Biology subjects

Cohen, N.

Publications and source records attributed to Cohen, N..

At least 19 recordsLinked to original sources

bioRxiv2026

Vaccination of pregnant mares with mRNA encoding a GPI-anchored form of the virulence-associated protein A of Rhodococcus equi induces strong IgG responses that are transferred to their foals

The bacterium Rhodococcus equi is a common cause of severe pneumonia in foals worldwide. Foals are exposed to R. equi from birth and are most susceptible in the first weeks of life. In North America, control measures are limited to transfusion to newborn foals of hyperimmune plasma whose potency is assessed by surpassing a threshold of IgG against the virulence-associated protein A (VapA) of R. equi. Here we report results of vaccination of pregnant mares with mRNA encoding a GPI-anchored virulence-associated protein A (VapA) of R. equi induced VapA-specific IgG subisotypes transferred to their foals during the 63-day period after birth. Pregnant mares were randomly assigned to receive either VapA mRNA vaccine (n=10) at 8 and 10 months of gestation or no vaccine (n=10). Serum was collected from mares at 8 months of gestation (before vaccination for immunized mares) and {approx}24 hours after foaling and from foals at ages 1, 35, and 63 days. Mean relative concentrations of were low before immunization in all mares and did not change over time in unvaccinated mares; however, mean relative concentrations of anti-VapA IgG1, IgG4/7, and IgG3/5 of mares were significantly increased after VapA mRNA immunization and strongly correlated with the corresponding concentrations in their foals sera at age {approx}24 hours. Concentrations of anti-VapA IgGs remained increased in foals of vaccinated mares compared with control foals for at least 35 days.

immunology↗

bioRxiv2026

Leveraging AI and structural proteomics for rational design of a KAT6A degrader

While targeted protein degraders such as PROTACs are a clinically proven therapeutic strategy, the discovery of novel degraders remains hampered by trial-and-error process. To address this challenge, we developed the AIMS platform, which combines structural proteomics with AI models for rational PROTAC design. AIMS is an end-to-end toolkit for PROTAC optimization, encompassing structure solving using proteomics and AI, prediction of ADME and degradation properties, and prospective ranking of compound design ideas. Altogether, this integrated platform successfully enabled the multi-parameter optimization of a potent and bioavailable in vivo validated KAT6A degrader, establishing a versatile framework for PROTAC development across various targets. O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=71 SRC="FIGDIR/small/727609v1_ufig1.gif" ALT="Figure 1"> View larger version (17K): org.highwire.dtl.DTLVardef@13596d0org.highwire.dtl.DTLVardef@140500eorg.highwire.dtl.DTLVardef@147e585org.highwire.dtl.DTLVardef@12dbdfe_HPS_FORMAT_FIGEXP M_FIG C_FIG

bioinformatics↗

bioRxiv2026

A hierarchy of locomotion costs shapes optimal foraging strategy

Animals must make decisions about how to move through their environment to find food, shelter and mates, yet the general principles underpinning effective search strategies remain poorly understood. Exploration patterns, such as alternating intensive local search with long-range relocation, are observed across many species, but whether these reflect universal organising principles, or are simply imposed by structured environments, is an open question. Here, we reconstruct hours of Caenorhabditis elegans foraging in homogeneous three-dimensional volumes and show that this animal achieves optimal volumetric coverage using a hierarchical search strategy of quasi-planar locomotion patches punctuated by costlier volumetric reorientations. We propose that underlying this cost hierarchy is the worms highly-constrained shape-space and show that despite being spanned by only five body-posture modes, it gives rise to effective locomotion, combining non-planar gaits, planar turns and three-dimensional reorientation manoeuvres. Our finding that action frequencies decay with inferred cost in a Boltzmann-like distribution is consistent with the Principle of Maximum Entropy: optimal search emerging from the tendency to maximise information gained subject to embodied constraints. This general principle offers a unifying framework to assess the optimality of animal decision-making, with implications for understanding nervous system function, foraging ecology, and the physics of living active matter.

