Inhibition of multiple staphylococcal growth states by a small molecule that disrupts membrane fluidity and voltage

ABSTRACT

New molecular approaches to disrupting bacterial infections are needed. The bacterial cell membrane is an essential structure with diverse potential lipid and protein targets for antimicrobials. While rapid lysis of the bacterial cell membrane kills bacteria, lytic compounds are generally toxic to whole animals. In contrast, compounds that subtly damage the bacterial cell membrane could disable a microbe, facilitating pathogen clearance by the immune system with limited compound toxicity. A previously described small molecule, D66, terminates Salmonella enterica serotype Typhimurium (S. Typhimurium) infection of macrophages and reduces tissue colonization in mice. The compound dissipates bacterial inner membrane voltage without rapid cell lysis under broth conditions that permeabilize the outer membrane or disable efflux pumps. In standard media, the cell envelope protects Gram-negative bacteria from D66. We evaluated the activity of D66 in Gram-positive bacteria because their distinct envelope structure, specifically the absence of an outer membrane, could facilitate mechanism of action studies. We observed that D66 inhibited Gram-positive bacterial cell growth, rapidly increased Staphylococcus aureus membrane fluidity, and disrupted membrane voltage while barrier function remained intact. The compound also prevented planktonic staphylococcus from forming biofilms and a disturbed three-dimensional structure in 1-day-old biofilms. D66 furthermore reduced the survival of staphylococcal persister cells and of intracellular S. aureus. These data indicate that staphylococcal cells in multiple growth states germane to infection are susceptible to changes in lipid packing and membrane conductivity. Thus, agents that subtly damage bacterial cell membranes could have utility in preventing or treating disease.

IMPORTANCE

An underutilized potential antibacterial target is the cell membrane, which supports or associates with approximately half of bacterial proteins and has a phospholipid makeup distinct from mammalian cell membranes. Previously, an experimental small molecule, D66, was shown to subtly damage Gram-negative bacterial cell membranes and to disrupt infection of mammalian cells. Here, we show that D66 increases the fluidity of Gram-positive bacterial cell membranes, dissipates membrane voltage, and inhibits the human pathogen Staphylococcus aureus in several infection-relevant growth states. Thus, compounds that cause membrane damage without lysing cells could be useful for mitigating infections caused by S. aureus.

INTRODUCTION

New clinical antibiotics are often chemical variants on natural products that microbes have deployed as antimicrobial weapons for millions of years. Bacteria quickly evolve resistance to chemical analogs of existing antibiotics by modifying endogenous pathways or by acquiring new resistance genes horizontally (1, 2). Moreover, during infection, bacteria are frequently differentially susceptible to antimicrobials because they adopt different growth states, including planktonic, biofilm, and persister states (3, 4). Therefore, compounds that interfere with infection by disrupting new targets across multiple growth states are needed.

Sources of new potential antimicrobials include screening platforms based on cell culture infection models, which recapitulate specific aspects of host soluble innate immunity better than traditional screens in standard microbiological media (5, 6). Multiple research groups have therefore developed screening platforms for compounds that prevent the survival of bacterial pathogens in the phagolysosomes of macrophages, which are professional phagocytes of the monocyte lineage that reside within tissues. In-macrophage screens for anti-infectives have utilized bacterial pathogens such as Salmonella enterica and Mycobacterium tuberculosis (7–10). These works have identified compounds that lack antibacterial activity in standard microbiological media but inhibit bacterial growth under broth conditions that mimic the macrophage phagosome environment. For instance, such compounds prevent S. Typhimurium replication in broth in growth media or altered genetic backgrounds that increase outer membrane permeability and/or inactivate efflux pumps (7, 9, 11–14). Thus, the relative impermeability of the Gram-negative cell envelope in standard media complicates the study of the mechanism of action for these compounds.

The present work focuses on the mechanism of action and the effect on disease-relevant growth states of the small molecule D66 (378.3 g/mol) in Gram-positive bacteria because they lack an outer membrane barrier. D66 is a screening hit from a commercially available drug-like discovery library (7, 15). The compound is minimally toxic to mammalian cells and mice, prevents the replication of S. Typhimurium in macrophages, and curbs infection in mice. Under broth conditions that permeabilize the outer membrane or reduce efflux pump activity, D66 inhibits the growth of Gram-negative bacteria and disrupts voltage across the inner membrane (11). In Gram-positive bacteria, we found that exposure to the compound inhibited growth, rapidly increased membrane fluidity, and disrupted membrane voltage without permeabilizing cells. D66 exposure inhibited biofilm formation and dispersed 1-day old biofilms. The compound also reduced the survival of staphylococcal persister cells in broth and in mammalian cells. These data suggest that compounds that disturb but do not readily lyse membranes have the potential to interfere with multiple bacterial growth states relevant to infection.

