarProtPred is available at http://polarprotpred.ttk.hu. SUPPLEMENTARY INFORMATION/BACKGROUND:Supplementary data are available at Bioinformatics online.

Many proteins contain intrinsically disordered regions (IDRs) which carry out important functions without relying on a single well-defined conformation. IDRs are increasingly recognized as critical elements of regulatory networks and have been also associated with cancer. However, it is unknown whether mutations targeting IDRs represent a distinct class of driver events associated with specific molecular and system-level properties, cancer types and treatment options. Here, we used an integrative computational approach to explore the direct role of intrinsically disordered protein regions driving cancer. We showed that around 20% of cancer drivers are primarily targeted through a disordered region. These IDRs can function in multiple ways which are distinct from the functional mechanisms of ordered drivers. Disordered drivers play a central role in context-dependent interaction networks and are enriched in specific biological processes such as transcription, gene expression regulation and protein degradation. Furthermore, their modulation represents an alternative mechanism for the emergence of all known cancer hallmarks. Importantly, in certain cancer patients, mutations of disordered drivers represent key driving events. However, treatment options for such patients are currently severely limited. The presented study highlights a largely overlooked class of cancer drivers associated with specific cancer types that need novel therapeutic options.

Transcription factor phosphorylation at specific sites often activates gene expression, but how environmental cues quantitatively control transcription is not well-understood. Activating protein 1 transcription factors are phosphorylated by mitogen-activated protein kinases (MAPK) in their transactivation domains (TAD) at so-called phosphoswitches, which are a hallmark in response to growth factors, cytokines or stress. We show that the ATF2 TAD is controlled by functionally distinct signaling pathways (JNK and p38) through structurally different MAPK binding sites. Moreover, JNK mediated phosphorylation at an evolutionarily more recent site diminishes p38 binding and made the phosphoswitch differently sensitive to JNK and p38 in vertebrates. Structures of MAPK-TAD complexes and mechanistic modeling of ATF2 TAD phosphorylation in cells suggest that kinase binding motifs and phosphorylation sites line up to maximize MAPK based co-regulation. This study shows how the activity of an ancient transcription controlling phosphoswitch became dependent on the relative flux of upstream signals.

Most cells in multicellular organisms are somehow asymmetric, polarized: maintaining separate membrane domains. Typical examples are the epithelial cells (apical-basal polarization), neurons (dendritic-axonal domains), or migratory cells (with a leading and a trailing edge). Here we present the most comprehensive database containing experimentally verified mammalian proteins that display polarized sorting or secretion, focusing on epithelial polarity. In addition to the source cells or tissues, homology-based inferences and transmembrane topology (if applicable) are all provided. PolarProtDb also offers a detailed interface displaying all information that may be relevant for trafficking: including post-translational modifications (glycosylations and phosphorylations), known or predicted short linear motifs conserved across orthologs, as well as potential interaction partners. Data on polarized sorting has so far been scattered across myriads of publications, hence difficult to access. This information can help researchers in several areas, such as scanning for potential entry points of viral agents like COVID-19. PolarProtDb shall be a useful resource to design future experiments as well as for comparative analyses. The database is available at http://polarprotdb.enzim.hu.

Cancer is a heterogeneous genetic disease that alters the proper functioning of proteins involved in key regulatory processes such as cell cycle, DNA repair, survival, or apoptosis. Mutations often accumulate in hot-spots regions, highlighting critical functional modules within these proteins that need to be altered, amplified, or abolished for tumor formation. Recent evidence suggests that these mutational hotspots can correspond not only to globular domains, but also to intrinsically disordered regions (IDRs), which play a significant role in a subset of cancer types. IDRs have distinct functional properties that originate from their inherent flexibility. Generally, they correspond to more recent evolutionary inventions and show larger sequence variations across species. In this work, we analyzed the evolutionary origin of disordered regions that are specifically targeted in cancer. Surprisingly, the majority of these disordered cancer risk regions showed remarkable conservation with ancient evolutionary origin, stemming from the earliest multicellular animals or even beyond. Nevertheless, we encountered several examples where the mutated region emerged at a later stage compared with the origin of the gene family. We also showed the cancer risk regions become quickly fixated after their emergence, but evolution continues to tinker with their genes with novel regulatory elements introduced even at the level of humans. Our concise analysis provides a much clearer picture of the emergence of key regulatory elements in proteins and highlights the importance of taking into account the modular organisation of proteins for the analyses of evolutionary origin.

