Unravelling bacterial virulence factors in yeast: From identification to the elucidation of their mechanisms of action

Pathogenic bacteria employ virulence factors (VF) to establish infection and cause disease in their host. Yeasts, Saccharomyces cerevisiae and Saccharomyces pombe, are useful model organisms to study the functions of bacterial VFs and their interaction with targeted cellular processes because yeast processes and organelle structures are highly conserved and similar to higher eukaryotes. In this review, we describe the principles and applications of the yeast model for the identification and functional characterisation of bacterial VFs to investigate bacterial pathogenesis. The growth inhibition phenotype caused by the heterologous expression of bacterial VFs in yeast is commonly used to identify candidate VFs. Then, subcellular localisation patterns of bacterial VFs can provide further clues about their target molecules and functions during infection. Yeast knockout and overexpression libraries are also used to investigate VF interactions with conserved eukaryotic cell structures (e.g., cytoskeleton and plasma membrane), and cellular processes (e.g., vesicle trafficking, signalling pathways, and programmed cell death). In addition, the yeast growth inhibition phenotype is also useful for screening new drug leads that target and inhibit bacterial VFs. This review provides an updated overview of new tools, principles and applications to study bacterial VFs in yeast.

Proteomic insight into the interaction of Paenibacillus larvae with honey bee larvae before capping collected from an American foulbrood outbreak: Pathogen proteins within the host, lysis signatures and interaction markers

American foulbrood (AFB) is a devastating disease of honey bees. There remains a gap in the understanding of the interactions between the causative agent and host, so we used shotgun proteomics to gain new insights. Nano-LC-MS/MS analysis preceded visual description and Paenibacillus larvae identification in the same individual sample. A further critical part of our methodology was that larvae before capping were used as the model stage. The identification of the virulence factors SplA, PlCBP49, enolase, and DnaK in all P. larvae-positive samples was consistent with previous studies. Furthermore, the results were consistent with the array of virulence factors identified in an in vitro study of P. larvae exoprotein fractions. Although an S-layer protein and a putative bacteriocin were highlighted as important, the microbial collagenase ColA and InhA were not found in our samples. The most important virulence factor identified was isoform of neutral metalloproteinase (UniProt: V9WB82), a major protein marker responsible for the shift in the PCA biplot. This protein is associated with larval decay and together with other virulence factors (bacteriocin) can play a key role in protection against secondary invaders. Overall, this study provides new knowledge on host-pathogen interactions and a new methodical approach to study the disease.

Molecular basis of antibiotic self-resistance in a bee larvae pathogen

Paenibacillus larvae, the causative agent of the devastating honey-bee disease American Foulbrood, produces the cationic polyketide-peptide hybrid paenilamicin that displays antibacterial and antifungal activity. Its biosynthetic gene cluster contains a gene coding for the N-acetyltransferase PamZ. We show that PamZ acts as self-resistance factor in Paenibacillus larvae by deactivation of paenilamicin. Using tandem mass spectrometry, nuclear magnetic resonance spectroscopy and synthetic diastereomers, we identified the N-terminal amino group of the agmatinamic acid as the N-acetylation site. These findings highlight the pharmacophore region of paenilamicin, which we very recently identified as a ribosome inhibitor. Here, we further determined the crystal structure of PamZ:acetyl-CoA complex at 1.34 Å resolution. An unusual tandem-domain architecture provides a well-defined substrate-binding groove decorated with negatively-charged residues to specifically attract the cationic paenilamicin. Our results will help to understand the mode of action of paenilamicin and its role in pathogenicity of Paenibacillus larvae to fight American Foulbrood.

The authors show that the N-acetyltransferase PamZ acts as a self-resistance factor disabling the antibacterial paenilamicin that is produced by the honey bee larvae pathogen Paenibacillus larvae.

A Structural Approach to Anti-Virulence: A Discovery Pipeline

The anti-virulence strategy is designed to prevent bacterial virulence factors produced by pathogenic bacteria from initiating and sustaining an infection. One family of bacterial virulence factors is the mono-ADP-ribosyltransferase toxins, which are produced by pathogens as tools to compromise the target host cell. These toxins are bacterial enzymes that exploit host cellular NAD+ as the donor substrate to modify an essential macromolecule acceptor target in the host cell. This biochemical reaction modifies the target macromolecule (often protein or DNA) and functions in a binary fashion to turn the target activity on or off by blocking or impairing a critical process or pathway in the host. A structural biology approach to the anti-virulence method to neutralize the cytotoxic effect of these factors requires the search and design of small molecules that bind tightly to the enzyme active site and prevent catalytic function essentially disarming the pathogen. This method requires a high-resolution structure to serve as the model for small molecule inhibitor development, which illuminates the path to drug development. This alternative strategy to antibiotic therapy represents a paradigm shift that may circumvent multi-drug resistance in the offending microbe through anti-virulence therapy. In this report, the rationale for the anti-virulence structural approach will be discussed along with recent efforts to apply this method to treat honey bee diseases using natural products.

