Protein design using structure-based residue preferences.

Recent developments in protein design rely on large neural networks with up to 100s of millions of parameters, yet it is unclear which residue dependencies are critical for determining protein function. Here, we show that amino acid preferences at individual residues-without accounting for mutation interactions-explain much and sometimes virtually all of the combinatorial mutation effects across 8 datasets (R2 ~ 78-98%). Hence, few observations (~100 times the number of mutated residues) enable accurate prediction of held-out variant effects (Pearson r > 0.80). We hypothesized that the local structural contexts around a residue could be sufficient to predict mutation preferences, and develop an unsupervised approach termed CoVES (Combinatorial Variant Effects from Structure). Our results suggest that CoVES outperforms not just model-free methods but also similarly to complex models for creating functional and diverse protein variants. CoVES offers an effective alternative to complicated models for identifying functional protein mutations.

A phage-encoded RNA-binding protein inhibits the antiviral activity of a toxin-antitoxin system.

Bacteria harbor diverse mechanisms to defend themselves against their viral predators, bacteriophages. In response, phages can evolve counter-defense systems, most of which are poorly understood. In T4-like phages, the gene tifA prevents bacterial defense by the type III toxin-antitoxin (TA) system toxIN, but the mechanism by which TifA inhibits ToxIN remains unclear. Here, we show that TifA directly binds both the endoribonuclease ToxN and RNA, leading to the formation of a high molecular weight ribonucleoprotein complex in which ToxN is inhibited. The RNA binding activity of TifA is necessary for its interaction with and inhibition of ToxN. Thus, we propose that TifA inhibits ToxN during phage infection by trapping ToxN on cellular RNA, particularly the abundant 16S rRNA, thereby preventing cleavage of phage transcripts. Taken together, our results reveal a novel mechanism underlying inhibition of a phage-defensive RNase toxin by a small, phage-encoded protein.

Selection of a de novo gene that can promote survival of Escherichia coli by modulating protein homeostasis pathways.

Cellular novelty can emerge when non-functional loci become functional genes in a process termed de novo gene birth. But how proteins with random amino acid sequences beneficially integrate into existing cellular pathways remains poorly understood. We screened ~108 genes, generated from random nucleotide sequences and devoid of homology to natural genes, for their ability to rescue growth arrest of Escherichia coli cells producing the ribonuclease toxin MazF. We identified ~2,000 genes that could promote growth, probably by reducing transcription from the promoter driving toxin expression. Additionally, one random protein, named Random antitoxin of MazF (RamF), modulated protein homeostasis by interacting with chaperones, leading to MazF proteolysis and a consequent loss of its toxicity. Finally, we demonstrate that random proteins can improve during evolution by identifying beneficial mutations that turned RamF into a more efficient inhibitor. Our work provides a mechanistic basis for how de novo gene birth can produce functional proteins that effectively benefit cells evolving under stress.

Toxic anti-phage defense proteins inhibited by intragenic antitoxin proteins.

Recombination-promoting nuclease (Rpn) proteins are broadly distributed across bacterial phyla, yet their functions remain unclear. Here we report these proteins are new toxin-antitoxin systems, comprised of genes-within-genes, that combat phage infection. We show the small, highly variable Rpn C -terminal domains (Rpn S ), which are translated separately from the full-length proteins (Rpn L ), directly block the activities of the toxic full-length proteins. The crystal structure of RpnA S revealed a dimerization interface encompassing a helix that can have four amino acid repeats whose number varies widely among strains of the same species. Consistent with strong selection for the variation, we document plasmid-encoded RpnP2 L protects Escherichia coli against certain phages. We propose many more intragenic-encoded proteins that serve regulatory roles remain to be discovered in all organisms. Significance: Here we document the function of small genes-within-genes, showing they encode antitoxin proteins that block the functions of the toxic DNA endonuclease proteins encoded by the longer rpn genes. Intriguingly, a sequence present in both long and short protein shows extensive variation in the number of four amino acid repeats. Consistent with a strong selection for the variation, we provide evidence that the Rpn proteins represent a phage defense system.

Toxic antiphage defense proteins inhibited by intragenic antitoxin proteins.

