Prophages and Past Prophage-Host Interactions Revealed by CRISPR Spacer Content in a Fish Pathogen

The role of prophages in the evolution, diversification, or virulence of the fish pathogen Flavobacterium columnare has not been studied thus far. Here, we describe a functional spontaneously inducing prophage fF4 from the F. columnare type strain ATCC 23463, which is not detectable with commonly used prophage search methods. We show that this prophage type has a global distribution and is present in strains isolated from Finland, Thailand, Japan, and North America. The virions of fF4 are myoviruses with contractile tails and infect only bacterial strains originating from Northern Finland. The fF4 resembles transposable phages by similar genome organization and several gene orthologs. Additional bioinformatic analyses reveal several species in the phylum Bacteroidetes that host a similar type of putative prophage, including bacteria that are important animal and human pathogens. Furthermore, a survey of F. columnare Clustered Regularly Interspaced Short Palindromic Repeats (CRISPR) spacers indicate a shared evolutionary history between F. columnare strains and the fF4 phage, and another putative prophage in the F. columnare strain ATCC 49512, named p49512. First, CRISPR spacer content from the two CRISPR loci (types II-C and VI-B) of the fF4 lysogen F. columnare ATCC 23463 revealed a phage terminase protein-matching spacer in the VI-B locus. This spacer is also present in two Chinese F. columnare strains. Second, CRISPR analysis revealed four F. columnare strains that contain unique spacers targeting different regions of the putative prophage p49512 in the F. columnare strain ATCC 49512, despite the geographical distance or genomovar of the different strains. This suggests a common ancestry for the F. columnare prophages and different host strains.

Distinct cavemoviruses interact synergistically with sweet potato chlorotic stunt virus (genus Crinivirus) in cultivated sweet potato

Two serologically unrelated sweet potato viruses causing symptoms of vein clearing in the indicator plant Ipomoea setosa were isolated and their genomes have been sequenced. They are associated with symptomless infections in sweet potato but distinct vein-clearing symptoms and higher virus titres were observed when these viruses co-infected with sweet potato chlorotic stunt virus (SPCSV), a virus that is distributed worldwide and is a mediator of severe virus diseases in this crop. Molecular characterization and phylogenetic analysis revealed an overall nucleotide identity of 47.6 % and an arrangement of the movement protein and coat protein domains characteristic of members of the genus Cavemovirus, in the family Caulimoviridae. We detected both cavemoviruses in cultivated sweet potato from East Africa, Central America and the Caribbean islands, but not in samples from South America. One of the viruses characterized showed a similar genome organization as, and formed a phylogenetic sublineage with, tobacco vein clearing virus (TVCV), giving further support to the previously suggested separation of TVCV, and related viral sequences, into a new caulimovirid genus. Given their geographical distribution and previous reports of similar but yet unidentified viruses, sweet potato cavemoviruses may co-occur with SPCSV more often than previously thought and they could therefore contribute to the extensive yield losses and cultivar decline caused by mixed viral infections in sweet potato.

Specificity and functional interaction of the polymerase complex proteins of human and avian metapneumoviruses

Human metapneumovirus (HMPV) and avian metapneumovirus (AMPV) have a similar genome organization and protein composition, but a different host range. AMPV subgroup C (AMPV-C) is more closely related to HMPV than other AMPVs. To investigate the specificity and functional interaction of the polymerase complex proteins of human and avian metapneumoviruses, a minireplicon system was generated for AMPV-C and used in combination with minireplicon systems for HMPV lineages A1 and B1. Viral RNA-like molecules representing HMPV-A1 and -B1, AMPV-A and -C and human respiratory syncytial virus were replicated efficiently by polymerase complexes of HMPV-A1 and -B1 and AMPV-C, but not by polymerase complexes of bovine parainfluenza virus 3. Upon exchange of HMPV and AMPV-C polymerase complex components, all chimeric polymerase complexes were functional; exchange between HMPVs did not result in altered polymerase activity, whereas exchange between HMPVs and AMPV-C did. Recombinant HMPV-B1 viruses in which polymerase genes were exchanged with those of HMPV-A1 replicated with normal kinetics in vitro, whilst replacement with AMPV-C genes resulted in moderate differences in virus replication. In hamsters, recombinant HMPV-B1 viruses in which individual polymerase genes were exchanged with those of AMPV-C were attenuated, irrespective of the results obtained with minireplicon systems or in vitro replication assays. This study provides insight into the specificity and functional interaction of polymerase complex proteins of human and avian metapneumoviruses, but neither minireplicon systems nor in vitro replication kinetics were found to be predictive for attenuation in permissive animals.