RESULTS

Growth of Staphylococcus species and Bacillus subtilis is inhibited by D66

To establish whether Gram-positive bacteria are sensitive to D66, we monitored bacterial growth across a dose range of D66 in four staphylococcal strains and in a Bacillus subtilis laboratory strain. The origins and defining characteristics of these strains are described in Materials and Methods. Bacterial growth (absorbance) over 18 hours was monitored after dilution of overnight cultures to an optical density at 600 nm (OD₆₀₀) of 0.01 in lysogeny broth (LB) containing D66. All strains tested were unable to grow in the presence of high concentrations of the compound, with values of half minimum inhibitory concentrations (MIC₅₀) ranging from 43 to 93 µM (Fig. 1A and B). In contrast, S. Typhimurium with an intact outer membrane grew normally even at 150 µM of D66, a concentration approaching the limit of solubility (11). Thus, bacteria lacking an outer membrane barrier have increased susceptibility to D66 upon outgrowth from overnight cultures.

Fig 1.

D66 inhibits the growth of Gram-positive bacteria in broth and is bactericidal to mid-log phase S. aureus. (A) Dose response curves monitoring staphylococcal and B. subtilis percent growth inhibition in the presence of D66. Data are normalized to growth in 2% DMSO after 18 hours. Mean and SEM of at least three biological replicates performed with technical triplicates. (B) Table of D66 concentrations that inhibit growth. S. Typhimurium is included for comparison. (C) Fold change in OD₆₀₀ and CFU/mL. Mid-log phase cultures of S. aureus FDA209 were treated at time 0 with DMSO or D66. Concentrations of D66 above and below the MIC₅₀ and cMIC₉₅, ranging from 40 to 160 µM were tested. Cultures were monitored for absorbance and plated for enumeration of CFU over 4 hours. Data are presented as fold change. Mean and SEM of three biological replicates performed with technical triplicates.

We evaluated whether D66 could affect bacterial survival in mid-log phase cells. S. aureus FDA209 was exposed to three concentrations of compound, and growth and survival (colony forming units) were monitored over 4 hours (Fig. 1C). At 40 µM of D66, strain FDA209 grew slightly less than DMSO control-treated cells. At 80 µM of D66, no growth was observed, and absorbance decreased over time, suggesting bacterial lysis. These observations were validated by the 10–100-fold decline in CFU from the initial cell count. We conclude that D66 is bactericidal to S. aureus in mid-log phase and that S. aureus could be used in experiments aimed at understanding D66 activity across bacterial growth states germane to infection.

S. aureus cell membranes rapidly increase fluidity in the presence of D66

Prior work in Gram-negative bacteria suggested the bacterial cell membrane is a target of D66 (11). Therefore, we established whether D66 could affect critical aspects of staphylococcal cell membrane function, including fluidity. Changes in membrane fluidity are detectable with the fluorescent probe Laurdan, which inserts into the phospholipid bilayer. The emission spectrum of Laurdan changes according to the polarity of the environment, as determined by calculating the generalized polarization (GP) (16–18). Treatment with the membrane fluidizer benzyl alcohol (BnOH) rapidly decreased GP, as expected (12, 19) (Fig. 2A). Exposure to 40 µM D66 similarly decreased GP, indicating membranes become more fluid in the presence of this compound. At 160 µM of D66, membrane fluidity quickly reversed, and rigidity increased, potentially reflecting D66 self-interactions. D66 thus rapidly increases S. aureus cell membrane fluidity.

Fig 2.

S. aureus cell membranes become more fluid and depolarize upon exposure to D66 but are not significantly permeabilized. Mid-log phase S. aureus FDA209 cells were used for all experiments. (A) Membrane fluidity was evaluated with a Laurdan GP assay. Cells were treated just after time 0 (arrow) with DMSO, the membrane fluidizer benzyl alcohol (BnOH) (50 mM), or D66. Mean and SEM of three biological replicates performed with technical triplicates normalized to DMSO at time 0. (B) Membrane voltage was monitored with the fluorescent dye DiSC3(5) and reported as relative fluorescent units (RFU). Cells were treated just after time 0 (arrow) with DMSO, the ionophore gramicidin (2 µg/mL; 1.1 mM), or D66. Mean and SEM of three biological replicates performed with technical triplicates normalized to DMSO at time 0. (C) Membrane potential in millivolts was calculated at 10 and 30 minutes after treatment based on calibration curves (Fig. S1B). Asterisks indicate P ≤ 0.01 compared with the corresponding DMSO sample, determined by one-way analysis of variance (ANOVA) with a Dunnett’s post-test; ns, not significant. (D) Membrane polarization and permeability were determined by flow cytometry using the fluorescent dyes TO-PRO-3 and DiOC₂(3). Cells were treated with DMSO, CCCP (30 µM), nisin (75 µg/mL), D66, or JD1 for 5 minutes. Data are representative of three biological replicates, with percentages shown in each quadrant.

D66 disrupts membrane voltage without creating large holes

Increases in membrane fluidity can impact ion and proton distribution across the cell membrane (20–22). To monitor membrane voltage upon exposure to D66, we used the fluorescent probe 3,3′-dipropylthiadicarbo-cyanine iodide [DiSC3(5)], which is quenched upon intercalation into charged lipid bilayers. Voltage disruption releases DiSC3(5) from the membrane and increases fluorescence (23, 24). Treatment of bacteria with gramicidin pore-forming peptides (1 mM) depolarized membranes and increased DiSC3(5) fluorescence, as expected (25), changing voltage from approximately −106 mV to −42 mV within 10 minutes (Fig. 2B and C; Fig. S1A). Treatment with D66 modestly increased the DiSC3(5) signal, for instance, changing voltage to −93 mV and −78 within 10 minutes at 40 and 160 µM, respectively. In summary, D66 disrupts membrane voltage in a dose-responsive manner.