Recombination-promoting nuclease (Rpn) proteins are broadly distributed across bacterial phyla, yet their functions remain unclear. Here, we report that these proteins are toxin-antitoxin systems, comprised of genes-within-genes, that combat phage infection. We show the small, highly variable Rpn C-terminal domains (RpnS), which are translated separately from the full-length proteins (RpnL), directly block the activities of the toxic RpnL. The crystal structure of RpnAS revealed a dimerization interface encompassing α helix that can have four amino acid repeats whose number varies widely among strains of the same species. Consistent with strong selection for the variation, we document that plasmid-encoded RpnP2L protects Escherichia coli against certain phages. We propose that many more intragenic-encoded proteins that serve regulatory roles remain to be discovered in all organisms.

Marginal specificity in protein interactions constrains evolution of a paralogous family.

The evolution of novel functions in biology relies heavily on gene duplication and divergence, creating large paralogous protein families. Selective pressure to avoid detrimental cross-talk often results in paralogs that exhibit exquisite specificity for their interaction partners. But how robust or sensitive is this specificity to mutation? Here, using deep mutational scanning, we demonstrate that a paralogous family of bacterial signaling proteins exhibits marginal specificity, such that many individual substitutions give rise to substantial cross-talk between normally insulated pathways. Our results indicate that sequence space is locally crowded despite overall sparseness, and we provide evidence that this crowding has constrained the evolution of bacterial signaling proteins. These findings underscore how evolution selects for "good enough" rather than optimized phenotypes, leading to restrictions on the subsequent evolution of paralogs.

Prophage-like gene transfer agents promote Caulobacter crescentus survival and DNA repair during stationary phase.

Gene transfer agents (GTAs) are prophage-like entities found in many bacterial genomes that cannot propagate themselves and instead package approximately 5 to 15 kbp fragments of the host genome that can then be transferred to related recipient cells. Although suggested to facilitate horizontal gene transfer (HGT) in the wild, no clear physiological role for GTAs has been elucidated. Here, we demonstrate that the α-proteobacterium Caulobacter crescentus produces bona fide GTAs. The production of Caulobacter GTAs is tightly regulated by a newly identified transcription factor, RogA, that represses gafYZ, the direct activators of GTA synthesis. Cells lacking rogA or expressing gafYZ produce GTAs harboring approximately 8.3 kbp fragment of the genome that can, after cell lysis, be transferred into recipient cells. Notably, we find that GTAs promote the survival of Caulobacter in stationary phase and following DNA damage by providing recipient cells a template for homologous recombination-based repair. This function may be broadly conserved in other GTA-producing organisms and explain the prevalence of this unusual HGT mechanism.

Direct activation of a bacterial innate immune system by a viral capsid protein.

Bacteria have evolved diverse immunity mechanisms to protect themselves against the constant onslaught of bacteriophages1-3. Similar to how eukaryotic innate immune systems sense foreign invaders through pathogen-associated molecular patterns4 (PAMPs), many bacterial immune systems that respond to bacteriophage infection require phage-specific triggers to be activated. However, the identities of such triggers and the sensing mechanisms remain largely unknown. Here we identify and investigate the anti-phage function of CapRelSJ46, a fused toxin-antitoxin system that protects Escherichia coli against diverse phages. Using genetic, biochemical and structural analyses, we demonstrate that the C-terminal domain of CapRelSJ46 regulates the toxic N-terminal region, serving as both antitoxin and phage infection sensor. Following infection by certain phages, newly synthesized major capsid protein binds directly to the C-terminal domain of CapRelSJ46 to relieve autoinhibition, enabling the toxin domain to pyrophosphorylate tRNAs, which blocks translation to restrict viral infection. Collectively, our results reveal the molecular mechanism by which a bacterial immune system directly senses a conserved, essential component of phages, suggesting a PAMP-like sensing model for toxin-antitoxin-mediated innate immunity in bacteria. We provide evidence that CapRels and their phage-encoded triggers are engaged in a 'Red Queen conflict'5, revealing a new front in the intense coevolutionary battle between phages and bacteria. Given that capsid proteins of some eukaryotic viruses are known to stimulate innate immune signalling in mammalian hosts6-10, our results reveal a deeply conserved facet of immunity.