animal behavior and cognition↗

bioRxiv2025

SARS-CoV-2 and MERS-CoV disrupt host protein synthesis via nsp1 with differential effects on the integrated stress response

Coronaviruses pose a serious threat to public health, driving the need for antiviral therapeutics and vaccines. Therefore, it is paramount to understand how this family of viruses evades cellular antiviral responses and establishes productive infection. The conserved coronavirus non-structural protein (nsp)1 has been shown to inhibit host protein synthesis and promote host mRNA degradation while viral mRNAs are protected. We showed previously that SARS-CoV-2 induces activation of host integrated stress response (ISR) kinases PKR and PERK, which promote phosphorylation of eIF2 and consequent inhibition of host protein synthesis. In contrast, eIF2 remains unphosphorylated during MERS-CoV infection. To investigate the interactions of nsp1 and the ISR kinases, we utilized recombinant SARS-CoV-2 and MERS-CoV expressing nsp1 with mutations in each of two conserved domains. Upon infection with SARS-CoV-2 nsp1 mutants, translation was shut down in wildtype (WT) and PKR knockout (KO) cells but rescued in PERK KO cells, likely due to reduced p-eIF2. In contrast, translation was rescued during infection with the analogous MERS-CoV nsp1 mutants even in WT cells. Moreover, SARS-CoV-2 WT suppressed expression of GADD34, a negative regulator of eIF2 phosphorylation, while SARS-CoV-2 nsp1 mutants induced GADD34. In contrast MERS-CoV WT induced GADD34. Utilizing single-molecule fluorescence in situ hybridization, we found that SARS-CoV-2 and MERS-CoV nsp1 promote host mRNA degradation during WT, but not nsp1 mutant, infection. Finally, while SARS-CoV-2 WT suppressed stress granule formation, nsp1 mutants induced stress granules containing host RNA. Thus, SARS-CoV-2 and MERS-CoV differ in interactions with the ISR and nsp1 control of host protein synthesis. SignificanceCoronaviruses cause disease across a wide range of animal species, and the human coronaviruses SARS-CoV-2 and MERS-CoV have caused epidemics of severe respiratory illness. Thus, it is imperative to understand how these viruses antagonize host responses and cause lethal disease. We show here that the betacoronavirus non-structural protein (nsp)1 promotes shutdown of host protein synthesis while preserving viral protein synthesis and, in addition, promotes degradation of host mRNAs. However, SARS-CoV-2 and MERS-CoV differ in their ability to manipulate the host integrated stress response, indicating that it is important to understand detailed coronavirus-host interactions and how they differ even between lethal coronaviruses. Such insights will inform the development of antiviral therapeutics to treat and prevent current and future coronavirus outbreaks.

microbiology↗

bioRxiv2025

A multiplexed approach for genetic screening of human cells by electron microscopy uncovers a critical effector of mitochondrial cristae shape

Mitochondria form complex and diverse membrane-architectures essential for their multiple functions. Whereas critical proteins sculpting mitochondrial membranes have been identified, the molecular basis for many key features remains enigmatic. Exploration of membrane ultrastructure, in general, is limited by the tradeoff between resolution and throughput: electron microscopy (EM) is essential to resolve their ultrastructure but lacks scalability for systematic functional discovery. To overcome this limitation, we developed a high-resolution screening pipeline for EM of multiplexed human cell pools, hMultiCLEM (human Multiplexed Correlative Light and EM). To showcase the power of hMultiCLEM we performed a genetic screen exploring mitochondrial ultrastructure. hMultiCLEM confirmed proposed cristae modulators and uncovered additional ones illuminating the protein networks driving cristae organization. Validation of candidates highlighted an intermembrane space (IMS) protein linked to Menieres disease, which we named MISHA (Mitochondrial-IMS membrane-SHApe-impacting protein). More broadly, hMultiCLEM transforms the EM field, enabling genetic/chemical screening in basic and medical research.

