Eran Bacharach

Deciphering protein function is fundamental to advancements in medicine and biotechnology. However, conventional experimental characterization remains resource-intensive. Public large language models (LLMs), though proficient in natural language processing, often fail to accurately interpret and predict the functional and structural properties of proteins, limiting their utility in bioinformatics. To address this gap, we introduce BetaDescribe, designed to generate detailed and rich textual descriptions of proteins, including their function, catalytic activity, involvement in specific metabolic pathways, subcellular localizations, and the presence of specific domains. The trained BetaDescribe model receives protein sequences as input and outputs a textual description of these properties. BetaDescribe starting point was the LLAMA2 model, which was trained on trillions of tokens. Our model was next trained on datasets containing both biological and English text, which allowed the incorporation of biological knowledge. In addition to the description generator, BetaDescribe comprises multiple validator models and a judge, which together enable accurate ranking of alternative generated descriptions. We demonstrate the utility of BetaDescribe by providing descriptions for proteins that share little to no sequence similarity to proteins with functional descriptions in public datasets. Using in silico mutagenesis, we further show that BetaDescribe relies on functionally important regions, as part of its prediction, suggesting that the model identifies regions of importance for the protein functionality without needing homologous sequence. BetaDescribe offers a powerful tool to explore protein functionality, augmenting existing approaches such as annotation transfer based on sequence or structure similarity.

The online version contains supplementary material available at 10.1186/s12964-026-02823-1. The online version contains supplementary material available at 10.1186/s12964-026-02823-1.

Rhabdoviruses exhibit a broad host range, yet the cellular receptors underlying their cross-species tropism remain poorly defined. Here, we identified tetraspanin CD9 as a conserved functional entry receptor for diverse rhabdoviruses across genera, including viral hemorrhagic septicemia virus (VHSV) ( Novirhabdovirus ), Siniperca chuatsi rhabdovirus (SCRV) ( Siniperhavirus ), and vesicular stomatitis virus (VSV) ( Vesiculovirus ). We demonstrated that the domain IV of VHSV glycoprotein G directly interacted with the large extracellular loop domain of Lateolabrax japonicus CD9 (LjCD9). CD9 knockout, CD9 protein, or CD9 antibody significantly reduced VHSV infection in vitro and CD9 knockout zebrafish, while HEK293T cells, which are nonsusceptible but permissive to VHSV, become susceptible when expressing LjCD9, suggesting that LjCD9 is an entry receptor for VHSV. We further confirmed that LjCD9 functions as a functional receptor of SCRV. Importantly, the human CD9 orthologue can serve as a receptor of VSV. Our findings also revealed that LjCD9 mediated VHSV entry via clathrin- and caveolae-mediated endocytosis. Notably, nitazoxanide (NTZ) was identified as a broad-spectrum inhibitor of VHSV, SCRV, and VSV likely by interfering with the G protein–CD9 interaction. This study establishes CD9 as a cross-species receptor for rhabdoviruses and highlights NTZ as a promising broad-spectrum antiviral agent.

The exogenous snakehead retrovirus (SnRV) is an unclassified member of the Orthoretrovirinae subfamily, discovered in cell lines derived from several fish species. SnRV resembles complex lentiviruses and potentially encodes accessory proteins, including the product of the 3’ open reading frame (3’ORF). The 3’ORF protein was suggested to function as a transactivator of transcription (Tat). Here, we constructed an infectious molecular clone for SnRV and tested the effects of 3’ORF mutations on SnRV transcription. Although replacing 3’ORF with foreign sequences strongly reduced virus expression and production, an out-of-frame point mutation in 3’ORF had only a minimal effect on SnRV replication. This latter result suggests that the 3’ORF protein does not function as Tat and that SnRV transcription is largely independent of the product of this ORF. We also show that in vitro, the SnRV promoter is versatile and robustly functioning in both fish and mammalian cultured cells. Finally, the SnRV promoter was transiently active in injected zebrafish embryos as early as the blastula stage. In transgenic zebrafish, this promoter drives enhanced expression in sensory organs and gonads, and its generated products are maternally inherited. Considering these characteristics, the SnRV promoter emerges as a promising candidate for developing versatile expression vectors applicable to research and biotechnological applications.