Changes in membrane fluidity and voltage disruption can reflect decreased membrane barrier function (16, 18, 26). However, we did not observe significant changes in the proton gradient, reduction potential, nor the ability to exclude propidium iodide upon exposure to D66 until at least 25 minutes after treatment at the highest dose (Fig. S1B through E). Since these assays were conducted across populations of bacteria, we turned to flow cytometry to determine whether, in individual cells, membrane permeability and potential change simultaneously upon treatment with D66. We used the fluorescent dyes 3,3′-diethyloxacarbocyanine iodide [DiOC₂(3)] and TO-PRO-3 to determine whether D66 exposure depolarizes cells and/or compromises membrane integrity, respectively. DiOC₂(3) accumulates in membranes and fluoresces red if the membrane is polarized. DiOC₂(3) fluorescence shifts to green if voltage decays; the red/green fluorescence ratio provides a measurement of depolarization. If the cell membrane is permeable, TO-PRO-3 diffuses into the nucleoid and binds DNA, increasing its fluorescence. Treatment of S. aureus with CCCP decreased the red/green DiOC₂(3) fluorescence ratio without affecting the TO-PRO-3 signal, as expected (Fig. 2D; Fig. S1F through H). Nisin, a small antimicrobial peptide that forms pores in the membrane, increased TO-PRO-3 fluorescence and decreased the DiOC₂(3) fluorescence ratio, as anticipated. JD1, a small molecule that lyses bacterial membranes (27), demonstrated increased TO-PRO-3 signal. Treatment of the bacteria with D66 decreased the DiOC₂(3) fluorescence ratio with only a marginal increase in TO-PRO-3 fluorescence, up to 11% ± 3% within 45 minutes at the highest concentrations of D66. These observations indicate that in most cells, D66 rapidly disrupts membrane voltage without creating large holes. Overall, the data suggest that D66 drives increases in membrane fluidity that perturb voltage but does not lyse cells.

D66 inhibits biofilm formation in the S. aureus HG001 strain

When bacteria transition from a planktonic growth state to a biofilm, their cell membranes undergo a decrease in fluidity (28–31). Therefore, we investigated whether exposure of planktonic cells to D66 affects biofilm formation. Staphylococcus strains were exposed to a range of subgrowth inhibitory concentrations of D66 over 24 hours, and the accumulation of extracellular mass was monitored with crystal violet staining (Fig. 3A through D). Rifampin (rifampicin) and vancomycin were selected as positive control antibiotics because they are effective against planktonic S. aureus and diffuse through biofilms (32, 33). Crystal violet staining revealed that in strain FDA209, only rifampin inhibited biofilm formation. In HG001 and Staphylococcus epidermidis, the 1/2× MIC concentrations of rifampin and D66 (0.025 µg/mL; 30 nM and 40 µM, respectively) reduced biofilm formation. The USA300 strain was prone to increased biofilm formation, especially at lower concentrations of rifampin, vancomycin, and D66. To establish whether D66 differentially affects the accumulation of live or dead cells in the maturing biofilm, we imaged live biofilms with a spinning disc confocal microscope. Relative levels of live (Syto9, green) and dead (PI, red) cells were quantified. Vancomycin exposure increased the volume of live and dead cells in the HG001 strain, mirroring observations in other vancomycin-sensitive S. aureus strains (34). In contrast, both rifampin and D66 reduced the volume of live and dead cells in HG001 (Fig. 3E). In the S. epidermidis strain, rifampin reduced cell volume, and D66 increased the volume of live and dead cells (Fig. 3F). Thus, D66 exposure has mixed effects but inhibits biofilm formation in the HG001 strain nearly as well as rifampin by reducing the volume of live and dead cells.

Fig 3.

D66 has modest effects on biofilm formation in staphylococcus. Biofilm formation was monitored after 18 hours in Trypticase soy broth (TSB) with subinhibitory concentrations of DMSO, rifampin (Rif; 1× MIC₉₅ = 0.05 µg/mL), vancomycin (Vanc; 1× MIC₉₅ = 1 µg/mL), or D66 (1× MIC₉₅ = 80 µM). (A–D) Crystal violet staining for accumulated biofilm mass measured at A₅₉₅. Concentrations of Rif, Vanc, and D66 were 1/2×, 1/4×, 1/8×, and 1/16× MIC₉₅ from left to right. Mean and SEM of three biological replicates performed with technical triplicates and normalized to untreated cells. Asterisks indicate P ≤ 0.05 compared with DMSO, determined by a one-way ANOVA with a Dunnett’s post-test. (E, F). Biofilms incubated with 1/2× MIC of the indicated compound were stained with Syto 9 (live cells) and PI (damaged/dead cells) and imaged with confocal fluorescent microscopy. Volume of live, dead, and total cells was calculated (positive volume = positive voxel; 1 voxel = 0.0367 µm³). Left, cell volume; right, ratio of live/dead cells. Mean and SD of one of two biological replicates derived from a minimum of four fields of view. Asterisks indicate P ≤ 0.01 compared with DMSO, determined by a one-way ANOVA with a Tukey post-test on log₁₀ transformed data.

