A Photoconvertible Reporter System for Bacterial Metabolic Activity Reveals That Staphylococcus aureus Enters a Dormant-Like State to Persist within Macrophages

Abstract

Staphylococcus aureus is a leading cause of difficult-to-treat infections. The capacity of S. aureus to survive and persist within phagocytic cells is an important factor contributing to therapy failures and infection recurrence. Therefore, interfering with S. aureus intracellular persistence is key to treatment success. In this study, we used a S. aureus strain carrying the reporter mKikumeGR that enables the monitoring of the metabolic status of intracellular bacteria to achieve a better understanding of the molecular mechanisms facilitating S. aureus survival and persistence within macrophages. We found that shortly after bacteria internalization, a large fraction of macrophages harbored mainly S. aureus with high metabolic activity. This population decreased gradually over time with the concomitant increase of a macrophage subpopulation harboring S. aureus with low metabolic activity, which prevailed at later times. A dual RNA-seq analysis performed in each macrophage subpopulation showed that the host transcriptional response was similar between both subpopulations. However, intracellular S. aureus exhibited disparate gene expression profiles depending on its metabolic state. Whereas S. aureus with high metabolic activity exhibited a greater expression of genes involved in protein synthesis and proliferation, bacteria with low metabolic activity displayed a higher expression of oxidative stress response-related genes, silenced genes involved in energy-consuming processes, and exhibited a dormant-like state. Consequently, we propose that reducing metabolic activity and entering into a dormant-like state constitute a survival strategy used by S. aureus to overcome the adverse environment encountered within macrophages and to persist in the intracellular niche. IMPORTANCE The capacity of Staphylococcus aureus to survive and persist within phagocytic cells has been associated with antibiotic treatment failure and recurrent infections. Here, we investigated the molecular mechanisms leading to S. aureus persistence within macrophages using a reporter system that enables to distinguish between intracellular bacteria with high and low metabolic activity in combinstion with a dual RNA-seq approach. We found that with the progression of infection, intracellular S. aureus transitions from a high metabolic state to a low metabolic dormant-like state by turning off major energy-consuming processes while remaining viable. This process seems to be driven by the level of stress encountered in the intracellular niche. Our study indicates that effective therapies by which to treat S. aureus infections should be able to target not only high metabolic bacteria but also intracellular dormant-like S. aureus.

Keywords: Staphylococcus aureus; dual RNA-seq; intracellular survival; macrophages.

FIG 1

Monitoring the metabolic activity of S. aureus pKikume within macrophages. (A) Flow cytometry histograms of S. aureus pKikume, showing the photoconversion from green to red fluorescence using a 405 nm light pulse for 10 sec, 30 sec, 1 min, or 2 min. (B) Recovery of green fluorescence by S. aureus pKikume 30 min after photoconversion with a 405 nm light pulse for 90 sec. (C) Schematic representation of the experimental model in which macrophages infected with S. aureus pKikume were photoconverted and sampled at defined time intervals. Macrophages were infected with S. aureus pKikume for 2 h (bacterial uptake), treated with lysostaphin/gentamicin for 15 min to lyse the remaining noninternalized extracellular bacteria, and further incubated for 2 h, 4 h, and 24 h. Infected macrophages were subjected to a 90 sec light pulse of 405 nm wavelength to switch the bacterial fluorescence from green to red 30 min before sampling, and this was followed by a flow cytometry analysis. (D) Representative flow cytometry contour plots showing S. aureus pKikume-infected macrophages at progressing times after infection. (E) ATP levels in green or red S. aureus isolated from infected macrophages and determined by luciferase assay at 4 h postinfection. Wells with medium alone were used for background determination. ATP levels are expressed in relative light units (RLU) per viable bacterium. Each symbol represents one individual measurement from three independent experiments (t test, *, P < 0.05).