cell biology↗

bioRxiv2025

USP15: the fourth Proteasome-associated DUB

The human genome encodes approximately 100 deubiquitinating enzymes (DUBs), but only three are considered proteasome-associated DUBs (pDUBs): PSMD14/Rpn11, USP14, and UCHL5. Among these, only PSMD14 is an integral 19S subunit, whereas USP14 and UCHL5 bind transiently to specific proteasomal subunits. Given the dynamic nature of proteasome composition, we searched for additional pDUBs. USP15 was found to be associated with 26S proteasomes purified from cultured cells. In proteasome preparations from erythrocytes, USP15 was identified as the most abundant transient pDUB. It was even feasible to separate proteasomes containing USP15 from those containing USP14. Although USP15 utilizes an internal ubiquitin-like (UBL) domain for positioning itself at the proteasome, it did not compete with the UBL-containing USP14. USP15 facilitated substrate selection at the proteasome by efficiently disassembling short polyubiquitin (polyUb) chains, while sparing K48-linked tetra-ubiquitin conjugates from deubiquitination. This feature may aid the proteasome in differentiating between substrates to be rescued from those committed to proteolysis. Identification of a fourth pDUB encourages the continued search for additional proteasome-interacting proteins that modulate its substrate specificity in a context-specific manner.

biochemistry↗

bioRxiv2025

Identification of a Thermogenic Target in the Dorsal Raphe Nucleus for Weight Management

Obesity emerges from a complex interplay of factors, including imbalanced interoception, genetic predisposition, and environmental cues, ultimately disrupting body weight homeostasis1. While much research has concentrated on strategies to suppress appetite for sustained weight loss, insufficient attention has been given to counterregulatory mechanisms that promote energy expenditure. Here, we show that chronic inhibition of GABAergic neurons in the Dorsal Raphe Nucleus (DRNVGAT) reduces body weight in diet-induced obese (DIO) mice. In this study, molecular profiling and in-situ hybridization in rodent and human brains revealed that the constitutively activated orphan receptor GPR6 is selectively enriched in DRNVGAT neurons. We next developed and administered a potent and highly selective GPR6 inverse agonist, which significantly reduced weight gain in DIO mice by stimulating brown adipose tissue thermogenesis without affecting appetite. Altogether, this study transitions from transcriptomic profiling, high-throughput drug screening and metabolic phenotyping to successfully identify a novel candidate to treat obesity.

neuroscience↗

bioRxiv2024

Drugs Commonly Used in Fungal Infection Risk Conditions Promote Drug Tolerance or Resistance in Candida albicans

BackgroundFungal infections are an increasing concern, particularly among immunocompromised patients and those with comorbidities who require multiple medications. However, the effects of drugs targeting human pathways on fungal cells, and whether they influence antifungal drug responses, are poorly understood. MethodsWe systematically analyzed clinical guidelines to shortlist non-antifungal drugs commonly used in conditions that increase the likelihood of fungal infections. Focusing on the most prevalent fungal pathogen, we then tested how these drugs affected the antifungal response of Candida albicans to two commonly used antifungals, fluconazole and anidulafungin. Drug interactions identified were further assessed using checkerboard and disk diffusion assays. Finally, antifungal treatment efficacy of fluconazole in combination with negatively interacting drugs was evaluated in an in vivo Galleria mellonella model of disseminated C. albicans infection. FindingsOut of 119 drugs frequently co-administered with antifungals in 40 conditions associated with a high risk of fungal infections, 34 compounds affected the antifungal drug response in C. albicans, with most drugs reducing or antagonising antifungal efficacy, several through increasing resistance or tolerance. Notably, fluconazole combinations with carvedilol and loperamide promoted antifungal resistance in both fungal cultures and in Galleria mellonella. InterpretationOur findings suggest that medications frequently taken by patients at risk of fungal infections regularly act on the fungal pathogens and can affect the effectiveness of antifungals. We propose that human drugs acting on fungal pathogens may be an underestimated factor contributing to the evolution of antifungal tolerance and resistance.