A functional selection reveals previously undetected anti-phage defence systems in the E. coli pangenome.

The ancient, ongoing coevolutionary battle between bacteria and their viruses, bacteriophages, has given rise to sophisticated immune systems including restriction-modification and CRISPR-Cas. Many additional anti-phage systems have been identified using computational approaches based on genomic co-location within defence islands, but these screens may not be exhaustive. Here we developed an experimental selection scheme agnostic to genomic context to identify defence systems in 71 diverse E. coli strains. Our results unveil 21 conserved defence systems, none of which were previously detected as enriched in defence islands. Additionally, our work indicates that intact prophages and mobile genetic elements are primary reservoirs and distributors of defence systems in E. coli, with defence systems typically carried in specific locations or hotspots. These hotspots encode dozens of additional uncharacterized defence system candidates. Our findings reveal an extended landscape of antiviral immunity in E. coli and provide an approach for mapping defence systems in other species.

The evolution of a counter-defense mechanism in a virus constrains its host range.

Bacteria use diverse immunity mechanisms to defend themselves against their viral predators, bacteriophages. In turn, phages can acquire counter-defense systems, but it remains unclear how such mechanisms arise and what factors constrain viral evolution. Here, we experimentally evolved T4 phage to overcome a phage-defensive toxin-antitoxin system, toxIN, in Escherichia coli. Through recombination, T4 rapidly acquires segmental amplifications of a previously uncharacterized gene, now named tifA, encoding an inhibitor of the toxin, ToxN. These amplifications subsequently drive large deletions elsewhere in T4's genome to maintain a genome size compatible with capsid packaging. The deleted regions include accessory genes that help T4 overcome defense systems in alternative hosts. Thus, our results reveal a trade-off in viral evolution; the emergence of one counter-defense mechanism can lead to loss of other such mechanisms, thereby constraining host range. We propose that the accessory genomes of viruses reflect the integrated evolutionary history of the hosts they infected.

The DarTG toxin-antitoxin system provides phage defence by ADP-ribosylating viral DNA.

Toxin-antitoxin (TA) systems are broadly distributed, yet poorly conserved, genetic elements whose biological functions are unclear and controversial. Some TA systems may provide bacteria with immunity to infection by their ubiquitous viral predators, bacteriophages. To identify such TA systems, we searched bioinformatically for those frequently encoded near known phage defence genes in bacterial genomes. This search identified homologues of DarTG, a recently discovered family of TA systems whose biological functions and natural activating conditions were unclear. Representatives from two different subfamilies, DarTG1 and DarTG2, strongly protected E. coli MG1655 against different phages. We demonstrate that for each system, infection with either RB69 or T5 phage, respectively, triggers release of the DarT toxin, a DNA ADP-ribosyltransferase, that then modifies viral DNA and prevents replication, thereby blocking the production of mature virions. Further, we isolated phages that have evolved to overcome DarTG defence either through mutations to their DNA polymerase or to an anti-DarT factor, gp61.2, encoded by many T-even phages. Collectively, our results indicate that phage defence may be a common function for TA systems and reveal the mechanism by which DarTG systems inhibit phage infection.

Ancestral reconstruction of duplicated signaling proteins reveals the evolution of signaling specificity.

Gene duplication is crucial to generating novel signaling pathways during evolution. However, it remains unclear how the redundant proteins produced by gene duplication ultimately acquire new interaction specificities to establish insulated paralogous signaling pathways. Here, we used ancestral sequence reconstruction to resurrect and characterize a bacterial two-component signaling system that duplicated in α-proteobacteria. We determined the interaction specificities of the signaling proteins that existed before and immediately after this duplication event and then identified key mutations responsible for establishing specificity in the two systems. Just three mutations, in only two of the four interacting proteins, were sufficient to establish specificity of the extant systems. Some of these mutations weakened interactions between paralogous systems to limit crosstalk. However, others strengthened interactions within a system, indicating that the ancestral interaction, although functional, had the potential to be strengthened. Our work suggests that protein-protein interactions with such latent potential may be highly amenable to duplication and divergence.

ssDNA is an allosteric regulator of the C. crescentus SOS-independent DNA damage response transcription activator, DriD.