The mammalian FBXL10-RNF68-RNF2 ubiquitin ligase complex (FRRUC) mono-ubiquitylates H2A at Lys119 to repress transcription in unstressed cells. We found that the FRRUC is rapidly and transiently recruited to sites of DNA damage in a PARP1- and TIMELESS-dependent manner to promote mono-ubiquitylation of H2A at Lys119, a local decrease of H2A levels, and an increase of H2A.Z incorporation. Both the FRRUC and H2A.Z promote transcriptional repression, double strand break signaling, and homologous recombination repair (HRR). All these events require both the presence and activity of the FRRUC. Moreover, the FRRUC and its activity are required for the proper recruitment of BMI1-RNF2 and MEL18-RNF2, two other ubiquitin ligases that mono-ubiquitylate Lys119 in H2A upon genotoxic stress. Notably, whereas H2A.Z is not required for H2A mono-ubiquitylation, impairment of the latter results in the inhibition of H2A.Z incorporation. We propose that the recruitment of the FRRUC represents an early and critical regulatory step in HRR.

One of the most popular means to follow interactions between bio(macro)molecules is Förster resonance energy transfer (FRET). There is large interest in widening the selection of fluorescent FRET pairs especially in the region of the red/far red range, where minimal autofluorescence is encountered. A set of bioorthogonally applicable fluorescent dyes, synthesized recently in our lab, were paired (Cy3T/Cy5T; Cy1A/Cy3T and Cy1A/CBRD1A) based on their spectral characteristics in order to test their potential in FRET applications. For fast elaboration of the selected pairs we have created a bioorthogonalized platform based on complementary 17-mer DNA oligomers. The cyclooctynylated strands were modified nearly quantitatively with the fluorophores via bioorthogonal chemistry steps, using azide- (Cy1; CBRD1) or tetrazine-modified (Cy3; Cy5) dyes. Reactions were followed by capillary electrophoresis using a method specifically developed for this project. FRET efficiencies of the fluorescent dye pairs were compared both in close proximity (5' and 3' matched) and at larger distance (5' and 5' matched). The specificity of FRET signals was further elaborated by denaturation and competition studies. Cy1A/Cy3T and Cy1A/CBRD1A introduced here as novel FRET pairs are highly recommended for FRET applications based on the significant changes in fluorescence intensities of the donor and acceptor peaks. Application of one of the FRET pairs was demonstrated in live cells, transfected with labeled oligos. Furthermore, the concise installation of the dyes allows for efficient fluorescence modification of any selected DNA strands as was demonstrated in the construction of Cy3T labeled oligomers, which were used in the FISH-based detection of Helicobacter pylori.

Short linear motifs (SLiMs) are protein binding modules that play major roles in almost all cellular processes. SLiMs are short, often highly degenerate, difficult to characterize and hard to detect. The eukaryotic linear motif (ELM) resource (elm.eu.org) is dedicated to SLiMs, consisting of a manually curated database of over 275 motif classes and over 3000 motif instances, and a pipeline to discover candidate SLiMs in protein sequences. For 15 years, ELM has been one of the major resources for motif research. In this database update, we present the latest additions to the database including 32 new motif classes, and new features including Uniprot and Reactome integration. Finally, to help provide cellular context, we present some biological insights about SLiMs in the cell cycle, as targets for bacterial pathogenicity and their functionality in the human kinome.