The first approved vaccines for human use against severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) are nanotechnology‐based. Although they are modular, rapidly produced, and can reduce disease severity, the currently available vaccines are restricted in preventing infection, stressing the global demand for novel preventive vaccine technologies. Bearing this in mind, we set out to develop a flexible nanovaccine platform for nasal administration to induce mucosal immunity, which is fundamental for optimal protection against respiratory virus infection. The next‐generation multiepitope nanovaccines co‐deliver immunogenic peptides, selected by an immunoinformatic workflow, along with adjuvants and regulators of the PD‐L1 expression. As a case study, we focused on SARS‐CoV‐2 peptides as relevant antigens to validate the approach. This platform can evoke both local and systemic cellular‐ and humoral‐specific responses against SARS‐CoV‐2. This led to the secretion of immunoglobulin A (IgA), capable of neutralizing SARS‐CoV‐2, including variants of concern, following a heterologous immunization strategy. Considering the limitations of the required cold chain distribution for current nanotechnology‐based vaccines, it is shown that the lyophilized nanovaccine is stable for long‐term at room temperature and retains its in vivo efficacy upon reconstitution. This makes it particularly relevant for developing countries and offers a modular system adaptable to future viral threats.

Secondary bacterial challenges during influenza virus infection “superinfection”) cause excessive mortality and hospitalization. Here, we present a longitudinal study of bulk gene expression changes in murine lungs during superinfection, with an initial influenza A virus infection and a subsequent Streptococcus pneumoniae infection. In addition to the well-characterized impairment of the host response, we identified superinfection-specific alterations in the global transcriptional program that are linked to the host’s ability to resist the pathogens. Particularly, whereas superinfected mice manifested an excessive rapid induction of the resistance-to-infection program, there was a substantial tissue-level rewiring of this program: upon superinfection, interferon-regulated genes were switched from positive to negative correlations with the host’s resistance state, whereas genes of fatty acid metabolism switched from negative to positive correlations with resistance states. Thus, the transcriptional resistance state in superinfection is reprogrammed toward repressed interferon signaling and induced fatty acid metabolism. Our findings suggest new insights into a tissue-level remodeling of the host defense upon superinfection, providing promising targets for future therapeutic interventions. IMPORTANCE: Secondary bacterial infections are the most frequent complications during influenza A virus (IAV) pandemic outbreaks, contributing to excessive morbidity and mortality in the human population. Most IAV-related deaths are attributed to Streptococcus pneumoniae (SP) infections, which usually begin within the first week of IAV infection in the respiratory tracts. Here, we focused on longitudinal transcriptional responses during a superinfection model consisting of an SP infection that follows an initial IAV infection, comparing superinfection to an IAV-only infection, an SP-only infection, and control treatments. Our longitudinal data allowed a fine analysis of gene expression changes during superinfection. For instance, we found that superinfected mice exhibited rapid gene expression induction or reduction within the first 12 h after encountering the second pathogen. Cell proliferation and immune response activation processes were upregulated, while endothelial processes, vasculogenesis, and angiogenesis were downregulated, providing promising targets for future therapeutic interventions. We further analyzed the longitudinal transcriptional responses in the context of a previously defined spectrum of the host’s resistance state, revealing superinfection-specific reprogramming of resistance states, such as reprogramming of fatty acid metabolism and interferon signaling. The reprogrammed functions are compelling new targets for switching the pathogenic superinfection state into a single-infection state.

Tilapia Lake Virus (TiLV) is an emerging virus lethal to tilapia, which threatens the global tilapia aquaculture with severe implications for food security. TiLV possesses similar features to orthomyxoviruses but is classified in the sole and the monotypic genus Tilapinevirus of the family Amnoonviridae. TiLV enveloped virions encapsidate a genome comprising ten segments of single-stranded, negative RNA. Remarkably, nine of TiLV’s ten major proteins lack sequence homology to any known viral or cellular proteins. The mode of TiLV entry into tilapia cells is not known. Following the measurement of the entry window of TiLV (∼3 h), we applied a panel of inhibitors of known regulators of endocytic functions to map the molecular requirements for TiLV entry. We identified productive entry by quantification of TiLV nucleoprotein expression and the generation of infectious particles. Inhibition of dynamin activity with dynasore or dynole, or depletion of cholesterol with methyl-β-cyclodextrin, strongly inhibited TiLV protein synthesis and infectious virion production. Moreover, inhibition of actin cytoskeleton polymerization with latrunculin A or microtubule polymerization with nocodazole within the entry window resulted in partial inhibition of TiLV infection. In contrast, inhibitors of endosomal acidification (NH₄Cl, bafilomycin A1, or chloroquine), an inhibitor of clathrin-coated pit assembly (pitstop 2), and erlotinib—an inhibitor of the endocytic Cyclin G-associated kinase (GAK), did not affect TiLV entry. Altogether, these results suggest that TiLV enters via dynamin-mediated endocytosis in a cholesterol-, cytoskeleton-dependent manner, and clathrin-, pH-independent manner. Thus, despite being an orthomyxo-like virus, when compared to the prototypical orthomyxovirus (influenza A virus), TiLV shows a distinct set of requirements for entry into cells.