Role of Capsule in Resistance to Disinfectants, Host Antimicrobials, and Desiccation in Acinetobacter baumannii

INTRODUCTION

Acinetobacter baumannii is a Gram-negative opportunistic pathogen commonly encountered in the nosocomial environment (1–3). A. baumannii bacteria can persist in the hospital environment due in part to their ability to resist desiccation and disinfectants (4, 5). This bacterium can cause a range of infections, including pneumonia, bacteremia, skin and soft tissue infections, and osteomyelitis (2, 6). Infections by A. baumannii are complicated by resistance to antimicrobial agents via intrinsic and acquired mechanisms, with many clinical isolates exhibiting multidrug resistance (7–9). The trend of increasing resistance to current antimicrobials combined with a paucity of new drugs in development has prompted the World Health Organization to place A. baumannii on their list of critical threats to humans (10).

Recently, a novel aspect of A. baumannii physiology was characterized in strain AB5075, in which cells can undergo a rapid switch between opaque (VIR-O) and translucent (AV-T) colony variants that are only detectable by oblique lighting (11, 12). Only the VIR-O variant is virulent in a mouse lung infection model (12). VIR-O cells possess a thicker capsule than AV-T cells and are also more resistant to desiccation, disinfectants, and host immune defenses. However, the actual role of capsule thickness in these phenotypes was unknown. Moreover, 116 genes are differentially expressed between VIR-O and AV-T variants, and any of these genes could influence resistance to desiccation, disinfectants, and host immune defenses independently of the capsule. To conclusively determine the role of the capsule, the gene encoding the protein-tyrosine kinase Wzc, required for capsule production in A. baumannii (13), was deleted in VIR-O and AV-T backgrounds by allelic replacement using procedures described previously (14). In VIR-O Δwzc cells, the extracellular capsule was undetectable by electron microscopy (12). This was further confirmed by capsule extraction and alcian blue staining in polyacrylamide gels according to a previously published method (13), in which no detectable capsule was observed in the VIR-O Δwzc or AV-T Δwzc mutants (Fig. 1A and B). Capsule production was restored by providing the plasmid pwzc, containing an intact copy of the wzc gene in pWH1266 (15) (Fig. 1A and B). In contrast to a previous report using the A. baumannii strain ATCC 17978, where mutations in wzc required second-site suppressors to grow (13), we had no difficulty in constructing the wzc deletion mutant in AB5075. Surprisingly, the VIR-O Δwzc and AV-T Δwzc colonies lacking the capsule still exhibited opacity differences, where the VIR-O Δwzc mutant was more opaque than the AV-T Δwzc mutant under oblique lighting (Fig. 1C). In addition, the loss of the capsule did not alter switching frequencies between the VIR-O and AV-T variants (see Fig. S1 in the supplemental material). This indicated that cell surface differences and/or cytoplasmic constituents other than the capsule also contribute to opacity differences between VIR-O and AV-T variants.

• Open in new tab
• Download powerpoint

FIG 1

Effects of capsule loss on colony opacity. (A and B) Representative gel image of capsule preparations from the indicated strains (A), and a graphic representation of these data (B). Bars represent the average of three separate preparations and gel analyses. SDS-PAGE gels were evaluated using the ImageJ program (https://imagej.nih.gov/ij/) (22). Bands were quantified with ImageJ by plotting peaks and calculating the area under the curve. These values were then used to calculate the percent capsule production compared with that of the wild-type AB5075 opaque variant. Error bars represent the standard deviation for each strain. *, P < 0.005; **, P < 0.0005. (C) Left, a mixture of VIR-O Δwzc (black arrow) and AV-T Δwzc (gray arrow) colonies, and right, a mixture of wild-type VIR-O (black arrow) and AV-T (gray arrow) colonies were photographed on a 0.5× LB plate viewed through a dissecting microscope with oblique lighting.