Anti-Virulence Strategy against the Honey Bee Pathogenic Bacterium Paenibacillus larvae via Small Molecule Inhibitors of the Bacterial Toxin Plx2A

American Foulbrood, caused by Paenibacillus larvae, is the most devastating bacterial honey bee brood disease. Finding a treatment against American Foulbrood would be a huge breakthrough in the battle against the disease. Recently, small molecule inhibitors against virulence factors have been suggested as candidates for the development of anti-virulence strategies against bacterial infections. We therefore screened an in-house library of synthetic small molecules and a library of flavonoid natural products, identifying the synthetic compound M3 and two natural, plant-derived small molecules, Acacetin and Baicalein, as putative inhibitors of the recently identified P. larvae toxin Plx2A. All three inhibitors were potent in in vitro enzyme activity assays and two compounds were shown to protect insect cells against Plx2A intoxication. However, when tested in exposure bioassays with honey bee larvae, no effect on mortality could be observed for the synthetic or the plant-derived inhibitors, thus suggesting that the pathogenesis strategies of P. larvae are likely to be too complex to be disarmed in an anti-virulence strategy aimed at a single virulence factor. Our study also underscores the importance of not only testing substances in in vitro or cell culture assays, but also testing the compounds in P. larvae-infected honey bee larvae.

The Buzz about ADP-Ribosylation Toxins from Paenibacillus larvae, the Causative Agent of American Foulbrood in Honey Bees

The Gram-positive, spore-forming bacterium Paenibacillus larvae is the etiological agent of American Foulbrood, a highly contagious and often fatal honey bee brood disease. The species P. larvae comprises five so-called ERIC-genotypes which differ in virulence and pathogenesis strategies. In the past two decades, the identification and characterization of several P. larvae virulence factors have led to considerable progress in understanding the molecular basis of pathogen-host-interactions during P. larvae infections. Among these virulence factors are three ADP-ribosylating AB-toxins, Plx1, Plx2, and C3larvin. Plx1 is a phage-born toxin highly homologous to the pierisin-like AB-toxins expressed by the whites-and-yellows family Pieridae (Lepidoptera, Insecta) and to scabin expressed by the plant pathogen Streptomyces scabiei. These toxins ADP-ribosylate DNA and thus induce apoptosis. While the presumed cellular target of Plx1 still awaits final experimental proof, the classification of the A subunits of the binary AB-toxins Plx2 and C3larvin as typical C3-like toxins, which ADP-ribosylate Rho-proteins, has been confirmed experimentally. Normally, C3-exoenzymes do not occur together with a B subunit partner, but as single domain toxins. Interestingly, the B subunits of the two P. larvae C3-like toxins are homologous to the B-subunits of C2-like toxins with striking structural similarity to the PA-63 protomer of Bacillus anthracis.

Mapping the DNA-Binding Motif of Scabin Toxin, a Guanine Modifying Enzyme from Streptomyces scabies

Scabin is a mono-ADP-ribosyltransferase toxin/enzyme and possible virulence factor produced by the agriculture pathogen, Streptomyces scabies. Recently, molecular dynamic approaches and MD simulations revealed its interaction with both NAD⁺ and DNA substrates. An Essential Dynamics Analysis identified a crab-claw-like mechanism, including coupled changes in the exposed motifs, and the R_β1-R_La-N_Lc-STT_β2-W_PN-W_ARTT-(QxE)_ARTT sequence motif was proposed as a catalytic signature of the Pierisin family of DNA-acting toxins. A new fluorescence assay was devised to measure the kinetics for both RNA and DNA substrates. Several protein variants were prepared to probe the Scabin-NAD-DNA molecular model and to reveal the reaction mechanism for the transfer of ADP-ribose to the guanine base in the DNA substrate. The results revealed that there are several lysine and arginine residues in Scabin that are important for binding the DNA substrate; also, key residues such as Asn110 in the mechanism of ADP-ribose transfer to the guanine base were identified. The DNA-binding residues are shared with ScARP from Streptomyces coelicolor but are not conserved with Pierisin-1, suggesting that the modification of guanine bases by ADP-ribosyltransferases is divergent even in the Pierisin family.

