Regulating alternative lifestyles in entomopathogenic bacteria

Abstract

Bacteria belonging to the genera Photorhabdus and Xenorhabdus participate in a trilateral symbiosis in which they enable their nematode hosts to parasitize insect larvae. The bacteria switch from persisting peacefully in a nematode's digestive tract to a lifestyle in which pathways to produce insecticidal toxins, degrading enzymes to digest the insect for consumption, and antibiotics to ward off bacterial and fungal competitors are activated. This study addresses three questions: (1) What molecular signal triggers antibiotic production in the bacteria? (2) What small molecules are regulated by the signal? And (3), how do the bacteria recognize the signal? Differential metabolomic profiling in Photorhabdus luminescens TT01 and Xenorhabdus nematophila revealed that L-proline in the insect's hemolymph initiates a metabolic shift. Small molecules known to be crucial for virulence and antibiosis in addition to previously unknown metabolites are dramatically upregulated by L-proline, linking the recognition of host environment to bacterial metabolic regulation. To identify the L-proline-induced signaling pathway, we deleted the proline transporters putP and proU in P. luminescens TT01. Studies of these strains support a model in which acquisition of L-proline both regulates the metabolic shift and maintains the bacterial proton motive force that ultimately regulates the downstream bacterial pathways affecting virulence and antibiotic production.

Figure 1

Selected P. luminescens (1-4) and X. nematophila (1, 5-9) metabolites.

Figure 2

Differential metabolomic profiling. Bacterial cultures with or without sterile-filtered insect hemolymph or L-proline were extracted after 72 hours, and organic extracts were assessed by HPLC (280 nm). Hemolymph supplementation (red traces, top) to P. luminescens (left) and X. nematophila (right) alters metabolite production compared to control cultures (blue traces). Proline supplementation (red traces, bottom) similarly regulates metabolite production in both bacteria compared to controls (blue traces). Numbers above peaks refer to metabolites in Fig. 1.

Figure 3

Fold change in metabolite production in P. luminescens (A) and X. nematophila (B) with increasing concentrations of supplementary L-proline. Numbers above curves refer to compounds in Fig. 1, and effective concentration, 50% (EC50) values are shown in red for metabolites with sigmoidal dose-response curves (error = s.d.). Metabolites were extracted at 72 hours, after the bacteria had reached stationary phase (Fig. S2). Rhabduscin (1) production remained constant in P. luminescens cultures upon addition of proline (but were stimulated in the presence of elevated osmolarity, Fig. S5).

Figure 4

Metabolic shift of wild-type (red triangles) versus ΔproU (blue squares) and ΔputP (green circles) P. luminescens with increasing proline. Stilbene production (3 and 4) was abolished in both mutants, but anthraquinone production (2) was upregulated in the mutants compared to WT (error bars = s.d.). While elevation of anthraquinone production by the ΔproU mutant was even more dramatic with addition of L-proline to the medium, levels were high, but unchanged over the proline concentration window for the ΔputP mutant, indicating relative insensitivity of the ProU transporter to high levels of proline in a non-osmotically challenged medium.

Figure 5

P. luminescens (A) and X. nematophila (B) L-proline growth effects. Bacteria were grown on rich solid media (medium used above: 2 g tryptone and 5 g yeast-extract per L) containing the dye bromothymol blue and the redox indicator TTC for 3-5 days, with and without 100 mM L-proline. Phenotypic growth effects on various media are illustrated in Figs. S8-S16. Staining patterns reveal a reddening of wild type colonies and the ΔproU mutant upon proline supplementation, indicating increased reduction of TTC. A darkening of colonies is also observed with proline, indicating increased dye uptake. Given our differential metabolic observations, it is logical that the P. luminescens ΔputP culture with proline resembled the wild-type strain grown in the absence of proline. P. luminescens ΔproU was visually similar to wild-type in these dye uptake studies in keeping with proline's proposed primary role.