Milan Gerovac

Bacteria encode diverse defence systems, including CRISPR–Cas, to recognize and cleave the DNA of bacteriophages (phages) and other mobile genetic elements¹. In response, phages encode anti-CRISPR (Acr) proteins that inhibit CRISPR–Cas activity by blocking DNA binding or cleavage². Here we report an unexpected mechanism by which the anti-CRISPR AcrVA2 inhibits Cas12a biogenesis. AcrVA2 binds conserved and functionally important amino acid residues near the Cas12a N-terminus and triggers selective degradation of cas12a mRNA as it is translated. Additionally, conserved residues in the AcrVA2 C-terminal domain enable co-sedimentation with ribosomes and polysomes, which is required to achieve targeted co-translational mRNA degradation. The AcrVA2 C-terminal domain is broadly conserved in homologs encoded by diverse mobile genetic elements, typically in hosts that lack cas12a, suggesting that these homologues may recognize and downregulate alternative substrates in other bacteria. These findings reveal a novel mechanism for molecular conflict and gene regulation in bacteria.

Virulent phages, characterized by an obligatory lytic cycle resulting in the death of their bacterial hosts, are assumed to lack the ability to persist within bacterial colonies. Consequently, when bacterial isolates are sequenced, the resulting assemblies are not expected to contain virulent phage genomes. Here we test this assumption at scale by analysing over 267,000 publicly available Escherichia assemblies. Surprisingly, we identified 373 virulent phage genomes within these assemblies. Such viral genomes are associated with specific phage taxa and especially with very large (jumbo) phages such as the phage nucleus-forming family Chimalliviridae. Multiple lines of evidence from culture experiments and bioinformatic analyses suggest that these virulent phage genomes arise from persistent infections without substantial host mortality. We also identified 285 related phage genomes in bacterial taxa beyond Escherichia, highlighting the frequency of as-yet-uncharacterized persistent phages. Our findings challenge the traditional virulent–temperate dichotomy and point to the prevalence of non-canonical phage lifestyles.

Bacteriophages are the most abundant entities on earth and exhibit vast genetic and phenotypic diversity. Exploitation of this largely unexplored molecular space requires identification and functional characterization of genes that act at the phage–host interface. So far, this has been restricted to few model phage–host systems that are amenable to genetic manipulation. Here, to overcome this limitation, we introduce a non-genetic mRNA targeting approach using exogenous delivery of programmable antisense oligomers to silence genes of DNA and RNA phages. A systematic knockdown screen of core and accessory genes of the nucleus-forming jumbo phage ΦKZ, coupled to RNA-sequencing and microscopy analyses, reveals previously unrecognized proteins that are essential for phage propagation and that, upon silencing, elicit distinct phenotypes at the level of the phage and host response. One of these factors is the RNase H-like protein ΦKZ155 (also known as Nlp2), which acts at a major decision point during infection, linking the formation of the protective phage nucleus to phage genome amplification. This non-genetic antisense oligomer-based gene silencing method promises to be a versatile tool for molecular discovery in phage biology, will help to elucidate defence and anti-defence mechanisms in non-model phage–host pairs, and offers potential for optimizing phage therapy and biotechnological procedures.

Bacteriophages are the most abundant entities on earth and exhibit vast genetic and phenotypic diversity. Exploitation of this largely unexplored molecular space requires identification and functional characterisation of genes that act at the phage-host interface. To date, this is restricted to few model phage-host systems that are amenable to genetic manipulation. To overcome this limitation, we introduce a non-genetic mRNA targeting approach using exogenous delivery of programmable antisense oligomers (ASOs) to silence genes of both DNA and RNA phages. A systematic knockdown screen of core and accessory genes of the nucleus-forming jumbo phage ΦKZ, coupled to RNA-sequencing and microscopy analyses, reveals previously unrecognised proteins that are essential for phage replication and that, upon silencing, elicit distinct phenotypes at the level of the phage and host response. One of these factors is a ΦKZ-encoded nuclease that acts at a major decision point during infection, linking the formation of the protective phage nucleus to phage genome amplification. The non-genetic ASO-based gene silencing used here promises to be a versatile tool for molecular discovery in phage biology, will help elucidate defence and anti-defence mechanisms in non-model phage-host pairs, and offers potential for optimising phage therapy and biotechnological procedures.

Antibacterial proteins inhibiting Pseudomonas aeruginosa have been identified in various phages and explored as antibiotic alternatives. Here, we isolated a phiKZ-like phage, Churi, which encodes 364 open reading frames. We examined 15 early-expressed phage proteins for their ability to inhibit bacterial growth, and found that gp335, closely related to phiKZ-gp14, exhibits antibacterial activity. Similar to phiKZ-gp14, recently shown to form a complex with the P. aeruginosa ribosome, we predict experimentally that gp335 interacts with ribosomal proteins, suggesting its involvement in protein translation. GFP-tagged gp335 clusters around the phage nucleus as early as 15 minutes post-infection and remains associated with it throughout the infection, suggesting its role in protein expression in the cell cytoplasm. CRISPR-Cas13-mediated deletion of gp355 reveals that the mutant phage has a prolonged latent period. Altogether, we demonstrate that gp335 is an antibacterial protein of nucleus-forming phages that associates with the ribosomes at the phage nucleus.

