The PhlA hemolysin from the entomopathogenic bacterium Photorhabdus luminescens belongs to the two-partner secretion family of hemolysins

Abstract

Photorhabdus is an entomopathogenic bacterium symbiotically associated with nematodes of the family Heterorhabditidae. Bacterial hemolysins found in numerous pathogenic bacteria are often virulence factors. We describe here the nucleotide sequence and the molecular characterization of the Photorhabdus luminescens phlBA operon, a locus encoding a hemolysin which shows similarities to the Serratia type of hemolysins. It belongs to the two-partner secretion (TPS) family of proteins. In low-iron conditions, a transcriptional induction of the phlBA operon was observed by using the chloramphenicol acetyltransferase reporter gene, causing an increase in PhlA hemolytic activity compared to iron-rich media. A spontaneous phase variant of P. luminescens was deregulated in phlBA transcription. The phlA mutant constructed by allelic exchange remained highly pathogenic after injection in the lepidopteran Spodoptera littoralis, indicating that PhlA hemolysin is not a major virulence determinant. Using the gene encoding green fluorescent protein as a reporter, phlBA transcription was observed in hemolymph before insect death. We therefore discuss the possible role of PhlA hemolytic activity in the bacterium-nematode-insect interactions.

FIG. 1.

Map of the P. luminescens phlBA hemolysin locus. (A) Transcription orientation of phlB and phlA (open arrows) are represented with the positions of the restriction endonucleases recognition sites used. The mutation obtained by Ω interposon insertion is indicated. Primers used and their orientation are represented by solid arrows. (B) Possible Fur-binding sites in the promoter region of the phlBA operon. The identities with the 19-nucleotide sequence of the Fur-binding consensus sequence reported previously (18) are indicated below the sequence, with matches represented by “+” and mismatches represented by “−”; the total number of identities is indicated at the end. The putative −10 box is overlined, and start codon is underlined.

FIG. 2.

Multiple alignments between the N-terminal regions of the hemolysins of Serratia marcescens (ShlA), P. luminescens (PhlA), E. tarda (EthA), Proteus mirabilis (HpmA), and H. ducreyi (HhdA) and the putative hemolysin of Y. pestis (YhlA) as determined by using Multalin version 5.4.1 (15). The total numbers of residues are as follows: 1,608 for ShlA, 1,481 for PhlA, 1,594 for EthA, 1,577 for HpmA, 1,175 for HhdA, and 1,635 for YhlA. Cystein residues are marked by an asterisk. Consensus levels: high, 90% (dark gray shading); low, 50% (light gray shading). Consensus symbols: $, any of LM; !, any of IV; %, any of FY; #, any of NDQEBZ.

FIG. 3.

PhlA hemolytic activity as represented by the percentage of hemoglobin release from horse erythrocytes. Bacterial strains were grown in LB broth supplemented with 100 μM FeSO₄ (▪) or iron depleted by the addition of 300 μM 2,2′-dipyridyl (░⃞) to an optical density at 540 nm (OD₅₄₀) of 1.0 prior to the hemolytic assay (see Materials and Methods). The data represent the mean values ± the standard error of the mean for triplicate determinations from one of four similar experiments.

FIG. 4.

RT-PCR detection of phlB, phlA, and a region overlapping the phlB and phlA open reading frames with P. luminescens TT01c RNA grown in iron-depleted LB broth. The following primers were used: lane 1, BF and BR (nucleotide positions in phlBA sequences 380 and 915, respectively); lane 2, 2F and 2R (positions 1471 and 1967); and lane 3, AF and AR (positions 3027 and 4685, see Materials and Methods).