DNA damage repair systems are critical for genomic integrity. However, they must be coordinated with DNA replication and cell division to ensure accurate genomic transmission. In most bacteria, this coordination is mediated by the SOS response through LexA, which triggers a halt in cell division until repair is completed. Recently, an SOS-independent damage response system was revealed in Caulobacter crescentus. This pathway is controlled by the transcription activator, DriD, but how DriD senses and signals DNA damage is unknown. To address this question, we performed biochemical, cellular, and structural studies. We show that DriD binds a specific promoter DNA site via its N-terminal HTH domain to activate transcription of genes, including the cell division inhibitor didA A structure of the C-terminal portion of DriD revealed a WYL motif domain linked to a WCX dimerization domain. Strikingly, we found that DriD binds ssDNA between the WYL and WCX domains. Comparison of apo and ssDNA-bound DriD structures reveals that ssDNA binding orders and orients the DriD domains, indicating a mechanism for ssDNA-mediated operator DNA binding activation. Biochemical and in vivo studies support the structural model. Our data thus reveal the molecular mechanism underpinning an SOS-independent DNA damage repair pathway.

A Silent Operon of Photorhabdus luminescens Encodes a Prodrug Mimic of GTP.

With the overmining of actinomycetes for compounds acting against Gram-negative pathogens, recent efforts to discover novel antibiotics have been focused on other groups of bacteria. Teixobactin, the first antibiotic without detectable resistance that binds lipid II, comes from an uncultured Eleftheria terra, a betaproteobacterium; odilorhabdins, from Xenorhabdus, are broad-spectrum inhibitors of protein synthesis, and darobactins from Photorhabdus target BamA, the essential chaperone of the outer membrane of Gram-negative bacteria. Xenorhabdus and Photorhabdus are symbionts of the nematode gut microbiome and attractive producers of secondary metabolites. Only small portions of their biosynthetic gene clusters (BGC) are expressed in vitro. To access their silent operons, we first separated extracts from a small library of isolates into fractions, resulting in 200-fold concentrated material, and then screened them for antimicrobial activity. This resulted in a hit with selective activity against Escherichia coli, which we identified as a novel natural product antibiotic, 3'-amino 3'-deoxyguanosine (ADG). Mutants resistant to ADG mapped to gsk and gmk, kinases of guanosine. Biochemical analysis shows that ADG is a prodrug that is converted into an active ADG triphosphate (ADG-TP), a mimic of GTP. ADG incorporates into a growing RNA chain, interrupting transcription, and inhibits cell division, apparently by interfering with the GTPase activity of FtsZ. Gsk of the purine salvage pathway, which is the first kinase in the sequential phosphorylation of ADG, is restricted to E. coli and closely related species, explaining the selectivity of the compound. There are probably numerous targets of ADG-TP among GTP-dependent proteins. The discovery of ADG expands our knowledge of prodrugs, which are rare among natural compounds. IMPORTANCE Drug-resistant Gram-negative bacteria have become the major problem driving the antimicrobial resistance crisis. Searching outside the overmined actinomycetes, we focused on Photorhabdus, gut symbionts of enthomopathogenic nematodes that carry up to 40 biosynthetic gene clusters coding for secondary metabolites. Most of these are silent and do not express in vitro. To gain access to silent operons, we first fractionated supernatant from Photorhabdus and then tested 200-fold concentrated material for activity. This resulted in the isolation of a novel antimicrobial, 3'-amino 3'-deoxyguanosine (ADG), active against E. coli. ADG is an analog of guanosine and is converted into an active ADG-TP in the cell. ADG-TP inhibits transcription and probably numerous other GTP-dependent targets, such as FtsZ. Natural product prodrugs have been uncommon; discovery of ADG broadens our knowledge of this type of antibiotic.

Robust and tunable signal processing in mammalian cells via engineered covalent modification cycles.

