Bacterial endosymbiosis is widely present among zygomycetes but does not contribute to the pathogenesis of mucormycosis

Abstract

Environmental isolates of the fungus Rhizopus have been shown to harbor a bacterial endosymbiont (Burkholderia) that produces rhixozin, a plant mycotoxin. We sought to define the role of rhizoxin production by endosymbionts in the pathogenesis of mucormycosis. Endosymbiotic bacteria were identified by polymerase chain reaction in 15 (54%) of 28 clinical isolates of Zygomycetes, with 33% of the bacterial strains showing 87% identity to Burkholderia 16S rDNA. The presence of rhizoxin in myclial extracts from fungi harboring bacteria was confirmed by high-performance liquid chromatography analysis. However, fungal strains with or without endosymbionts did not differ in their ability to cause endothelial cell injury in vitro, nor did antibiotic-mediated eradication of endosymbionts and rhizoxin production decrease the virulence of fungal strains in mice or flies. In summary, although bacterial endosymbiosis is widely detected in clinical isolates of Zygomycetes, including Rhizopus oryzae strains, we found no evidence that bacterial endosymbionts and rhizoxin contribute to the pathogenesis of mucormycosis in the models studied.

Figure 1

Wide presence of bacterial symbiosis in Zygomycetes. A, Polymerase chain reaction of genomic DNA extracted from Zygomycetes, demonstrating the presence of bacterial endosymbionts. B, Dendrogram showing the close homology of 16S bacterial rDNA extracted from multiple Rhizopus species with 16S rDNA of Burkholderia.

Figure 2

High-performance liquid chromatography profiles of representative culture extracts monitored for the detection of rhizoxin. The rhizoxin peak is marked with 1. Other peaks likely represent putative rhizoxin derivatives [13]. “Plus cipro” denotes treatment of the culture with 60 µg/mL ciprofloxacin to eliminate the symbiotic bacteria.

Figure 3

Results of endothelial cell injury experiments. Although rhizoxin causes damage to endothelial cells, endothelial cell injury caused by agents of mucormycosis was neither affected by the presence of bacterial endosymbionts nor altered by bacterial eradication with ciprofloxacin treatment. A, Demonstration that pure rhizoxin toxin at concentrations ≥50 nmol/L is capable of inducing endothelial cell injury. Purified rhizoxin (Sigma) was incubated with endothelial cells for 6 h, and injury to endothelial cells was measured using our ⁵¹Cr-release assay on a 96-well plate [18]. B, Polymerase chain reaction of genomic DNA extracted from Zygomycetes, demonstrating the presence of rhizoxin-producing bacterial endosymbionts in some Rhizopus isolates but not others. C, Pooled results for endothelial cell injury caused by Rhizopus that harbor bacteria (+ bacteria) and bacteria-free Rhizopus (n = 8 in each arm). D, Eradication of bacteria by means of ciprofloxacin. Bacteria-harboring Rhizopus strains were rendered free of bacteria by treating the fungus with ciprofloxacin at 60 µg/mL, and the absence of bacteria was verified by a lack of amplificatio of 1.6 kb of 16S rDNA. E, No effect of ciprofloxacin treatment on the ability of selected Rhizopus isolates to cause endothelial cell injury. The bacteria-free organisms generated in panel D were compared with their corresponding parent strains that harbored bacteria with respect to their ability to cause endothelial cell injury (n = 8). Data are displayed as medians plus interquartile ranges. *P > .05, compared with Rhizopus-harboring bacteria.

Figure 4

No effect of the eradication of bacteria from Rhizopus by ciprofloxacin treatment on fungal virulence in diabetic ketoacidotic mice. A, Survival of mice (n = 8 per group) infected with 1 of 3 bacteria-free Rhizopus strains or their corresponding parent strains at 1 × 10⁴ spores per mouse. B, Polymerase chain reaction demonstrating the presence of bacteria in non–ciprofloxacin-treated Rhizopus and the absence of bacteria in ciprofloxacin-treated Rhizopus, both before infection and after fungal strains were retrieved from expired mice. C, Survival of mice (n = 8 per group) infected with 1 × 10⁴ spores of Rhizopus species ATCC 20577 and treated with liposomal amphotericin B (LAmB) at 5 mg/kg/day, ciprofloxacin at 80 mg/kg twice a day, or a combination of both drugs. Treatment began 24 h after infection and continued for 4 consecutive days.

Figure 5

No effect of the eradication of bacteria from Rhizopus by ciprofloxacin treatment on fungal virulence in a fly model of infection. Shown is the survival of Drosophila flies infected with 1 of 3 bacteria-free Rhizopus strains (as proved by lack of amplification of rDNA) or their corresponding parent strains. Each experiment was conducted at least in triplicate, using 25 female flies.