Aminoglycoside Antibiotics Inhibit Phage Infection by Blocking an Early Step of the Infection Cycle

Abstract

In response to viral predation, bacteria have evolved a wide range of defense mechanisms, which rely mostly on proteins acting at the cellular level. Here, we show that aminoglycosides, a well-known class of antibiotics produced by Streptomyces, are potent inhibitors of phage infection in widely divergent bacterial hosts. We demonstrate that aminoglycosides block an early step of the viral life cycle, prior to genome replication. Phage inhibition was also achieved using supernatants from natural aminoglycoside producers, indicating a broad physiological significance of the antiviral properties of aminoglycosides. Strikingly, we show that acetylation of the aminoglycoside antibiotic apramycin abolishes its antibacterial effect but retains its antiviral properties. Altogether, our study expands the knowledge of aminoglycoside functions, suggesting that aminoglycosides not only are used by their producers as toxic molecules against their bacterial competitors but also could provide protection against the threat of phage predation at the community level. IMPORTANCE Predation by phages is a major driver of bacterial evolution. As a result, elucidating antiphage strategies is crucial from both fundamental and therapeutic standpoints. While protein-mediated defense mechanisms, like restriction-modification systems or CRISPR/Cas, have been extensively studied, much less is known about the potential antiphage activity of small molecules. Focusing on the model bacteria Escherichia coli and Streptomyces venezuelae, our findings revealed significant antiphage properties of aminoglycosides, a major class of translation-targeting antibiotics produced by Streptomyces. Further, we demonstrate that supernatants from natural aminoglycoside producers protect bacteria from phage propagation, highlighting the physiological relevance of this inhibition. Suppression of phage infection by aminoglycosides did not result from the indirect inhibition of bacterial translation, suggesting a direct interaction between aminoglycosides and phage components. This work highlights the molecular versatility of aminoglycosides, which have evolved to efficiently block protein synthesis in bacterial competitors and provide protection against phages.

Keywords: Streptomyces; aminoglycosides; antibiotics; bacteriophages; phage defense; phage-host interaction.

FIG 1

Aminoglycosides inhibit a wide range of phages. (a) Schematic representation of the screening for the antiphage effect of different aminoglycosides. Strains resistant to the aminoglycosides were constructed using plasmid-borne resistance cassettes and subsequently challenged by phages in the presence of increasing aminoglycoside concentrations. (b) Overview of the screening results, showing the log₁₀ fold change in plaque formation by tested phages relative to the aminoglycoside-free control. Molecular structures of the aminoglycosides tested are indicated on the left. High concentrations of aminoglycosides prevented in some cases either the formation of plaque or lysis zone by the spotted phages (“no lysis”) or bacterial growth (“no lawn”). n = 2 independent biological replicates. The different phage morphologies are depicted with icons according to the following color scheme: blue, Siphoviridae; red, Myoviridae; green, Podoviridae; purple, Inoviridae; yellow, Leviviridae. (c) Exemplary pictures from propagation assays performed in the presence of the indicated aminoglycoside concentration. Results are representative of two biological replicates.

FIG 2

Aminoglycosides strongly inhibit phage amplification in liquid cultures. (a) Infection curves for Streptomyces venezuelae infected by phage Alderaan in the presence of different aminoglycosides (concentrations, in μg/mL, are indicated with subscripts; AB, antibiotic). (b) Time-lapse micrographs of S. venezuelae cultivated in a microfluidics system and challenged with Alderaan (insets show time after infection). (c) Infection curves for E. coli DSM 4230 infected by λ in the presence of 25 μg/mL apramycin. (d) Phage titers determined over two successive rounds of infection. A first infection round of S. venezuelae by Alderaan was performed in the presence or absence of apramycin. At the end of the cultivation, surviving cells from the apramycin-treated cultures were collected and exposed to phage Alderaan again, this time in the absence of apramycin. (e) Effect of MgCl₂ on infection of S. venezuelae by Alderaan, assessed by infection curves and determination of the corresponding phage titers over time. (a, d, and e) Alderaan was added to an initial titer of 10⁷ PFU/mL; (c) λ was added to an initial titer of 10⁸ PFU/mL. For growth curves and phage titers in panels a, c, d, and e, data are averages for three independent biological replicates (n = 3).