In 1-day-old biofilms, D66 reduces the number of live cells and disrupts biofilm architecture

We next examined whether D66 disrupts the maintenance of 1- and 5-day-old biofilms. Biofilms were treated with DMSO, rifampin, vancomycin, or D66 for 18 hours. The mass of 1-day-old biofilms, as measured by crystal violet, was reduced in the FDA209 S. aureus and the S. epidermidis strains in response to rifampin. Vancomycin treatment only affected the S. epidermidis strain (Fig. 4A through D). In contrast, D66 reduced 1-day-old biofilm mass in the FDA209 and USA300 strains. Rifampin and D66 at 4× MIC disrupted the biofilm structure and reduced the volume of live cells in S. aureus HG001 and in S. epidermidis (Fig. 4E and F). Volume reconstructions of the imaged biofilms confirmed that after 1 day, S. aureus HG001 formed thinner (~38 µm) biofilms than S. epidermidis (~80 µm) (4, 35). Exposure to D66 at 4× MIC reduced biofilm height and complexity similarly to rifampin (Fig. 4G and H). In 5-day old biofilms, vancomycin had little effect on any strain, but rifampin and D66 reduced mass in HG001 (Fig. S1). In summary, D66 treatment reduces the number of live cells in HG001 and in S. epidermidis 1-day-old biofilms. D66 also diminishes biofilm structure, suggesting that compounds that alter membrane fluidity can disrupt biofilms in the early stages of maturation.

Fig 4.

D66 decreases the number of live cells in 1-day-old staphylococcal biofilms. For the indicated strain, biofilms established in TSB for 24 hours were treated for 18 hours with DMSO, rifampin (1× MIC₉₅ = 0.05 µg/mL), vancomycin (1× MIC₉₅ = 1 µg/mL), or D66 (Fig. 1B). (A–D) Remaining biofilm matrix was quantified with crystal violet across three compound concentrations (4×, 2×, and 1× MICs from left to right). Mean and SEM from three biological replicates performed in triplicate. Asterisks indicate P ≤ 0.05 compared with DMSO, determined by one-way ANOVA with a Dunnett’s post-test. (E, F) Biofilm volume was measured after staining with Syto9 (live cells) and PI (dead cells). Mean and SD of one of two replicates derived from a minimum of four fields of view. Asterisks indicate P ≤ 0.05 compared with DMSO, determined by a one-way ANOVA with Dunnett’s post-test. (G, H) Representative Syto 9- and PI-stained biofilm images. Z-Stacked images were converted to volumes. Volume data were compiled and reconstructed in Nikon Elements Advanced Research software using a color-coded volume display where blue is the bottom, lowest Z-stack, and magenta is the highest point for each sample. Scale bar is in micrometers. Mean and SD of one of two biological replicates derived from a minimum of four fields of view. Asterisks indicate P ≤ 0.01 compared with DMSO, determined by a one-way ANOVA with a Tukey post-test on log₁₀ transformed data. Panels G and H are representative images from the same replicate as in panels E and F. The DMSO, rifampin, and vancomycin control samples were from the same experiments as the D66-treated samples but were published previously (27).

Broth persister cells and intracellular S. aureus are sensitive to D66

Staphylococcal biofilms include live cells that are growing and are antibiotic sensitive, in addition to non-growing persister cells that are recalcitrant to antibiotics (4, 36). Since D66 reduced the number of live cells during biofilm formation and early development, we tested whether broth persister cells are susceptible to this compound. Persister cells were generated by growing cultures overnight (37, 38). Samples were monitored by enumeration of CFU after 3 and 24 hours for recalcitrance to high doses of vancomycin (10 µg/mL) and ciprofloxacin (4 µg/mL). In the three S. aureus strains, ciprofloxacin prevented growth, an indication of persister cells (39) (Fig. 5A through C). In the S. epidermidis strain, vancomycin prevented growth (Fig. 5D). Exposure to D66 reduced survival relative to antibiotic-treated cells in S. aureus HG001, indicating that D66 can kill staphylococcal persister cells.

Fig 5.

Treatment with D66 reduced the number of viable S. aureus broth HG001 persister cells and modestly decreased the load of USA300 in mammalian cells. (AD) Overnight cultures of the indicated strains were treated at time 0 with DMSO, ciprofloxacin (4 µg/mL), vancomycin (10 µg/mL), or four different concentrations of D66. Cultures were plated for enumeration of CFU at the time points indicated. Mean and SEM of three biological replicates are shown. (E) RAW 264.7 macrophage-like cells (blue) and HeLa cells (red) were infected with S. aureus USA300, at multiplicities of infection (MOIs) of 0.5 and 2.5, respectively. Cells were treated 2 hours after infection with DMSO (circles on y axis), vancomycin (vanc, 50 µg/mL, upside-down triangles), daptomycin (dapto, 50 µg/mL, right-side up triangles), or dilutions of D66 from 40 µM (stars for RAW264.7 cells, squares for HeLa cells). After 8 hours of infection, cells were lysed and plated for enumeration of CFU. Mean and SEM of biological duplicates performed with technical triplicates are shown. Non-linear regression; four-parameter variable slope. The IC₅₀ values are indicated.