microbiology↗

bioRxiv2024

The Role of Metabolism in Shaping Enzyme Structures Over 400 Million Years of Evolution

The functions of cells and proteins depend on their biochemical microenvironment. To understand how biochemical constraints shaped protein structural evolution, we coupled the extensive genetic and metabolic data from the Saccharomycotina subphylum with the capability of AlphaFold2 to systematically predict protein structures from sequence. Determining how 11,269 enzyme structures catalysing 361 different metabolic reactions evolved over 400 million years alongside their molecular functions, we report that metabolism has shaped the structural evolution of enzymes at different levels: the organisms overall metabolism; the topological organisation of the metabolic network; and each enzymes molecular properties. For example, structural evolution depends on each enzymes reaction mechanism, on the variability rather than the amount of metabolic flux, and on biosynthetic cost. Evolutionary cost-optimization is stronger on highly abundant enzymes and acts differently on different structural domains, with the exception of small-molecule binding sites, which are prioritised over other structural domains and lack cost-optimisation. Finally, while enzyme surfaces are less constrained, surface residues can also be exposed to positive selection for the co-evolution of protein-protein interaction sites. Accessing AlphaFolds power to predict protein structures systematically and across species barriers, facilitating the integration of protein structures with functional genomics, we were thus able to map biological constraints which shape protein structural evolution at scale and over long timelines.

systems biology↗

bioRxiv2024

Transcriptomic landscape of human induced pluripotent stem cell-derived osteogenic differentiation identifies a regulatory role of KLF16

Osteogenic differentiation is essential for bone development, metabolism, and repair; however, the underlying regulatory relationships among genes remain poorly understood. To elucidate the transcriptomic changes and identify novel regulatory genes involved in osteogenic differentiation, we differentiated mesenchymal stem cells (MSCs) derived from 20 human iPSC lines into preosteoblasts (preOBs) and osteoblasts (OBs). We then performed transcriptome profiling of MSCs, preOBs and OBs. The iPSC-derived MSCs and OBs showed similar transcriptome profiles to those of primary human MSCs and OBs, respectively. Differential gene expression analysis revealed global changes in the transcriptomes from MSCs to preOBs, and then to OBs, including the differential expression of 840 genes encoding transcription factors (TFs). TF regulatory network analysis uncovered a network comprising 451 TFs, organized into five interactive modules. Multiscale embedded gene co-expression network analysis (MEGENA) identified gene co-expression modules and key network regulators (KNRs). From these analyses, KLF16 emerged as an important TF in osteogenic differentiation. We demonstrate that overexpression of Klf16 in vitro inhibited osteogenic differentiation and mineralization, while Klf16+/- mice exhibited increased bone mineral density, trabecular number, and cortical bone area. Our study underscores the complexity of osteogenic differentiation and identifies novel regulatory genes such as KLF16, which plays an inhibitory role in osteogenic differentiation both in vitro and in vivo.

molecular biology↗

bioRxiv2023

Protein language models enable prediction of polyreactivity of monospecific, bispecific, and heavy-chain-only antibodies

Early assessment of antibody off-target binding is essential for mitigating developability risks such as fast clearance, reduced efficacy, toxicity, and immunogenicity. The baculovirus particle (BVP) binding assay has been widely utilized to evaluate polyreactivity of antibodies. As a complementary approach, computational prediction of polyreactivity is desirable for counter-screening antibodies from in silico discovery campaigns. However, there is a lack of such models. Herein, we present the development of an ensemble of three deep learning models based on two pan-protein foundational protein language models (ESM2 and ProtT5) and an antibody-specific protein language model (Antiberty). These models were trained in a transfer learning network to predict the outcomes in the BVP assay and the bovine serum albumin (BSA) binding assay which was developed as a complement to the BVP assay. The training was conducted on a large dataset of antibody sequences augmented with experimental conditions, which were collected through a highly efficient application system. The resulting models demonstrated robust performance on normal mAbs (monospecific with heavy and light chain), bispecific Abs, and single-domain Fc (VHH-Fc). Protein language models outperformed a model built using molecular descriptors calculated from AlphaFold 2 predicted structures. Embeddings from the antibody-specific and foundational protein language models resulted in similar performance. To our knowledge, this represents the first application of protein language models to predict assay data on bispecifics and VHH-Fcs. Our study yields valuable insights on building infrastructures to support machine learning activities and training models for critical assays in antibody discovery.