Engineered signaling networks can impart cells with new functionalities useful for directing differentiation and actuating cellular therapies. For such applications, the engineered networks must be tunable, precisely regulate target gene expression, and be robust to perturbations within the complex context of mammalian cells. Here, we use bacterial two-component signaling proteins to develop synthetic phosphoregulation devices that exhibit these properties in mammalian cells. First, we engineer a synthetic covalent modification cycle based on kinase and phosphatase proteins derived from the bifunctional histidine kinase EnvZ, enabling analog tuning of gene expression via its response regulator OmpR. By regulating phosphatase expression with endogenous miRNAs, we demonstrate cell-type specific signaling responses and a new strategy for accurate cell type classification. Finally, we implement a tunable negative feedback controller via a small molecule-stabilized phosphatase, reducing output expression variance and mitigating the context-dependent effects of off-target regulation and resource competition. Our work lays the foundation for establishing tunable, precise, and robust control over cell behavior with synthetic signaling networks.

Co-evolution of interacting proteins through non-contacting and non-specific mutations.

Proteins often accumulate neutral mutations that do not affect current functions but can profoundly influence future mutational possibilities and functions. Understanding such hidden potential has major implications for protein design and evolutionary forecasting but has been limited by a lack of systematic efforts to identify potentiating mutations. Here, through the comprehensive analysis of a bacterial toxin-antitoxin system, we identified all possible single substitutions in the toxin that enable it to tolerate otherwise interface-disrupting mutations in its antitoxin. Strikingly, the majority of enabling mutations in the toxin do not contact and promote tolerance non-specifically to many different antitoxin mutations, despite covariation in homologues occurring primarily between specific pairs of contacting residues across the interface. In addition, the enabling mutations we identified expand future mutational paths that both maintain old toxin-antitoxin interactions and form new ones. These non-specific mutations are missed by widely used covariation and machine learning methods. Identifying such enabling mutations will be critical for ensuring continued binding of therapeutically relevant proteins, such as antibodies, aimed at evolving targets.

Toxin-Antitoxin Systems as Phage Defense Elements.

Toxin-antitoxin (TA) systems are ubiquitous genetic elements in bacteria that consist of a growth-inhibiting toxin and its cognate antitoxin. These systems are prevalent in bacterial chromosomes, plasmids, and phage genomes, but individual systems are not highly conserved, even among closely related strains. The biological functions of TA systems have been controversial and enigmatic, although a handful of these systems have been shown to defend bacteria against their viral predators, bacteriophages. Additionally, their patterns of conservation-ubiquitous, but rapidly acquired and lost from genomes-as well as the co-occurrence of some TA systems with known phage defense elements are suggestive of a broader role in mediating phage defense. Here, we review the existing evidence for phage defense mediated by TA systems, highlighting how toxins are activated by phage infection and how toxins disrupt phage replication. We also discuss phage-encoded systems that counteract TA systems, underscoring the ongoing coevolutionary battle between bacteria and phage. We anticipate that TA systems will continue to emerge as central players in the innate immunity of bacteria against phage.

Escherichia coli SymE is a DNA-binding protein that can condense the nucleoid.

Type I toxin-antitoxin (TA) systems typically consist of a protein toxin that imbeds in the inner membrane where it can oligomerize and form pores that change membrane permeability, and an RNA antitoxin that interacts directly with toxin mRNA to inhibit its translation. In Escherichia coli, symE/symR is annotated as a type I TA system with a non-canonical toxin. SymE was initially suggested to be an endoribonuclease, but has predicted structural similarity to DNA binding proteins. To better understand SymE function, we used RNA-seq to examine cells ectopically producing it. Although SymE drives major changes in gene expression, we do not find strong evidence of endoribonucleolytic activity. Instead, our biochemical and cell biological studies indicate that SymE binds DNA. We demonstrate that the toxicity of symE overexpression likely stems from its ability to drive severe nucleoid condensation, which disrupts DNA and RNA synthesis and leads to DNA damage, similar to the effects of overproducing the nucleoid-associated protein H-NS. Collectively, our results suggest that SymE represents a new class of nucleoid-associated proteins that is widely distributed in bacteria.