Common Mechanism for Target Specificity of Protein- and DNA-Targeting ADP-Ribosyltransferases

Many bacterial pathogens utilize ADP-ribosyltransferases (ARTs) as virulence factors. The critical aspect of ARTs is their target specificity. Each individual ART modifies a specific residue of its substrates, which could be proteins, DNA, or antibiotics. However, the mechanism underlying this specificity is poorly understood. Here, we review the substrate recognition mechanism and target residue specificity based on the available complex structures of ARTs and their substrates. We show that there are common mechanisms of target residue specificity among protein- and DNA-targeting ARTs.

The N-terminus of Paenibacillus larvae C3larvinA modulates catalytic efficiency

C3larvinA was recently described as a mono-ADP-ribosyltransferase (mART) toxin from the enterobacterial repetitive intergenic consensus (ERIC) III genotype of the agricultural pathogen, Paenibacillus larvae. It was shown to be the full-length, functional version of the previously described C3larvin_trunc toxin, due to a 33-residue extension of the N-terminus of the protein. In the present study, a series of deletions and substitutions were made to the N-terminus of C3larvinA to assess the contribution of the α₁-helix to toxin structure and function. Catalytic characterization of these variants identified Asp²³ and Ala³¹ residues as supportive to enzymatic function. A third residue, Lys³⁶, was also found to contribute to the catalytic activity of the enzyme. Analysis of the C3larvinA homology model revealed that these three residues were participating in a series of interactions to properly orient both the Q-X-E and S-T-S motifs. Ala³¹ and Lys³⁶ were found to associate with a structural network of residues previously identified in silico, whereas Asp²³ forms novel interactions not previously described. At last, the membrane translocation activity into host target cells of each variant was assessed, highlighting a possible relationship between protein dipole and target cell entry.

Development of Anti-Virulence Therapeutics against Mono-ADP-Ribosyltransferase Toxins

Mono-ADP-ribosyltransferase toxins are often key virulence factors produced by pathogenic bacteria as tools to compromise the target host cell. These toxins are enzymes that use host cellular NAD⁺ as the substrate to modify a critical macromolecule target in the host cell machinery. This post-translational modification of the target macromolecule (usually protein or DNA) acts like a switch to turn the target activity on or off resulting in impairment of a critical process or pathway in the host. One approach to stymie bacterial pathogens is to curtail the toxic action of these factors by designing small molecules that bind tightly to the enzyme active site and prevent catalytic function. The inactivation of these toxins/enzymes is targeted for the site of action within the host cell and small molecule therapeutics can function as anti-virulence agents by disarming the pathogen. This represents an alternative strategy to antibiotic therapy with the potential as a paradigm shift that may circumvent multi-drug resistance in the offending microbe. In this review, work that has been accomplished during the past two decades on this approach to develop anti-virulence compounds against mono-ADP-ribosyltransferase toxins will be discussed.

Several New Putative Bacterial ADP-Ribosyltransferase Toxins Are Revealed from In Silico Data Mining, Including the Novel Toxin Vorin, Encoded by the Fire Blight Pathogen Erwinia amylovora

Mono-ADP-ribosyltransferase (mART) toxins are secreted by several pathogenic bacteria that disrupt vital host cell processes in deadly diseases like cholera and whooping cough. In the last two decades, the discovery of mART toxins has helped uncover the mechanisms of disease employed by pathogens impacting agriculture, aquaculture, and human health. Due to the current abundance of mARTs in bacterial genomes, and an unprecedented availability of genomic sequence data, mART toxins are amenable to discovery using an in silico strategy involving a series of sequence pattern filters and structural predictions. In this work, a bioinformatics approach was used to discover six bacterial mART sequences, one of which was a functional mART toxin encoded by the plant pathogen, Erwinia amylovora, called Vorin. Using a yeast growth-deficiency assay, we show that wild-type Vorin inhibited yeast cell growth, while catalytic variants reversed the growth-defective phenotype. Quantitative mass spectrometry analysis revealed that Vorin may cause eukaryotic host cell death by suppressing the initiation of autophagic processes. The genomic neighbourhood of Vorin indicated that it is a Type-VI-secreted effector, and co-expression experiments showed that Vorin is neutralized by binding of a cognate immunity protein, VorinI. We demonstrate that Vorin may also act as an antibacterial effector, since bacterial expression of Vorin was not achieved in the absence of VorinI. Vorin is the newest member of the mART family; further characterization of the Vorin/VorinI complex may help refine inhibitor design for mART toxins from other deadly pathogens.

