Non-deacetylated poly- N-acetylglucosamine-hyperproducing Staphylococcus aureus undergoes immediate autoaggregation upon vortexing

Abstract

Biofilms are microbial communities of cells embedded in a matrix of extracellular polymeric substances generated and adhering to each other or to a surface. Cell aggregates formed in the absence of a surface and floating pellicles that form biofilms at the air-liquid interface are also considered to be a type of biofilm. Staphylococcus aureus is a well-known cause of biofilm infections and high-molecular-weight polysaccharides, poly-N-acetylglucosamine (PNAG) is a main constituent of the biofilm. An icaADBC operon comprises major machinery to synthesize and extracellularly secrete PNAG. Extracellular PNAG is partially deacetylated by IcaB deacetylase, and the positively charged PNAG hence interacts with negatively charged cell surface to form the major component of biofilm. We previously reported a new regulator of biofilm (Rob) and demonstrated that Rob binds to a unique 5-bp motif, TATTT, present in intergenic region between icaADBC operon and its repressor gene icaR in Yu et al. The deletion of the 5-bp motif induces excessive adherent biofilm formation. The real function of the 5-bp motif is still unknown. In an attempt to isolate the 5-bp motif deletion mutant, we isolated several non-adherent mutants. They grew normally in turbid broth shaking culture but immediately auto-aggregated upon weak vortexing and sedimented as a lump resulting in a clear supernatant. Whole genome sequencing of the mutants identified they all carried mutations in icaB in addition to deletion of the 5-bp motif. Purification and molecular characterization of auto-aggregating factor in the culture supernatant of the mutant identified that the factor was a massively produced non-deacetylated PNAG. Therefore, we created a double deficient strain of biofilm inhibitory factors (5-bp motif, icaR, rob) and icaB to confirm the aggregation phenomenon. This peculiar phenomenon was only observed in Δ5bpΔicaB double mutant but not in ΔicaR ΔicaB or ΔrobΔicaB mutant. This study explains large amount of extracellularly produced non-deacetylated PNAG by Δ5bpΔicaB double mutation induced rapid auto-aggregation of S. aureus cells by vortexing. This phenomenon indicated that Staphylococcus aureus may form biofilms that do not adhere to solid surfaces and we propose this as a new mechanism of non-adherent biofilm formation of S. aureus.

Keywords: PNAG; Staphylococcus aureus; aggregation; biofilm; icaB; poly-N-acetylglucosamine.

Figure 1

Autoaggregation assay of S. aureus FK300 5 bp deletion mutant culture. Autoaggregation assay of S. aureus FK300 5-bp deletion mutant culture. During the process of isolating the 5-bp (icaR-icaA intergenic region) deletion mutant from S. aureus FK300, we obtained colonies showing a distinct phenotype and the isolates were designated as Δ5bpBm 1, 2, and 3. These isolates were cultured in TSB for 6 h (before), and vortexed for 10 s (after), following which images of the culture were obtained (Supplementary Video 1). Shown is an image of the shape of the colonies growing on TSA plate. Left, genuine Δ5bp mutant; right, Δ5bpBm 2. Other Δ5bpBm mutants revealed a similar phenotype (Supplementary Figure 1).

Figure 2

Dysfunction of IcaB and advanced biofilm production cause aggregation. Combination of IcaB dysfunction and 5-bp deletion cause autoaggregation. Autoaggregation assay (A) and biofilm assay (B) of various Staphylococcus aureus FK300 and the associated mutants: FK300Δ5bp; FK300Δ5bpBm 2; FK300Δ5bpΔicaB; and FK300Δ5bpBm 2 complemented with pKAT carrying icaR (picaR), icaB (picaB), or pKAT, whereas FK300Δ5bpΔicaB complemented with picaB or pKAT. Autoaggregation assay was carried out as shown in Figure 1. Biofilm assay was conducted using a microtiter plate. Bacteria were grown in TSB in the presence (Glc) or absence (Glc-) of 1% glucose. Biofilm stained with crystal violet was solubilized and OD 590 nm was measured using the polystyrene microtiter plate as described in the Methods section. Bars indicate mean values, error bars indicate standard error of the mean (n = 3). WT, wild type strain FK300. pKAT, empty vector control.

Figure 3

Comparison of each biofilm formation inhibitory factor and icaB double deficient strains. Autoaggregation and biofilm formation of Δ5bpΔicaB, ΔicaRΔicaB, and ΔrobΔicaB double mutants. Autoaggregation assay (A) and biofilm formation (B) of FK300 and FK300Δ5bp, FK300Δ5bpBm 2, FK300Δ5bpΔicaB, FK300ΔicaR, FK300ΔicaRΔicaB, FK300Δrob, FK300ΔrobΔicaB and FK300ΔicaB. Status before and after mixing for a few seconds on a vortex mixer is shown. Each bacterium was cultured in TSB medium at 37°C for 6 h with shaking. (B) Bacteria were grown in TSB in the presence (Glc) or absence (Glc-) of 1% glucose. Biofilm formation was measured using the polystyrene microtiter plate assay as described in the Methods section. Bars indicate mean values, error bars indicate standard error of the mean (n = 3). WT; wild type strain FK300.