Some S. aureus strains can infect and survive within macrophages and epithelial cells, including the USA300 strain (40–43). We therefore tested whether D66 reduces viable intracellular USA300 bacteria in RAW264.7 macrophage-like cells and in HeLa cells. After 8 hours of infection, D66 treatment decreased the number of recoverable bacteria by approximately fivefold; IC₅₀ values could be calculated only for S. aureus in HeLa cells (Fig. 5E). The vancomycin and daptomycin controls performed similarly in HeLa cells and somewhat better in the macrophage-like cells, reducing bacterial load by approximately 10-fold. We conclude that D66 and control antibiotics have modest activity against intracellular S. aureus survival/growth.

DISCUSSION

The bacterial cell membrane is an untapped potential source of new antimicrobial targets. While compounds that, like detergents, rapidly permeabilize cell membranes can inhibit persister cells and biofilms (44, 45), whether compounds with more subtle effects on membranes could be prototypes for antibiotics or antibiotic adjuvants has not been well explored. In this study, we used Gram-positive bacteria to examine the mechanism of action and potential utility of a small molecule, D66. This compound was previously demonstrated to be minimally toxic to mammalian cells and mice. D66 inhibits S. Typhimurium survival in macrophages and curbs systemic infection in mice. However, D66 has been challenging to study in broth because it has difficulty passing through the Gram-negative bacterial outer membrane barrier and is expelled by efflux pumps (11). We found that in Gram-positive bacteria, the compound prevented regrowth from stationary phase, suggesting that it can reach a bacterial target(s). D66 was bactericidal to planktonic mid-log phase S. aureus in broth at concentrations of 1× MIC (80 µM), and the compound fluidized the S. aureus cell membrane with a rapidity that suggests fluidization as a primary or near-primary effect of the compound. D66 also modestly disrupted membrane voltage, potentially a secondary effect of membrane fluidization. However, S. aureus cells treated with D66 did not rapidly or strongly become permeabilized to large (600 MW) molecules, and the cells maintained a proton gradient, indicating that the compound has a significant but non-catastrophic effect on bacterial cell membrane integrity.

While compounds that lyse membranes disrupt biofilms (46), whether compounds that increase membrane fluidity affect biofilms is not clear. Under growth conditions that promote biofilm formation, D66 had varied and mild effects on the extracellular matrix and on the volume of live and dead cells between strains. In 1-day-old biofilms, D66 reduced the volume of live cells as well as rifampin and had a more modest effect on reducing the mass of extracellular matrix. The compound strongly disrupted the structure of 1-day-old biofilms, reducing the depth of the mat apparently by reducing the volume and height of the live cell mass. In 5-day-old biofilms, D66 exposure reduced biofilm mass only in one staphylococcal strain tested. The negative effects of D66 on biofilms may reflect the membrane-fluidizing activity of this compound. Bacterial membrane fluidity and permeability are typically a function of the overall acyl chain composition of the lipids (47). Upon transition from a planktonic to a biofilm growth mode, bacterial cell membranes become increasingly rigid. For example, in Pseudomonas aeruginosa, the inner membranes of cells within biofilms are less fluid than those of planktonic bacteria (28, 29). Across Gram-negative and Gram-positive bacteria, cells within biofilms have membranes with a higher percentage of saturated fatty acids than planktonic cells, consistent with increased membrane rigidity, decreased exchange with the environment, and resistance to toxic molecules (30, 31). As bacteria shift from a planktonic to a biofilm growth state, enzymes involved in lipid saturation and membrane rigidification become increasingly expressed (31). Therefore, compounds that enhance membrane fluidity have the potential to interfere with biofilm formation.

D66 exposure reduces staphylococcal persister cell survival in broth and has a modest inhibitory effect on S. aureus cells within mammalian cells. These observations suggest that, as in biofilms, bacteria in different growth states germane to infection need to maintain membrane fluidity within bounds to remain viable. Alternatively, or in addition, bacteria in distinct growth states could have varying amounts of a D66 target(s) or enable more D66 to reach a target. Along these lines, persister cells and biofilms are both recalcitrant to antibiotics, as are stationary phase bacteria, in comparison to mid-log phase bacteria, potentially due to differences in membrane permeability (48, 49). In macrophages, D66 could have activity against pathogens within vesicles in part due to lysosomal trapping of the compound with the acidification of the phagosome (13). Phagosomes that harbor S. aureus also become acidified, and, as with S. Typhimurium, low pH induces virulence gene expression (50–52). However, some S. aureus strains escape low-pH phagosomes and survive for a period of time within the cytosol (53). We do not yet know whether in macrophages, D66 specifically affects S. Typhimurium persister cells, growing cells, or both. Given the emerging lifestyle diversity of S. aureus strains and S. Typhimurium strains within host cells (54, 55), multiple approaches will likely be needed to thwart intracellular infection states across staphylococcal species and strains.