Since A. baumannii is known to colonize the hospital environment and can persist despite decontamination procedures (16), the role of the capsule in resistance to commonly used disinfectants was assessed. Treatments with benzethonium chloride (BZT), benzalkonium chloride (BAC), or chlorhexidine gluconate (CHG) showed that the Δwzc mutant exhibited a statistically significant increase in sensitivity to these disinfectants compared with that of the wild type (P < 0.05) (Fig. 2A). Complementation of the Δwzc mutant with pwzc restored the level of survival back to that of the wild type (Fig. 2A).

• Open in new tab
• Download powerpoint

FIG 2

Role of the capsule in disinfectant and desiccation survival. A. baumannii strain AB5075 VIR-O/vector (pWH1266), VIR-O Δwzc/vector, and VIR-O Δwzc/pwzc were subcultured in LB broth with 5 µg/ml tetracycline and grown at 37°C with shaking to an OD₆₀₀ of ∼0.15. Nine hundred ninety microliters of the culture was then added to tubes, each containing 10 µl of one of the appropriated disinfectants, benzethonium chloride (BZT; Sigma-Aldrich), benzalkonium chloride (BAC; Sigma-Aldrich), or chlorhexidine gluconate (CHG; Sigma-Aldrich). Tubes were incubated at room temperature for 30 min. Cultures were then serially diluted and plated for CFU enumerations. (A) BZT treatment at 0.01%, BAC treatment at 0.004%, and CHG treatment at 0.008%. Bars represent the average of five to seven replicates, with error bars denoting standard deviation. *, P < 0.05; **, P < 0.01. (B) Wild-type VIR-O and VIR-O Δwzc mutant cells were mixed and incubated at 37°C with the specified antimicrobials. Data are presented as the average ratio of the percent survival of VIR-O Δwzc cells to wild-type VIR-O cells or VIR-O Δwzc/vector to VIR-O Δwzc/pwzc cells. Error bars denote standard deviation. ***, P < 0.005. (C) AB5075 VIR-O/vector, VIR-O Δwzc/vector, and VIR-O Δwzc/pwzc were tested for desiccation survival. Bacteria were rehydrated and plated on day 12 of desiccation to determine viability. Bars represent the average of 7 replications, with error bars indicating standard deviation. **, P = 0.0082; ***, P = 0.0009.

Resistance to host-associated factors, such as antimicrobial peptides, hydrogen peroxide, and lysozyme, may contribute to A. baumannii pathogenesis in a mouse lung model of infection. Furthermore, A. baumannii capsule production is required for resistance to killing when grown in human ascites fluid or human serum in vitro (13, 17). We sought to examine the role of the capsule in protection against specific antimicrobial molecules produced by the immune system. Equal amounts of wild-type VIR-O and VIR-O Δwzc variants were grown to early log phase in 2 ml LB broth. A 5-µl aliquot of each culture was mixed together in 250 µl of 0.1× tryptone-yeast extract (1 g tryptone, 0.5 g yeast extract/liter). Cells were mixed by vortexing, and serial dilutions were plated onto 0.5× LB agar plates to assess the initial CFU for both strains. Various antimicrobials were then added at the following final concentrations: cathelicidin-related antimicrobial peptide (CRAMP; murine homolog of human LL-37), 10 µg/ml; H₂O₂, 0.01%; and lysozyme, 10 mg/ml. Cells were treated for 1 h, dilutions were plated onto 0.5× LB agar plates, and colonies were visualized under oblique lighting and counted to enumerate the surviving variants. Wild-type VIR-O cells were more susceptible (2-fold) than the Δwzc mutant to the antimicrobial peptide CRAMP and to hydrogen peroxide, although these differences were not statistically significant (P = 0.51 for CRAMP; P = 0.42 for hydrogen peroxide) (Fig. 2B). However, the A. baumannii VIR-O Δwzc mutant was 137-fold more susceptible to lysozyme than was the isogenic wild-type VIR-O parent (P < 0.005) (Fig. 2B). Complementation with pwzc increased the levels of lysozyme survival back to those of the wild type (Fig. 2B).