Figure 4

Measurements of icaA transcription by qPCR. Ica operon expression of hyper-biofilm elaborating mutants. Total RNA preparation, cDNA synthesis, and quantitative PCR were performed as described in the Methods section. Relative expression level of ica operon; transcript levels in the Δ5bp, icaR, or rob deletion mutants as compared to those in WT strain FK300 are shown. The expression of gyrB was used for sample normalization. Bars indicate mean values, error bars indicate standard error of the mean (n = 3). Kruskal-Wallis test results p < 0.05. There were significant differences between all deficient strains and WT. *: p < 0.05 (Mann–Whitney U test). WT; wild type strain FK300.

Figure 5

PIA / PNAG production on the culture supernatant and cell surface in each strain. PIA/PNAG production in the culture supernatant and on cell surface of the mutants. PIA/PNAG collected from the culture supernatant and cell surface of FK300 and FK300Δ5bpΔicaB, FK300ΔicaRΔicaB, FK300ΔrobΔicaB, FK300Δ5bp, FK300ΔicaR, FK300Δrob and FK300ΔicaB cultured for 6 h were evaluated via dot blotting on nitrocellulose using rabbit antibody against PIA. (A) supernatant, (B) cell surface. WT; wild type strain FK300.

Figure 6

Swapping experiment of culture supernatant and bacterial cells. Swapping of culture supernatant and bacterial cells. Filtered culture supernatant of FK300, FK300Δ5bp, or FK300Δ5bpΔicaB was incubated with bacterial cells of FK300 or FK300Δ5bpΔicaB (before) and vortexed for 10 s (after), and photographic images of the culture were obtained. Aggregates identified after vortexing were indicated by red arrows. WT; wild type strain FK300.

Figure 7

Searching for autoaggregation factors using electron microscopy and elemental analysis. Morphological and elemental analysis of autoaggregation factor(s) using electron microscopy. (A) Ultrastructural observation of Staphylococcus aureus FK300Δ5bpBm 2 using transmission electron microscope. S. aureus FK300 (left) and S. aureus FK300Δ5bpBm 2 (right). (B) Elemental analysis of the autoaggregate from FK300Δ5bpΔicaB supernatant. Autoaggregates obtained by vortexing the culture supernatant of FK300Δ5bpΔicaB were dried and examined via a scanning electron microscope (Mini Scope TM-3030) for element analysis. As a control, the value of N-acetylglucosamine is shown. WT; wild type strain FK300. The top left figure is SEM image of a sample irradiated with characteristic X-rays. The bottom left figure is EDS spectrum. The right is a table of elemental ratios of flocculence samples and N-acetylglucosamine.

Figure 8

Gel permeation high performance liquid chromatography analyses of culture supernatants of FK300Δ5bpΔicaB and WT. Gel permeation high performance liquid chromatography analyses of culture supernatants of FK300Δ5bpΔicaB and WT. Concentrated culture supernatants of FK300Δ5bpΔicaB and FK300 were subjected to gel permeation chromatography (GPC) using Shim-pack Diol-300 (500 × 7.9 mm) under water as the mobile phase with a flow rate 0.5 ml/min. Effluents were fractionated every 2 ml from fraction (Fr.) 1 to 21, 11 min after sample injection. Each fraction, with or without hydrolysis, was analyzed for amino sugar (A) using the Morgan-Elson colorimetric method and the PNAG (B) signal using dot blot with the rabbit anti-PNAG as described in the Methods section. WT; wild type strain FK300.

Figure 9

The typical ESI positive ion mass spectrum of PIA fractionated by GPC, showing the major mass differences to be deacetylation of poly-N-acetyl-D-glucosamine polymer. Typical ESI-positive ion mass spectrum of PNAG from FK300Δ5bpΔicaB fractionated using GPC. Samples of cell surface FK300Δrob (A) and culture supernatant FK300Δ5bpΔicaB (B) from the collected fraction by GPC were analyzed using ESI-MS. MS analyses of authentic PNAG (A) showed fragmentation of polysaccharide homopolymers resulting in a series of ESI-MS of m/z = 203 Da, which corresponded to GlcNAc. Besides these spectra, several peaks defecting of m/z = 42 Da, which corresponded to deacetylation of [GlcNAc] polymer were observed in FK300Δrob (A). By contrast, the culture supernatant of autoaggregation inducing strain FK300Δ5bpΔicaB (B) contained non deacetylated [GlcNAc] polymer. N-acetyl-D-glucosamine is expressed as GlcNAC, Glucosamine as GlcN. WT; wild type strain FK300.