FIG 2

Viability and proliferation of intracellular green or red S. aureus. (A) Numbers of viable intracellular green (green symbols) or red (red symbols) S. aureus at 2 h (open symbols) and 4 h (solid symbols) postinfection. Each symbol represents an independent determination with mean values indicated by horizontal bars (t test, *, P < 0.05; **, P < 0.01). (B) Time-lapse imaging of photoconverted S. aureus pKikume within macrophages. The upper panels show the recovery of green fluorescence by proliferating bacteria, and the lower panels show the recovery of green fluorescence by nonproliferating S. aureus. Scale bar, 10 μm. (C) Confocal microscopy photograph of macrophages infected with S. aureus at 4 h postinfection. Bacteria exhibiting green fluorescence are shown in (i), bacteria exhibiting red fluorescence are shown in (ii), lysosomal compartments labeled using Alexa 647 conjugated dextran beads are shown in (iii), and a merged image is shown in (iv). Scale bars, 2 μm. (D) Lactate dehydrogenase (LDH) released in the culture supernatant of S. aureus-infected macrophages during the course of infection. Results are displayed as the percentage of the maximum LDH release achieved after the disruption of the macrophages with 0.1% Triton X-100. Each bar represents the average ± standard deviation of three independent experiments.

FIG 3

Transcriptional analysis of macrophages harboring green or red S. aureus. (A) Schematic outline of the experimental design for a dual RNA-seq analysis of the S. aureus-infected macrophages. The macrophages were infected with S. aureus for 2 h (bacterial uptake), treated with lysostaphin/gentamicin to lyse the remaining noninternalized extracellular bacteria (0 h), and further incubated for 4 h. Infected macrophages were subjected to a 90 sec light pulse of 405 nm wavelength to switch the bacterial fluorescence from green to red 30 min before sampling. Infected macrophages were harvested, and the subpopulations harboring green S. aureus were separated from the subpopulations harboring red S. aureus by cell sorting. RNA was isolated from both macrophage subpopulations and analyzed by dual RNA-seq to investigate the gene expression profiles of the host and the pathogen simultaneously. (B) Heat map showing gene expression levels (top 500 genes) in infected macrophage subpopulations harboring either green or red S. aureus as well as in uninfected macrophages. The color key represents the z score normalized transcripts per million reads (TPM). (C) PCA clustering of the transcriptomes of infected macrophages harboring green S. aureus versus uninfected macrophages (upper panel), macrophages harboring red S. aureus versus uninfected macrophages (middle panel), and macrophages harboring green versus macrophages harboring red S. aureus (lower panel) based on a Euclidean distance matrix of normalized RNA-seq data. Circles enclose replicates that cluster together. Each dot represents one biological replicate. (D) MA plots showing the log₂-fold change in gene expression for macrophages harboring green S. aureus versus uninfected macrophages (upper panel) and macrophages harboring red S. aureus versus uninfected macrophages (lower panel). Genes with adjusted P < 0.05 and log₂-fold change > 2 or log₂-fold change < −2 are labeled in green (upper panel) and red (lower panel).

FIG 4

Transcriptional analysis of intracellular green or red S. aureus. (A) Heat map showing gene expression levels (top 500 genes) in intracellular green or red S. aureus as well as in S. aureus in the infection inoculum. The color key represents the z score normalized transcripts per million reads (TPM). (B) PCA clustering of the transcriptomes of intracellular green S. aureus versus S. aureus in the infection inoculum (upper panel), intracellular red S. aureus versus S. aureus in the infection inoculum (middle panel), and intracellular green versus intracellular red S. aureus (lower panel) based on a Euclidean distance matrix of normalized RNA-seq data. Circles enclose replicates that cluster together. Each dot represents one biological replicate. (C) Enriched KEGG pathways in genes with significantly greater expression (P < 0.05) and a log₂-fold change of >1 (over the red line) or with significantly lower expression and a log₂-fold change of <−1 (under the red line) in intracellular green S. aureus versus S. aureus in the infection inoculum (upper panel) or intracellular red S. aureus versus S. aureus in the infection inoculum (lower panel). The colors of the dots reflect the P values calculated by the DAVID software program, using a modification of Fisher’s exact test. The sizes of the dots reflects the number of genes in the pathway (count). (D) Expression levels of a set of DEGs between intracellular green and intracellular red S. aureus, shown as a log₂-fold change. Positive values, depicted as green bars, indicate higher expression levels in green S. aureus, and negative values, depicted as red bars, indicate higher expression levels in red S. aureus.

FIG 5

Effect of DPI treatment on the metabolic activity of intracellular S. aureus. (A) Percentage of macrophages harboring green S. aureus in the presence (open symbols) or absence (closed symbols) of DPI. (B) Percentage of macrophages harboring red S. aureus in the presence (open symbols) or absence (closed symbols) of DPI. (C) Viability of intracellular S. aureus wild-type and saHPF mutant strains during the course of infection. Each bar represents the average ± standard deviation of three independent experiments. *, P < 0.05; **, P < 0.01; ***, P < 0.001.