Bacteriophages must seize control of the host gene expression machinery to replicate. To bypass bacterial anti-phage defence systems, this host takeover occurs immediately upon infection. A general understanding of phage mechanisms for immediate targeting of host transcription and translation processes is lacking. Here we introduce an integrative high-throughput approach to uncover phage-encoded proteins that target the gene expression machinery of Pseudomonas aeruginosa immediately upon infection with the jumbo phage ΦKZ. By integrating biochemical, genetic and structural analyses, we identify an abundant and conserved phage factor ΦKZ014 that targets the large ribosomal subunit by binding the 5S ribosomal RNA, and rapidly promotes replication in several clinical isolates. ΦKZ014 is among the earliest ΦKZ proteins expressed after infection and remains bound to ribosomes during the entire translation cycle. Our study provides a strategy to decipher molecular components of phage-mediated host takeover and argues that phage genomes represent an untapped discovery space for proteins that modulate the host gene expression machinery.

The obligate anaerobic, enteric pathogen Clostridioides difficile persists in the intestinal tract by forming antibiotic‐resistant endospores that contribute to relapsing and recurrent infections. Despite the importance of sporulation for C. difficile pathogenesis, environmental cues and molecular mechanisms that regulate sporulation initiation remain ill‐defined. Here, by using RIL‐seq to globally capture the Hfq‐dependent RNA–RNA interactome, we discovered a network of small RNAs that bind to mRNAs encoding sporulation‐related genes. We show that two of these small RNAs, SpoX and SpoY, regulate translation of the master regulator of sporulation, Spo0A, in an opposing manner, which ultimately leads to altered sporulation rates. Infection of antibiotic‐treated mice with SpoX and SpoY deletion mutants revealed a global effect on gut colonization and intestinal sporulation. Our work uncovers an elaborate RNA–RNA interactome controlling the physiology and virulence of C. difficile and identifies a complex post‐transcriptional layer in the regulation of spore formation in this important human pathogen.

Bacteriophages must seize control of the host gene expression machinery to promote their own protein synthesis. Since the bacterial hosts are armed with numerous anti-phage defence systems, it is essential that mechanisms of host take-over act immediately upon infection. Although individual proteins that modulate components of the bacterial gene expression apparatus have been described in several different phages, systematic approaches which capture the phage’s arsenal for immediate targeting of host transcription and translation processes have been lacking. In particular, there are no known phage factors that associate directly with host ribosomes to modulate protein synthesis. Here, we take an integrative high-throughput approach to uncover numerous new proteins encoded by the jumbo phage ΦKZ that target the gene expression machinery of the Gram-negative human pathogen Pseudomonas aeruginosa immediately upon infection. By integrating biochemical and structural analyses, we identify a conserved phage factor that associates with the large ribosomal subunit by binding the 5S ribosomal RNA. This highly abundant factor is amongst the earliest ΦKZ proteins expressed after infection and stays bound to ribosomes during the entire translation cycle. Our study provides a general strategy to decipher molecular components of phage-mediated host take-over and argues that phage genomes represent a large discovery space for proteins that modulate the host gene expression machinery.

Enterococcus faecalis and Enterococcus faecium are major nosocomial pathogens. Despite their relevance to public health and their role in the development of bacterial antibiotic resistance, relatively little is known about gene regulation in these species. RNA–protein complexes serve crucial functions in all cellular processes associated with gene expression, including post-transcriptional control mediated by small regulatory RNAs (sRNAs). Here, we present a new resource for the study of enterococcal RNA biology, employing the Grad-seq technique to comprehensively predict complexes formed by RNA and proteins in E. faecalis V583 and E. faecium AUS0004. Analysis of the generated global RNA and protein sedimentation profiles led to the identification of RNA–protein complexes and putative novel sRNAs. Validating our data sets, we observe well-established cellular RNA–protein complexes such as the 6S RNA–RNA polymerase complex, suggesting that 6S RNA-mediated global control of transcription is conserved in enterococci. Focusing on the largely uncharacterized RNA-binding protein KhpB, we use the RIP-seq technique to predict that KhpB interacts with sRNAs, tRNAs, and untranslated regions of mRNAs, and might be involved in the processing of specific tRNAs. Collectively, these datasets provide departure points for in-depth studies of the cellular interactome of enterococci that should facilitate functional discovery in these and related Gram-positive species. Our data are available to the community through a user-friendly Grad-seq browser that allows interactive searches of the sedimentation profiles (https://resources.helmholtz-hiri.de/gradseqef/).