Limitations of this study include the small number of bacterial species in which the effect of D66 on membrane fluidity, permeability, and biofilms were examined. In addition, whether chemical analogs of D66 could separate the effects of the compound on membrane fluidity and growth in broth and in a biofilm state remains to be explored. It will further be of interest to establish whether D66 affects cell membrane fluidity uniformly or in patches throughout the membrane and the cell (56). A more precise understanding of the downstream cellular effects of D66 will be needed to clarify how this compound and possibly others that fluidize membranes act and could be deployed.

Overall, the work highlights that the search for compounds that interfere with biofilms need not be constrained to those that rapidly lyse and/or permeabilize membranes. As existing antibiotics and biocides continue to lose their utility (57), the future of antibacterials and biofilm prevention is likely to require cocktails of functionally complementary agents. It makes sense to include in these cocktails compounds with diverse mechanisms of actions, such as those that modify the fluidity of bacterial cell membranes.

MATERIALS AND METHODS

Bacterial strains

S. aureus FDA209 (ATCC 6538), S. aureus HG001 (AH2183), S. aureus USA300, and S. epidermidis 1457 (AH2490) are the bacterial strains used. S. aureus FDA209 was isolated from a skin lesion, is sensitive to antibiotics, and has historically been used in antimicrobial and quality control testing (58–60). S. aureus HG001 is a common lab strain derivative of NCTC835, a methicillin-sensitive strain isolated from a septic patient but with a restored rsbU1 gene to increase its biofilm production (61, 62). S. aureus USA300 grows rapidly and is a community-acquired methicillin-resistant strain that causes intracellular infections (62). S. epidermidis 1457 is antibiotic sensitive and forms robust biofilms (63, 64). B. subtilis 168 is a standard laboratory strain originally isolated from soil (65).

Media and reagents

Unless otherwise stated, bacteria were grown in LB at 37°C with aeration. D66 (AW00798) was purchased from MolPort. To obtain mid-log phase cells, bacteria were grown overnight in LB, diluted the next morning 1:100 in fresh LB, and then grown to mid-log phase (OD₆₀₀ 0.4–0.6).

Minimum inhibitory concentrations

Overnight cultures were diluted in LB to an OD₆₀₀ of 0.01 and distributed into polystyrene 96-well flat-bottom plates (Greiner, 655185). Compound was added to the desired final concentration, and the final DMSO concentration never exceeded 2%. Plates were grown at 37°C with shaking for 18 hours, and OD₆₀₀ was monitored (BioTek Synergy H1 or BioTek Eon). MICs were defined as the concentration at which 95% of growth was inhibited (OD₆₀₀).

Growth and kill curves

Mid-log phase cultures were sampled at time 0 and then compound or vehicle control (DMSO) was added. Cultures were incubated at 37°C with agitation. At the time intervals indicated, aliquots were monitored for OD₆₀₀ and plated for CFU enumeration. Data for OD₆₀₀ and CFU/mL were normalized to time 0.

Membrane fluidity

Cells were grown to mid-log phase in LBg (LB with 0.2% glucose). Laurdan (Invitrogen) was added to a final concentration of 10 µM and incubated at 37°C with rotation for 30 minutes. Cells were harvested by centrifugation, washed three times, and resuspended in prewarmed PBSg (PBS with 0.2% glucose). Cells (200 µL) were transferred to a black polystyrene 96-well plate (Greiner, 655076) and monitored (ex 360/em 450 and 500 nm) on a BioTek Synergy H1 plate reader. Baseline fluorescence was recorded for five minutes prior to addition of compound. Fluorescence was recorded for 25 additional minutes. Laurdan GP was calculated as follows: ${\displaystyle \mathrm{G}\mathrm{P}=({\mathit{I}}_{460}-{\mathit{I}}_{500})/({\mathit{I}}_{460}+{\mathit{I}}_{500})}$.

Membrane potential

Membrane potential was measured using the potentiometric fluorescent probe DiSC₃(5) (Invitrogen). Mid-log phase cells were diluted to an OD₆₀₀ of 0.4. DiSC₃(5) was added to a final concentration of 2 µM, and the culture was incubated at 37°C in a rotator for 15 minutes. Cells were captured on a 0.45-µm Metricel membrane filter (Pall), resuspended in fresh LB, and distributed (200 µL) into black polystyrene 96-well plates (Greiner, 655076). Plates were monitored (ex 650/em 680 nm) on a BioTek Synergy H1 plate reader. After baseline fluorescence was recorded, compound was added to the desired final concentration and measurements were recorded for an additional 30 minutes. The relationship between membrane potential and DiSC₃(5) fluorescence intensity (calibration) was established as described (23, 66). Briefly, S. aureus FDA209 was grown overnight in LB media with 50 mM Hepes pH 7.5 and three concentrations of KCl (46 mM, 18.1 mM, and 7.1 mM), with three corresponding concentrations of NaCl to maintain ionic strength (254 mM, 281.9 mM, and 292.9 mM, respectively). Cells were diluted 1:100 in these same media and grown until mid-log phase. Cells were then diluted to an OD₆₀₀ of 0.3 and incubated with 2 µM DiSC3(5) for 15 minutes. Valinomycin (K+ carrier) was added to a final concentration of 5 µM, and fluorescence intensity was monitored with a BioTek Synergy H1 (ex 650/em 680 nm) for 30 minutes.