A. baumannii is also known to effectively resist desiccation and has been shown to survive up to several months on dry solid surfaces (4). The desiccation survival of wild-type VIR-O cells and the isogenic Δwzc mutant was determined by growing cells to an optical density at 600 nm (OD₆₀₀) of ∼0.15, and 25 µl of bacterial culture was desiccated on a 96-well flat-bottom polystyrene plate at room temperature. At daily intervals, wells were rehydrated in 100 µl sterile phosphate-buffered saline (PBS) for 30 min and then serially diluted and plated onto 0.5× LB plates to enumerate the surviving cells. After 12 days of desiccation, the viability of the Δwzc mutant was decreased by approximately 2.5-fold relative to the wild type (P = 0.0009) (Fig. 2C). This loss of viability was restored by the addition of the wild-type wzc gene (Fig. 2C). Taken together, these data highlight that the capsule contributes to disinfectant and desiccation resistance, both of which facilitate survival in the hospital environment.

A. baumannii strains that lack the capsule are known to be attenuated in mammalian infection models (17–19). However, to our knowledge, a defined capsule mutant has not been tested in a lung model of infection. Approximately 5 × 10⁷ CFU was administered per mouse by intranasal inoculation. At 24 h postinfection, mice were sacrificed, and the lungs, spleen, and liver were harvested, homogenized, and plated to determine the CFU onto 0.5× LB plates. The VIR-O Δwzc mutant exhibited an 8-log reduction in CFU at 24 h postinfection compared to that of the encapsulated wild-type VIR-O variant (Fig. 3A). Similar trends were apparent when assessing bacterial load in the spleen and liver (Fig. 3B and C). These results highlight that the capsule is essential for A. baumannii virulence in a lung model of infection.

• Open in new tab
• Download powerpoint

FIG 3

The capsule is required for virulence in a murine pneumonia model of infection. The wild-type VIR-O and VIR-O Δwzc strains were inoculated intranasally into mice (n = 10/group), and at 24 h postinfection, animals were euthanized and internal organs collected to determine the CFU per gram tissue in lungs (A), spleen (B), and liver (C). Bars represent geometric mean by a two-tailed Mann-Whitney U test. ***, P < 0.0001.

In summary, this study demonstrates the important role of the capsule in disinfectant, desiccation, and lysozyme resistance. This adds to previous work demonstrating that proper lipid A acylation is required for A. baumannii to survive desiccation and to resist the actions of cationic antimicrobial peptides in vitro (20). In addition, resistance to disinfectants by Gram-negative bacteria has been typically linked to the low permeability of lipopolysaccharides and the outer membrane, and our work reveals an additional role for the capsule in this process (21). The capsule did not contribute to resistance to cationic antimicrobial peptides or hydrogen peroxide. As previous work revealed that the AV-T variant with a thinner capsule was more susceptible than the VIR-O variant to these defenses (12), the increased susceptibility of the AV-T variant must be due to factors other than the capsule.

The Δwzc mutant survived poorly in the mouse lung, with an 8-log reduction in CFU at 24 h postinfection relative to the wild type (Fig. 3). This is in agreement with previous studies indicating that the capsule is an important virulence factor in a soft tissue infection model (17). To our knowledge, this is the first study to test a defined capsule mutant in a lung model of infection; however, it should be noted that transposon insertions in a UDP-glucose 4-epimerase–encoding gene predicted to be involved in capsule synthesis failed to be recovered from the mouse lung in a transposon insertion sequencing (TnSeq) screen using A. baumannii ATCC 17978 (19). Overall, in addition to a role in virulence, this study highlights the role of the capsule in desiccation and disinfectant survival, two important factors that allow A. baumannii to persist in the hospital environment. Targeted approaches that inhibit capsule synthesis may have a combined benefit of reducing hospital colonization and the ability to cause disease.