bioinformatics↗

bioRxiv2023

Detecting Anomalous Proteins Using Deep Representations

Many advances in biomedicine can be attributed to identifying unusual proteins and genes. Many of these proteins unique properties were discovered by manual inspection, which is becoming infeasible at the scale of modern protein datasets. Here, we propose to tackle this challenge using anomaly detection methods that automatically identify unexpected properties. We adopt a state-of-the-art anomaly detection paradigm from computer vision, to highlight unusual proteins. We generate meaningful representations without labeled inputs, using pretrained deep neural network models. We apply these protein language models (pLM) to detect anomalies in function, phylogenetic families, and segmentation tasks. We compute protein anomaly scores to highlight human prion-like proteins, distinguish viral proteins from their host proteome, and mark non-classical ion/metal binding proteins and enzymes. Other tasks concern segmentation of protein sequences into folded and unstructured regions. We provide candidates for rare functionality (e.g., prion proteins). Additionally, we show the anomaly score is useful in 3D folding-related segmentation. Our novel method shows improved performance over strong baselines and has objectively high performance across a variety of tasks. We conclude that the combination of pLM and anomaly detection techniques is a valid method for discovering a range of global and local protein characteristics.

bioinformatics↗

bioRxiv2023

Biological representation disentanglement of single-cell data

Due to its internal state or external environment, a cells gene expression profile contains multiple signatures, simultaneously encoding information about its characteristics. Disentangling these factors of variations from single-cell data is needed to recover multiple layers of biological information and extract insight into the individual and collective behavior of cellular populations. While several recent methods were suggested for biological disentanglement, each has its limitations; they are either task-specific, cannot capture inherent nonlinear or interaction effects, cannot integrate layers of experimental data, or do not provide a general reconstruction procedure. We present biolord, a deep generative framework for disentangling known and unknown attributes in single-cell data. Biolord exposes the distinct effects of different biological processes or tissue structure on cellular gene expression. Based on that, biolord allows generating experimentally-inaccessible cell states by virtually shifting cells across time, space, and biological states. Specifically, we showcase accurate predictions of cellular responses to drug perturbations and generalization to predict responses to unseen drugs. Further, biolord disentangles spatial, temporal, and infection-related attributes and their associated gene expression signatures in a single-cell atlas of Plasmodium infection progression in the mouse liver. Biolord can handle partially labeled attributes by predicting a classification for missing labels, and hence can be used to computationally extend an infected hepatocyte population identified at a late stage of the infection to earlier stages. Biolord applies to diverse biological settings, is implemented using the scvi-tools library, and is released as open-source software at https://github.com/nitzanlab/biolord.

bioinformatics↗

bioRxiv2022

An experimental and theoretical approach to understand the interaction between particles and mucosal tissues