Cytosolic pH

BCECF [2′,7′-bis-(2-carboxyethyl)-5-(and-6)-carboxyfluorescein], acetoxymethyl ester (BCECF-AM) (Molecular Probes) was added to mid-log phase cells in BCECF LB (LB, 0.1% glucose, 300 mM KCl, 50 mM HEPES, and pH 7) to a final concentration of 10 µM and incubated at 37°C in a rotator for 1 hour. Cells were diluted 1:10 and pipetted into a black polystyrene 96-well plate (Greiner, 655076). After 5 minutes of equilibration, DMSO, nigericin (Sigma-Aldrich, N7143), or D66 were added and fluorescence (ex 490/em 535 nm and ex 440/em 535 nm) was monitored every 2.5 minutes for 20 minutes using a BioTek Synergy H1 plate reader. BCECF fluorescence was calibrated at seven pH levels between 5.5 and 8 (every 0.5 pH): pH = pK_a – log(I₄₉₀/I₄₄₀); the pK_a of BCECF is 6.97 (12).

Reduction potential

Mid-log-phase cells (200 mL per well) were transferred to a black polystyrene 96-well plate (Greiner; 655076) containing compound. Compound was added, and the plate was incubated with shaking. Resazurin (alamarBlue; Invitrogen) (67) was added to a final concentration of 100 mg/mL 5 minutes at the indicated time point. The plate was incubated with shaking in the dark at room temperature for 5 minutes. Fluorescence readings were taken (ex 570/em 650 nm) using a BioTek Synergy H1 plate reader.

Cytosolic ATP

Intracellular ATP levels were measured using BacTiter-Glo Microbial Cell Viability Assay (Promega) according to the manufacturer’s instructions. Overnight cultures were subcultured in fresh LB and grown to an OD₆₀₀ of 0.35–0.45. Cells (100 µL) were added to 2 µL of compound in a 96-well plate and incubated for 10 minutes at 37°C with agitation. Reagent (100 µL) was added and incubated in the dark with agitation for 5 minutes. Luminescence was read on a BioTek Synergy H1 plate reader.

Membrane barrier assays

Compound, DMSO, or SDS as stated in the figure was added to mid-log phase cells to the desired concentration, and cultures were sampled at 0, 10, 15, 20, 30, and 45 minutes. Five minutes before harvesting, PI (10 µg/mL) (Life Technologies) was added. Cells were pelleted, washed twice, resuspended in PBS, and monitored (ex 535/em 617 nm) using a BioTek Synergy H1 plate reader.

Single-cell membrane potential and permeability

Overnight cultures of S. aureus strain FDA 209 were diluted 1:100 in fresh LB and incubated until grown to mid-log phase (OD₆₀₀ 0.3–0.6). Cultures were treated with varying concentrations of D66, JD1, Nisin, CCCP, or DMSO and incubated at 37°C rotating for either 5 minutes, 20 minutes, or 45 minutes. Thirty micromolars of DiOC2(3) and 200 nM TO-PRO-3 were then added, and the cultures were incubated in the dark at 37°C rotating for an additional 15 minutes. Samples were run using the Beckman Coulter CyAn ADP Analyzer with Summit version 4.4. Bacteria were identified based on SSC versus FSC. DiOC2(3) (ex 488 nm/em PMT FL1; green, and FL3; red) and TO-PRO-3 (ex 635 nm/em PMT FL8) were quantified across at least 40,000 events per sample. Data were analyzed using FlowJo version 10 9.0.

Biofilm growth and treatment

As previously described (27), cultures were grown to mid-log phase in TSB, diluted to 4 × 10⁶ CFU/mL in TSB, and 200 µL was added to each well of a flat-bottomed, polystyrene, 96-well (Greiner 655185) plate (edge wells on the plate were filled with PBS to minimize evaporation of experimental wells). After 24 hours of incubation at 37°C without agitation, wells were carefully washed twice with PBS followed by the addition of 200 µL of TSB containing 4 µL DMSO or antibiotic or compound to achieve the desired final concentration. For 5-day biofilm assays, wells were washed and received fresh media daily and DMSO, antibiotic, or compound was added on day 5. Plates were incubated at 37°C for 18 hours without agitation. Prior to staining, the biofilm was washed twice with PBS to remove non-adherent cells.

