B. henselae Genomic Islands and Islets

The three B. henselae genomic islands are all flanked by tRNAs, encode mobility factors, and have a base composition slightly different from the surrounding regions. The two largest islands have a defunct copy of the prophage integrase, suggesting a phage origin. The intact prophage is most likely the result of a site-specific integration into a tRNALeu gene [37]. The islands, however, appear to have used different integration sites since they are found next to other tRNA genes.

The two largest islands encode several copies of a gene cluster consisting of genes coding for filamentous hemagglutinin fhaB) and an outer-membrane hemolysin activator/transporter (fhaC/hecB). These form a two-partner secretion system, where fhaC/hecB product mediates transport of the fhaB product. The 9-kbp island also contains hecB, adjacent to a presumably inactivated copy of an fhaB homologue.

13.5 Genomic Islands and Phages in Bartonella Species | 289

There are many phage genes present on the islands, both with and without homologues on the prophage. Examples of genes that are not present on the prophage include multiple copies of a gene homologous to a virulence-associated gene from B. melitensis and two apparently truncated genes related to a virulence locus in Photorhabdus luminescens, a symbiont of nematodes and pathogen of insects [65]. A few other genes related to the parasitism of insects are found outside the genomic islands, e.g., six copies of the phage lysozyme, which is related to a protein from a plasmid-encoded virulence determinant from the insect pathogen Serratia entomophila. This plasmid also encodes proteins homologous to insecticidal toxins from the virulence locus of P. luminescens [66], suggesting horizontal transfer between the genomes of insect-associated species.

Also encoded on the islands are a number of genes that have paralogues in the backbone of the genome, i.e., the large segments found in both B. henselae and B. quintana. This backbone copy is often located next to an integrase remnant, as for instance two inactivated copies of a DNA helicase on the largest genomic island and a region often phage genes on the intermediate island, suggesting that they might have been duplicated and relocated in the genome by the action of a phage or island. The islands also harbor a few plasmid maintenance genes, suggesting that at least some of the island genes were originally plasmid-borne. The largest islands both contain blocks of plasmid genes, and like the prophage they encode a homologue of vapA, a killer suppressor protein involved in plasmid maintenance, from a plasmid-borne virulence-associated island of the animal pathogen Dichelobacter nodosus.

Of the 68 B. henselae islets, 41 are remnants of genes present on the phage or large genomic islands, 18 carry genes or pseudogenes found elsewhere in the B. henselae genome, and 9 seem to be strictly noncoding, possibly representing former genes that are degraded beyond recognition. Twenty-five of the phage-related islets contain only an integrase remnant and are not located next to a tRNA, suggesting that they were integrated with a different mechanism than the one used by the prophage. In total, there are 43 integrase copies in various stages of decay in the B. henselae genome and they all appear to be closely related to the prophage integrase. However, since some of them are heavily degraded, it cannot be ruled out that they are remnants of a different, albeit related phage. Many of the integrase remnants represent only the 5' part of the gene [37]. A possible explanation is that this region carries the leftward phage promoter that is altering the expression of downstream genes.

B. quintana has only four integrase remnants; however, there is evidence that B. quintana once had a larger number of integrase genes. For example, B. quintana spacers orthologous to B. henselae integrase remnants are significantly longer than other spacers in the genome, perhaps indicating that they represent degraded genes [37].

Was this article helpful?

0 0

Post a comment