Characterization of C3larvinA, a novel RhoA-targeting ADP-ribosyltransferase toxin produced by the honey bee pathogen, Paenibacillus larvae

C3larvinA is a putative virulence factor produced by Paenibacillus larvae enterobacterial-repetitive-intergenic-consensus (ERIC) III/IV (strain 11-8051). Biochemical, functional and structural analyses of C3larvinA revealed that it belongs to the C3-like mono-ADP-ribosylating toxin subgroup. Mammalian RhoA was the target substrate for its transferase activity suggesting that it may be the biological target of C3larvinA. The kinetic parameters of the NAD+ substrate for the transferase (KM = 75 ± 10 µM) and glycohydrolase (GH) (KM = 107 ± 20 µM) reactions were typical for a C3-like bacterial toxin, including the Plx2A virulence factor from Paenibacillus larvae ERIC I. Upon cytoplasmic expression in yeast, C3larvinA caused a growth-defective phenotype indicating that it is an active C3-like toxin and is cytotoxic to eukaryotic cells. The catalytic variant of the Q187-X-E189 motif in C3larvinA showed no cytotoxicity toward yeast confirming that the cytotoxicity of this factor depends on its enzymatic activity. A homology consensus model of C3larvinA with NAD+ substrate was built on the structure of Plx2A, provided additional confirmation that C3larvinA is a member of the C3-like mono-ADP-ribosylating toxin subgroup. A homology model of C3larvinA with NADH and RhoA was built on the structure of the C3cer-NADH-RhoA complex which provided further evidence that C3larvinA is a C3-like toxin that shares an identical catalytic mechanism with C3cer from Bacillus cereus. C3larvinA induced actin cytoskeleton reorganization in murine macrophages, whereas in insect cells, vacuolization and bi-nucleated cells were observed. These cellular effects are consistent with C3larvinA disrupting RhoA function by covalent modification that is shared among C3-like bacterial toxins.

Comprehensive proteomic analysis of exoproteins expressed by ERIC I, II, III and IV Paenibacillus larvae genotypes reveals a wide range of virulence factors

American foulbrood is a quarantine disease of the honeybee Apis mellifera L. in many countries and contributes greatly to colony losses. We performed a label-free proteomics study of exoprotein fractions produced in vitro by Paenibacillus larvae reference strains of the ERIC I-IV genotypes. A quantitative comparison was performed of previous studied protein-based virulence factors and many newly identified putative virulence factors. Among the multiple proteases identified, key virulence factors included the microbial collagenase ColA and immune inhibitor A (InhA, an analog of the Bacillus thuringiensis protein InhA). Both of these virulence factors were detected in ERICs II-IV but were absent from ERIC I. Furthermore, the different S-layer proteins and polysaccharide deacetylases prevailed in ERICs II-IV. Thus, the expression patterns of these virulence factors corresponded with the different speeds at which honeybee larvae are known to be killed by ERICs II-IV compared to ERIC I. In addition, putative novel toxin-like proteins were identified, including vegetative insecticidal protein Vip1, a mosquitocidal toxin, and epsilon-toxin type B, which exhibit similarity to homologs present in Bacillus thuringiensis or Lysinibacillus sphaericus. Furthermore, a putative bacteriocin similar to Lactococcin 972 was identified in all assayed genotypes. It appears that P. larvae shares virulence factors similar to those of the Bacillus cereus group. Overall, the results provide novel information regarding P. larvae virulence potential, and a comprehensive exoprotein comparison of all four ERICs was performed for the first time. The identification of novel virulence factors can explain differences in the virulence of isolates.

The biological role of the enigmatic C3larvinAB toxin of the honey bee pathogenic bacterium Paenibacillus larvae

Paenibacillus larvae is the causative agent of the notifiable epizootic American foulbrood, a fatal bacterial disease of honey bee larvae. The species P. larvae has been classified into four differentially virulent and prevalent genotypes (ERIC I-IV), which also differ in their virulence factor equipment. Recently, a novel P. larvae toxin, the C3-like C3larvin, has been described. Genome analysis now revealed that the C3larvin gene is actually a part of a toxin locus encompassing two genes encoding a binary AB toxin with the A subunit being C3larvin (C3larvinA) and a putative B subunit (C3larvinB) encoded by the second gene. Sequence and structural analyses demonstrated that C3larvinB is a homologue of the Bacillus anthracis protective antigen (PA), the B subunit of anthrax toxin. The C3larvinAB toxin locus was interrupted by point mutations in all analysed P. larvae ERIC I and ERIC II strains. Only one P. larvae ERIC III/IV strain harboured an uninterrupted toxin locus comprising full-length genes for C3larvinA and B. Exposure bioassays did not substantiate a role as virulence factor for C3larvinAB in P. larvae ERIC I/II. However, the PA homologue C3larvinB had an influence on the virulence of the unique P. larvae strain expressing the functional C3larvinAB locus.