Nanonization of poorly water-soluble drugs has shown great potential in improving their oral bioavailability by enhancing the dissolution rate and saturation solubility. Moreover, due to particle size reduction and larger surface area, the number of contact points with the gastrointestinal mucus favors adhesion. Similar phenomena could be anticipated when nano-pollutants come into direct contact with mucosal tissues. However, the fundamental features that govern the interaction of particles with mucus have not been investigated in a systematic and rational way before. In this work, we synthesize mucin hydrogels of different pore sizes with rheological properties that closely mimic the properties of freshly extracted porcine mucin. By using fluorescent pure curcumin particles, we characterize the effect of particle size (hydrodynamic diameter of 200 nm, and 1.2 and 1.3 m), concentration (18, 35, and 71 g mL-1), and hydrogel crosslinking density (which is directly related to the stiffness and governs the average pore size) on the diffusion-driven particle penetration in vitro. Next, we derive a phenomenological model that describes the physics behind the diffusion-derived penetration of particles into the mucin network and considers the contributions of the particle size, the particle concentration, and the crosslinking density of the mucin hydrogel. Finally, we challenge our experimental-theoretical approach by preliminarily assessing the oral pharmacokinetics of an anti-cancer model drug, namely dasatinib, in pristine and nanonized forms and two clinically relevant doses in rats. For of a dose of 10 mg kg-1, drug nanonization leads to a significant ~8- and ~21-fold increase of the drug oral bioavailability and half-life, respectively, with respect to the unprocessed micron-sized drug. When the drug dose of pure drug nanoparticles (which is directly related to the local concentration of the drug in the gastrointestinal tract) was increased to 15 mg mL-1, the oral bioavailability increased though not significantly, suggesting the saturation of the penetration sites in the mucus, as demonstrated by the in vitro model. Our overall results reveal the potential of this experimental-theoretical approach, shed light on the interaction of particulate matter and mucosal tissues, and pave the way for the development of tools that enable a more rational design of nano-drug delivery systems for mucosal administration and the assessment of risk factors related to the exposure of mucosal tissues to nano-pollutants.

bioengineering↗

bioRxiv2022

Anellovirus Structure Reveals a Mechanism for Immune Evasion

The significant impact of the human virome on human physiology is beginning to emerge thanks to modern sequencing methods and bioinformatic tools1. Anelloviruses, the principal constituent of the commensal human virome, are universally acquired in infancy and found throughout the body2,3,4. Since the discovery of the original torque teno virus in 19975, three genera of the Anelloviridae family, each extremely diverse genetically, have been found in humans. These viruses elicit weak immune responses that permit multiple strains to co-exist and persist for years in a typical individual6. However, because they do not cause disease7 and due to the lack of an in vitro culture system, anelloviruses remain poorly understood8,9. Basic features of the virus, such as the identity of its structural protein, have been unclear until now. Here, we describe the first structure of an anellovirus particle, which includes a jelly roll domain that forms a 60-mer icosahedral particle core from which spike domains extend to form a salient part of the particle surface. The spike domains come together around the 5-fold symmetry axes to form crown-like features. Relatively conserved patches of amino acids are near the base of the spike domain while a hypervariable region is at the apex. We propose that this structure renders the particle less susceptible to antibody neutralization by hiding vulnerable conserved epitopes while exposing highly diverse epitopes as immunological decoys, thereby contributing to the immune evasion properties of anelloviruses. This would contrast with viruses such as beak and feather disease virus, canine parvovirus or adeno-associated virus which lack such pronounced surface features. These results shed light on the structure of anelloviruses and provide a framework to understand their interactions with the immune system.

biochemistry↗

bioRxiv2022

A systematic approach to study protein-substrate specificity enables the identification of Ssh1 substrate range

Many cellular functions are carried out by protein pairs, or families, providing robustness alongside functional diversity. For such processes, it remains a challenge to map the degree of specificity versus promiscuity. Protein-protein interactions (PPIs) can be used to inform on these matters as they highlight cellular locals, regulation and, in cases where proteins affect other proteins - substrate range. However, methods to study transient PPIs systematically are underutilized. In this study we create a novel approach to study stable as well as transient PPIs in yeast. Our approach, Cel-lctiv (CELlular biotin-Ligation for Capturing Transient Interactions in Vivo), uses high- throughput pairwise proximity biotin ligation for uncovering PPIs systematically and in vivo. As a proof of concept we study the homologous translocation pores Sec61 and Ssh1. We show how Cel-lctiv can uncover the unique substrate range for each translocon allowing us to pinpoint a specificity determinator driving interaction preference. More generally this demonstrates how CEl-lctiv can provide direct information on substrate specificity even for highly homologous proteins.

