Transcriptional analysis of a Photorhabdus sp. variant reveals transcriptional control of phenotypic variation and multifactorial pathogenicity in insects

Abstract

Photorhabdus luminescens lives in a mutualistic association with entomopathogenic nematodes and is pathogenic for insects. Variants of Photorhabdus frequently arise irreversibly and are studied because they have altered phenotypic traits that are potentially important for the host interaction. VAR* is a colonial and phenotypic variant displaying delayed pathogenicity when directly injected into the insect, Spodoptera littoralis. In this study, we evaluated the role of transcriptomic modulation in determining the phenotypic variation and delayed pathogenicity of VAR* with respect to the corresponding wild-type form, TT01α. A P. luminescens microarray identified 148 genes as differentially transcribed between VAR* and TT01α. The net regulator status of VAR* was found to be significantly modified. We also observed in VAR* a decrease in the transcription of genes supporting certain phenotypic traits, such as pigmentation, crystalline inclusion, antibiosis, and protease and lipase activities. Three genes encoding insecticidal toxins (pit and pirB) or putative insecticidal toxins (xnp2) were less transcribed in VAR* than in the TT01α. The overexpression of these genes was not sufficient to restore the virulence of VAR* to the levels of ΤΤ01α, which suggests that the lower virulence of VAR* does not result from impaired toxemia in insects. Three loci involved in oxidative stress responses (sodA, katE, and the hca operon) were found to be downregulated in VAR*. This is consistent with the greater sensitivity of VAR* to H(2)O(2) and may account for the impaired bacteremia in the hemolymph of S. littoralis larvae observed with VAR*. In conclusion, we demonstrate here that some phenotypic traits of VAR* are regulated transcriptionally and highlight the multifactorial nature of pathogenicity in insects.

FIG. 1.

qRT-PCR data on six genes displaying lower transcription in TT01α than in VAR*. qRT-PCR was carried out with total RNA from TT01α (white bars) and VAR* (gray bars) variants in the growth exponential phase (plain bars) and the stationary phase (hatched bars). The specific internal primers for each gene are listed in Table 2. The data are presented as a ratio, with gyrB used as the control gene. Values are means of six assays (three technical replicates on each biological replicates) and were compared using a Student t test (P > 0.05). The confidence limits are shown.

FIG. 2.

Mortality in S. littoralis. Shown is the mortality in S. littoralis infected with the TT01α P. luminescens wild-type form (○), the VAR* phenotypic variant (▪), and VAR* overexpressing toxins or potential toxin genes: the pirAB locus (✳), the pit gene (⧫), and the xnp2 gene (▴). Bacteria obtained at the end of the exponential growth phase were injected into fourth-instar larvae. Mortality values are based on data obtained after the injection of 20 larvae. Note that the virulence of the VAR* variant harboring the pBB-xnp2 plasmid is more attenuated than the VAR* virulence; this is probably due to the energy cost necessary to replicate a plasmid with a large insert (7.3 kb) compared to the 0.5- and 1.7-kb inserts of the pBB-pit and pBB-pirAB plasmids, respectively.

FIG. 3.

Bacterial growth after the injection of TT01α (○) and VAR* (▪) into S. littoralis. The graphs show the mean numbers of CFU recovered from the total hemolymph of single larvae (four larvae per time point). Larvae were each injected with 5 × 10⁴ bacteria at time zero. Errors bars indicate the standard errors of the means.