An In-Silico Sequence-Structure-Function Analysis of the N-Terminal Lobe in CT Group Bacterial ADP-Ribosyltransferase Toxins

The C3-like toxins are single-domain proteins that represent a minimal mono-ADP-ribosyl transferase (mART) enzyme with a simple model scaffold for the entire cholera toxin (CT)-group. These proteins possess a single (A-domain) that modifies Rho proteins. In contrast, C2-like toxins require a binding/translocation partner (B-component) for intoxication. These are A-only toxins that contain the E-x-E motif, modify G-actin, but are two-domains with a C-domain possessing enzymatic activity. The N-domain of the C2-like toxins is unstructured, and its function is currently unknown. A sequence-structure-function comparison was performed on the N-terminal region of the mART domain of the enzymatic component of the CT toxin group in the CATCH fold (3.90.210.10). Special consideration was given to the N-domain distal segment, the α-lobe (α₁–α₄), and its different roles in these toxin sub-groups. These results show that the role of the N-terminal α-lobe is to provide a suitable configuration (i) of the α₂–α₃ helices to feature the α3-motif that has a role in NAD⁺ substrate binding and possibly in the interaction with the protein target; (ii) the α₃–α₄ helices to provide the α_3/4-loop with protein-protein interaction capability; and (iii) the α₁-N_tail that features specialized motif(s) according to the toxin type (A-only or A-B toxins) exhibiting an effect on the catalytic activity via the ARTT-loop, with a role in the inter-domain stability, and with a function in the binding and/or translocation steps during the internalization process.

Targeting ADP-ribosylation as an antimicrobial strategy

ADP-ribosylation (ADPr) is an ancient reversible modification of cellular macromolecules controlling major biological processes as diverse as DNA damage repair, transcriptional regulation, intracellular transport, immune and stress responses, cell survival and proliferation. Furthermore, enzymatic reactions of ADPr are central in the pathogenesis of many human diseases, including infectious conditions. By providing a review of ADPr signalling in bacterial systems, we highlight the relevance of this chemical modification in the pathogenesis of human diseases depending on host-pathogen interactions. The post-antibiotic era has raised the need to find alternative approaches to antibiotic administration, as major pathogens becoming resistant to antibiotics. An in-depth understanding of ADPr reactions provides the rationale for designing novel antimicrobial strategies for treatment of infectious diseases. In addition, the understanding of mechanisms of ADPr by bacterial virulence factors offers important hints to improve our knowledge on cellular processes regulated by eukaryotic homologous enzymes, which are often involved in the pathogenesis of human diseases.

Swarming motility and biofilm formation of Paenibacillus larvae, the etiological agent of American Foulbrood of honey bees (Apis mellifera)

American Foulbrood is a worldwide distributed, fatal disease of the brood of the Western honey bee (Apis mellifera). The causative agent of this fatal brood disease is the Gram-positive, spore-forming bacterium Paenibacillus larvae, which can be classified into four different genotypes (ERIC I-IV), with ERIC I and II being the ones isolated from contemporary AFB outbreaks. P. larvae is a peritrichously flagellated bacterium and, hence, we hypothesized that P. larvae is capable of coordinated and cooperative multicellular behaviors like swarming motility and biofilm formation. In order to analyze these behaviors of P. larvae, we firstly established appropriate functional assays. Using these assays we demonstrated that P. larvae ERIC II, but not P. larvae ERIC I, was capable of swarming. Swarming motility was hampered in a P. larvae ERIC II-mutant lacking production of paenilarvin, an iturin-like lipopeptide exclusively expressed by this genotype. Both genotypes were able to form free floating biofilm aggregates loosely attached to the walls of the culture wells. Visualizing the biofilms by Congo red and thioflavin S staining suggested structural differences between the biofilms formed. Biofilm formation was shown to be independent from paenilarvin production because the paenilarvin deficient mutant was comparably able to form a biofilm.

Bacterial pathogens of bees

Pollination is an indispensable ecosystem service provided by many insects, especially by wild and managed bee species. Hence, reports on large scale honey bee colony losses and on population declines of many wild bees were alarming and resulted in increased awareness of the importance of bee health and increased interest in bee pathogens. To serve this interest, this review will give a comprehensive overview on bacterial bee pathogens by covering not only the famous pathogens (Paenibacillus larvae, Melissococcus plutonius), but also the orphan pathogens which have largely been neglected by the scientific community so far (spiroplasmas) and the pathogens which were only recently discovered as being pathogenic to bees (Serratia marcescens, Lysinibacillus sphaericus).