FIG 3

Secondary metabolites produced by Streptoalloteichus tenebrarius inhibit phage infection. (a) Influence of spent medium from S. tenebrarius on infection of S. venezuelae by Alderaan. Data are averages for three independent biological replicates; error bars represent standard deviations. (b) Determination of the final phage titers of infected cultures shown in panel a. Results are representative of two biological replicates. (c) Extracted ion chromatogram of samples analyzed by LC-MS assessing the presence of apramycin (molecular weight, 539.58 g/mol) in spent medium (SM) of S. tenebrarius. The indicated concentrations of apramycin are close to the detection limit under these measuring conditions. GYM, glucose-yeast extract-malt extract medium.

FIG 4

Apramycin blocks the phage life cycle at an early stage—before replication and transcription of phage DNA. (a) Scheme of the phage lytic life cycle, highlighting the different steps which could be inhibited by antiphage metabolites. (b) Infection of S. venezuelae by Alderaan; time-resolved quantification of phage DNA by qPCR in the intracellular fraction. To quantify the relative concentration of phage DNA per host DNA, a gene coding for the minor tail protein of Alderaan (HQ601_00028) and the housekeeping gene atpD of S. venezuelae were used. The corresponding oligonucleotide sequences are provided in Table S2D. Data are means for three independent biological replicates measured as technical duplicates. The range of relative concentrations measured for the uninfected controls (measured 120 min postinfection) is marked in gray. Note that the values measured for apramycin-treated samples are close to or even below the detection limit. (c) Time-resolved determination of Alderaan titers in the extracellular medium via double-agar overlays. n = 3 independent replicates. (d) RNA-seq coverage of the Alderaan genome (39 kbp) during infection in the presence and absence of apramycin.

FIG 5

Visualization of intracellular phage DNA by phage targeting direct-geneFISH. (a and c) Phage-targeting direct-geneFISH micrographs of (a) E. coli DSM4230 infected with λ and (c) S. venezuelae infected with Alderaan in the presence and absence of 25 μg/mL and 10 μg/mL apramycin, respectively. (First and third rows) Phase-contrast pictures merged with fluorescence signal from bacterial DNA (DAPI, blue) and phage DNA (Alexa647, red). (Second and fourth rows) Fluorescence signal from phage DNA only (Alexa647, red). Bar, 10 μm. (b and d) Quantification of Alexa647 fluorescence in (b) E. coli cells infected with λ and (d) S. venezuelae cells infected with Alderaan, shown as density plots of pixel counts relative to their fluorescence intensity. Data are averages for biological three independent biological replicates (n = 3); the data for all replicates are shown in Fig. S6a and b.

FIG 6

Acetylated apramycin strongly inhibits phage infection, despite the loss of its antibacterial properties. (a) Acetylation reaction of apramycin catalyzed by the AAC(3)IV acetyltransferase. (b) Total ion chromatogram and extracted ion chromatograms of samples analyzed by LC-MS assessing the presence of apramycin (molecular weight, 539.58 g/mol; m/z 540) and acetylated apramycin (molecular weight, 581.62 g/mol; m/z 582) after in vitro acetylation of apramycin. (c and d) Effect of acetylated apramycin on infection of wild-type S. venezuelae with Alderaan, performed in liquid (c) and solid (d) media. For panel d, the reaction mixtures of the in vitro acetylation assays containing apramycin, acetyl-CoA, the AAC(3)IV acetyltransferase, or different combinations of these were used to supplement the plates. A piece of paper was placed below plates to facilitate assessment of bacterial growth.