Bacteriophage 7-7-1, a member of the family, infects the soil bacterium sp. H13-3. Infection requires attachment to actively rotating bacterial flagellar filaments, with flagellar number, length, and rotation speed being important determinants for infection efficiency. To identify secondary receptor(s) on the cell surface, we isolated motile, phage-resistant sp. H13-3 transposon mutants. Transposon insertion sites were pinpointed using arbitrary-primed polymerase chain reaction and bioinformatics analyses. Three genes were recognized, whose corresponding proteins had the following computationally predicted functions: AGROH133_07337, a glycosyl transferase, AGROH133_13050, a UDP-glucose 4-epimerase, and AGROH133_08824, an integral cytoplasmic membrane protein. The first two gene products are part of the lipopolysaccharide (LPS) synthesis pathway, while the latter is predicted to be a relatively small (13.4 kDa) cytosolic membrane protein with up to four transmembrane helices. Phenotypes of transposon mutants were verified by complementation and site-directed mutagenesis. Additional characterization of motile, phage resistant mutants is also described. Given these findings, we propose a model for sp. H13-3 infection by bacteriophage 7-7-1 where the phage initially attaches to the flagellar filament and is propelled down towards the cell surface by clockwise flagellar rotation. The phage then attaches to and degrades the LPS to reach the outer membrane, and ejects its DNA into the host using its syringe-like contractile tail. We hypothesize that the integral membrane protein plays an important role in events following viral DNA ejection or in LPS processing and/or deployment. The proposed two-step attachment mechanism may be conserved among other flagellotropic phages infecting Gram-negative bacteria. Flagellotropic bacteriophages belong to tailed phage order , which are the most abundant in the virome. While it is known that these viruses adhere to the bacterial flagellum and use flagellar rotation to reach the cell surface, their infection mechanisms are poorly understood. Characterizing flagellotropic phage-host interactions is crucial to understanding how microbial communities are shaped. Using a transposon mutagenesis approach combined with a screen for motile, phage-resistant mutants, we identified lipopolysaccharides as the secondary cell surface receptor for phage 7-7-1. This is the first cell surface receptor identified for flagellotropic phages. One hypothetical membrane protein was also recognized as essential for infection. These new findings, together with previous results, culminated in an infection model for phage 7-7-1.

The flagellotropic phage 7-7-1 specifically adsorbs to Agrobacterium sp. strain H13-3 (formerly Rhizobium lupini H13-3) flagella for efficient host infection. The Agrobacterium sp. H13-3 flagellum is complex and consists of three flagellin proteins: the primary flagellin FlaA, which is essential for motility, and the secondary flagellins FlaB and FlaD, which have minor functions in motility. Using quantitative infectivity assays, we showed that absence of FlaD had no effect on phage infection, while absence of FlaB resulted in a 2.5-fold increase in infectivity. A flaA deletion strain, which produces straight and severely truncated flagella, experienced a significantly reduced infectivity, similar to that of a flaB flaD strain, which produces a low number of straight flagella. A strain lacking all three flagellin genes is phage resistant. In addition to flagellation, flagellar rotation is required for infection. A strain that is nonmotile due to an in-frame deletion in the gene encoding the motor component MotA is resistant to phage infection. We also generated two strains with point mutations in the motA gene resulting in replacement of the conserved charged residue Glu98, which is important for modulation of rotary speed. A change to the neutral Gln caused the flagellar motor to rotate at a constant high speed, allowing a 2.2-fold-enhanced infectivity. A change to the positively charged Lys caused a jiggly motility phenotype with very slow flagellar rotation, which significantly reduced the efficiency of infection. In conclusion, flagellar number and length, as well as speed of flagellar rotation, are important determinants for infection by phage 7-7-1.

A B S T R A C TBacteria, including strains of Salmonella, have been researched and applied as therapeutic cancer agents for centuries. Salmonella are particularly of interest due to their facultative anaerobic nature, facilitating colonization of differentially oxygenated tumor regions. Additionally, Salmonella can be manipulated with relative ease, resulting in the ability to attenuate the pathogen or engineer vectors for drug delivery. It was recently discovered that the anti-cancer Salmonella enterica serovar Typhimurium strain VNP20009 is lacking in chemotactic ability, due to a non-synonymous single nucleotide polymorphism in cheY. Replacing the mutated copy of cheY with the wild-type sequence restored chemotaxis to 70% of the parental strain. We aimed to investigate further if chemotaxis of VNP20009 can be optimized. By restoring the gene msbB in VNP20009 cheY + , which confers attenuation by lipid A modification, we observed a 9% increase in swimming speed, 13% increase in swim plate performance, 19% increase in microfluidic device partitioning towards the attractant at the optimum concentration gradient, and mitigation of a non-motile cell subpopulation. We conclude that chemotaxis can be enhanced further but at the cost of changing one defining characteristic of VNP20009. A less compromised strain might be needed to employ for investigating bacterial chemotaxis in tumor interactions.

No abstract

A variety of bacterial strains have been evaluated as bio-therapeutic and immunomodulatory agents to treat cancer. One such strain, Salmonella enterica serovar Typhimurium VNP20009, which is attenuated by a purine auxotrophic mutation and modified lipid A, is characterized in previous models as a safely administered, tumor colonizing agent. However, earlier work tended to use less aggressive cancer cell lines and immunocompromised animal models. Here, we investigated the safety and efficacy of VNP20009 in a highly malignant murine model of human breast cancer. Additionally, as VNP20009 has recently been found to have a defective chemotaxis system, we tested whether restoring chemotaxis would improve anti-cancer properties in this model system. Exposure to VNP20009 had no significant effect on primary mammary tumor size or pulmonary metastasis, and the tumor colonizing process appeared chemotaxis independent. Moreover, tumor-bearing mice exposed to Salmonella exhibited increased morbidity that was associated with significant liver disease. Our results suggest that VNP20009 may not be safe or efficacious when used in aggressive, metastatic breast cancer models utilizing immunocompetent animals.