Biofilm inhibition assays

Cultures were grown to mid-log phase in TSB, diluted to 4 × 10⁶ CFU/mL in TSB, and 200 µL of this dilution was added to each well of a flat-bottomed, polystyrene, 96-well (Greiner 655185) plate, and 4 µL of DMSO, antibiotic, or compound was added to achieve the desired final concentration. Edge wells on the plate were filled with PBS to minimize evaporation of experimental wells. After 24 hours of incubation at 37°C with no agitation, the media and planktonic cells were carefully removed, and wells were washed two times with PBS. A multichannel pipette was used for removal of media and subsequent PBS washes.

Biofilm mass assessment with crystal violet

Plates containing washed biofilms were allowed to air dry; then, 200 µL of 0.01% crystal violet was added to each well and incubated for 20 minutes. The biofilms were washed once with Deionized water and air dried. The stained biofilm was resuspended in 200 µL 70% ethanol, and the A₅₉₅ was measured.

Biofilm volume microscopy and analysis

Biofilms were grown as described above with the following differences (27). Cellvis 35-mm # 1.5H Glass Bottom Dish with 20-mm bottom well dishes were used for biofilm growth, and PBS washes to remove non-adherent cells (cells not attached to the biofilm) were performed with serological pipettes to minimize disruption. Biofilms were stained for 15 minutes prior to imaging with PI (10 µg/mL) and Syto9 (3 µM) resuspended in PBS. The dye solution was gently removed with a pipette, and samples were washed twice with PBS and imaged live (fixation with 4% PFA resulted in a PI staining artifact). Images were acquired using the Yokogawa CellVoyager CV1000 Confocal Scanner System with a 40×/0.6NA WD 2.7–4.0 (mm) objective, a Microlens-enhanced Nipkow Disk Scanner with a pinhole size of 50 μm and Hamamatsu Photonics|ImagEM X2 EM-CCD Camera C9100-14 High Resolution 16-bit format. Prior to image acquisition, each biofilm Z range was determined by manually establishing the top and bottom of the biofilm, and the z-step size was set to 1 µm to simplify volume analysis (conversion between volume and voxels). Images were acquired in two color channels: ex/em for Syto 9 and PI at 488 nm_525/50 and 561 nm_617/73, respectively. A minimum of eight randomly selected fields of view were acquired, and at least four images were used for analysis. Images were imported to MATLAB R2020b as multidimensional tiff stacks and processed as volume data using a custom script designed to store all results in data structures for user review. To create 3D binary masks for extracting the Syto 9 and PI foreground signals from the background, Otsu’s threshold was chosen over manual thresholding due to widely differing levels of background signal observed across strains and/or caused by biofilm thickness (68). The total volume for each of the two channels was quantified as the summation of voxels identified above the threshold per channel. The total number of voxels was converted to the total volume (1 voxel = 0.0367 µm³) using post-analysis of metadata. Select volumes were reconstructed in Nikon Elements using thresholds determined by Otsu’s method to provide a color-coded quantitative volume display based on depth in micrometers. The DMSO, rifampin, and vancomycin control samples were from the same experiments as the D66-treated samples but were published previously (27).

Persister assays

Cultures were grown overnight in TSB at 37°C with aeration for 18 hours. Cultures were divided into samples, and antibiotic or compound was added to the indicated final concentration. Samples were incubated at 37°C with aeration and plated at 0, 3, and 24 hours for enumeration of CFU.

Intracellular infection assays

RAW 264.7 (TIB-71) macrophages (5 × 10⁴ macrophages in 100 µL of complete DMEM) were seeded in 96-well tissue culture plates (Greiner, 655180) and were incubated at 37°C with 5% CO₂. For experiments performed with HeLa cells (ATCC CCL-2), 1 × 10⁴ cells were seeded per well. S. aureus USA300 was grown overnight in TSB and sub-cultured in TSB to an OD₆₀₀ of 0.4 for regrowth to an OD₆₀₀ of 0.6. Cultures were diluted into to a final concentration of 5 × 10⁵ CFU/mL in complete DMEM. Twenty-four hours after seeding of the mammalian cells, 50 µL of bacterial cultures was added to each cell culture well, at approximate MOIs of 0.5 and 2.5 for RAW 264.7 cells and HeLa cells, respectively. Plates were centrifuged at 500 × g for 2 minutes to synchronize infection. Thirty or 45 minutes (for RAW 264.7 or HeLa cells, respectively) after bacterial addition, wells were washed one time with PBS. DMEM containing 100 µg/mL gentamicin (Sigma-Aldrich) was added and cells were incubated for 90 minutes. Wells were then washed twice with PBS. Complete DMEM was added to the wells followed by 1 µL of DMSO or D66 to the stated final concentration. After 8 hours of infection, cells were lysed and plated for CFU.

Statistical analyses

GraphPad Prism version 10.11 was used for statistical comparisons.

SUPPLEMENTAL MATERIAL

The following material is available online at https://doi.org/10.1128/msphere.00772-23.

Supplemental Figures. msphere.00772-23-s0001.pdf.

Supporting data for Fig. 2 and 4.

ASM does not own the copyrights to Supplemental Material that may be linked to, or accessed through, an article. The authors have granted ASM a non-exclusive, world-wide license to publish the Supplemental Material files. Please contact the corresponding author directly for reuse.