cell biology↗

bioRxiv2022

A toolbox for systematic discovery of stable and transient protein interactors in baker's yeast

Identification of both stable and transient interactions is essential for understanding protein function and regulation. While assessing stable interactions is more straightforward, capturing transient ones is challenging. In recent years, sophisticated tools have emerged to improve transient interactor discovery, with many harnessing the power of evolved biotin ligases for proximity labelling. However, biotinylation-based methods have lagged behind in the model eukaryote, Saccharomyces cerevisiae, possibly due to the presence of several abundant, endogenously biotinylated proteins. In this study, we optimised robust biotin- ligation methodologies in yeast and increased their sensitivity by creating a bespoke technique for downregulating endogenous biotinylation which we term ABOLISH (Auxin- induced BiOtin LIgase diminiSHing). We used the endoplasmic reticulum insertase complex (EMC) to demonstrate our approaches and uncover new substrates. To make these tools available for systematic probing of both stable and transient interactions, we generated five full-genome collections of strains in which every yeast protein is tagged with each of the tested biotinylation machineries; some on the background of the ABOLISH system. This comprehensive toolkit enables functional interactomics of the entire yeast proteome.

molecular biology↗

bioRxiv2021

MERS-CoV endoribonuclease and accessory proteins jointly evade host innate immunity during infection of lung and nasal epithelial cells

Middle East respiratory syndrome coronavirus (MERS-CoV) emerged into humans in 2012, causing highly lethal respiratory disease. The severity of disease may be in part because MERS-CoV is adept at antagonizing early innate immune pathways - interferon (IFN) production and signaling, protein kinase R (PKR), and oligoadenylate synthetase ribonuclease L (OAS/RNase L) - generated in response to viral double-stranded (ds)RNA generated during genome replication. This is in contrast to SARS-CoV-2, which we recently reported activates PKR and RNase L and to some extent, IFN signaling. We previously found that MERS-CoV accessory proteins NS4a (dsRNA binding protein) and NS4b (phosphodiesterase) could weakly suppress these pathways, but ablation of each had minimal effect on virus replication. Here we investigated the antagonist effects of the conserved coronavirus endoribonuclease (EndoU), in combination with NS4a or NS4b. Inactivation of EndoU catalytic activity alone in a recombinant MERS-CoV caused little if any effect on activation of the innate immune pathways during infection. However, infection with recombinant viruses containing combined mutations with inactivation of EndoU and deletion of NS4a or inactivation of the NS4b phosphodiesterase promoted robust activation of the dsRNA-induced innate immune pathways. This resulted in ten-fold attenuation of replication in human lung derived A549 and primary nasal cells. Furthermore, replication of these recombinant viruses could be rescued to the level of WT MERS-CoV by knockout of host immune mediators MAVS, PKR, or RNase L. Thus, EndoU and accessory proteins NS4a and NS4b together suppress dsRNA-induced innate immunity during MERS-CoV infection in order to optimize viral replication. ImportanceMiddle East Respiratory Syndrome Coronavirus (MERS-CoV) causes highly lethal respiratory disease. MERS-CoV encodes several innate immune antagonists, accessory proteins NS4a and NS4b unique to the merbeco lineage and the nsp15 protein endoribonuclease (EndoU), conserved among all coronaviruses. While mutation of each antagonist protein alone has little effect on innate immunity, infections with recombinant MERS-CoVs with mutations of EndoU in combination with either NS4a or NS4b, activate innate signaling pathways and are attenuated for replication. Our data indicate that EndoU and accessory proteins NS4a and NS4b together suppress innate immunity during MERS-CoV infection, to optimize viral replication. This is in contrast to SARS-CoV-2 which activates these pathways and consistent with greater mortality observed during MERS-CoV infection compared to SARS-CoV-2.

microbiology↗