Molecular Characterization, Phylogenetic Relationships, and Specific Detection of Peach mosaic virus

Peach mosaic virus (PcMV) and Cherry mottle leaf virus (CMLV) are serologically related viruses that cause distinct diseases, have a different host range, and are vectored by different eriophyid mites. Sequence analysis of the genome of PcMV indicates that it is closely related genetically to CMLV but distinct, with similar genome organization and a member of the genus Trichovirus. The genome of PcMV consists of 7,988 nucleotides, excluding a poly(A) tail at the 3′ end of the genome. Four putative open reading frames (ORF1 to 4) were identified coding for proteins of 216.3, 47.2, 21.7, and 15.7 kDa, respectively. Also, three noncoding regions were identified, including an intergenic region separating ORF3 and ORF4. The complete nucleotide sequence of PcMV shares 73% identity with CMLV. The CP amino acid sequence identity between isolates of PcMV ranged from 97 to 99% versus 83% identity when compared with the CP of CMLV. In vitro expression and subsequent western blot analysis confirmed ORF3 as encoding the CP gene of PcMV. Phylogenetic analysis supports classification of PcMV and CMLV as members of the genus Trichovirus. They are unique members of this genus with an extra ORF (ORF4). PcMV ORF4 appears to code for a putative nucleic acid-binding (NB) protein which has identity with the NB protein of CMLV and members of the genera Allexivirus, Carlavirus, and Vitivirus. PcMV and CMLV appear to be the products of recombination between members of the genus Trichovirus and a virus group containing the putative NB protein. Alternatively, PcMV and CMLV may represent the intact genome, with a deletion event producing members that lack ORF4. A reverse transcription-polymerase chain reaction procedure was developed for reliable and specific detection of PcMV. This will be an asset for stone fruit virus certification.

TheglcBlocus ofRhizobium leguminosarumVF39 encodes an arabinose-inducible malate synthase

In the course of a study conducted to isolate genes upregulated by plant cell wall sugars, we identified an arabinose-inducible locus from a transcriptional fusion library of Rhizobium leguminosarum VF39, carrying random insertions of the lacZ transposon Tn5B22. Sequence analysis of the locus disrupted by the transposon revealed a high similarity to uncharacterized malate synthase G genes from Sinorhizobium meliloti, Agrobacterium tumefaciens, and Mesorhizobium loti. This enzyme catalyzes the condensation of glyoxylate and acetyl-CoA to yield malate and CoA and is thought to be a component of the glyoxylate cycle, which allows microorganisms to grow on two carbon compounds. Enzyme assays showed that a functional malate synthase is encoded in the glcB gene of R. leguminosarum and that its expression is induced by arabinose, glycolate, and glyoxylate. An Escherichia coli aceB glcB mutant, complemented with the R. leguminosarum PCR-amplified gene, recovered malate synthase activity. A very similar genome organization of the loci containing malate synthase and flanking genes was observed in R. leguminosarum, S. meliloti, and A. tumefaciens. Pea plants inoculated with the glcB mutant or the wild-type strain showed no significant differences in nitrogen fixation. This is the first report regarding the characterization of a mutant in one of the glyoxylate cycle enzymes in the rhizobia.Key words: Rhizobium, malate synthase, glyoxylate cycle, arabinose metabolism.