attenuated virulence and biofilm formation in staphylococcus

1

Attenuated Virulence and Biofilm 1

Formation in Staphylococcus aureus 2

following Sublethal Exposure to Triclosan 3 4

Joe Latimer#, Sarah Forbes# and Andrew J. McBain* 5

School of Pharmacy and Pharmaceutical Sciences, 6 The University of Manchester, Manchester, M13 9PT, U.K. 7

8

9

Running title: Triclosan-adapted Staphylococcus aureus 10

Key Words: Small colony variants, Galleria mellonella infection model, triclosan 11 susceptibility 12 13 #These authors contributed equally to this work 14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

*For correspondence: Andrew McBain, School of Pharmacy, The University of Manchester, 30 Oxford Road, Manchester M13 9PT, UK. Tel: 00 44 161 275 2360; Fax: 00 44 161 275 2396; 31 Email: [email protected] 32

33

Copyright © 2012, American Society for Microbiology. All Rights Reserved.Antimicrob. Agents Chemother. doi:10.1128/AAC.05904-11 AAC Accepts, published online ahead of print on 19 March 2012

on February 13, 2018 by guest

http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

2

Sub-effective exposure of Staphylococcus aureus to the biocide triclosan can reportedly 34

induce a small colony variant (SCV) phenotype in Staphylococcus aureus. S. aureus SCVs 35

are characterised by slow growth rates, reduced pigmentation and lowered antimicrobial 36

susceptibility. Whilst they may exhibit enhanced intracellular survival, there are conflicting 37

reports regarding their pathogenicity. The current study reports the characteristics of a SCV-38

like strain of S. aureus, created by repeated passage on sub-lethal triclosan concentrations. S. 39

aureus ATCC 6538 (P0) was serially exposed ten times to concentration gradients of triclosan 40

to generate strain P10. This strain was then further passaged ten times on triclosan-free 41

medium (designated x10). The minimum inhibitory and bactericidal concentrations of 42

triclosan for P0, P10 and x10 were determined and growth rates measured in biofilm and 43

planktonic culture. Haemolysin, DNAse and coagulase activities were measured and virulence 44

determined using a Galleria mellonella pathogenicity model. Strain P10 exhibited decreased 45

susceptibility to triclosan and characteristics of a SCV phenotype, including considerably 46

reduced growth rate and the formation of pinpoint colonies. However, this strain also had 47

delayed coagulase production, impaired haemolysis (p<0.01), was defective in biofilm 48

formation and DNAase activity, and displayed significantly attenuated virulence. Colony size, 49

haemolysis, coagulase activity and virulence were only partially restored in strain x10, 50

whereas planktonic growth rate was fully restored. However, x10 was at least as defective in 51

biofilm formation and DNAse production as P10. These data suggest that although repeated 52

exposure to triclosan may result in a SCV-like phenotype, this is not necessarily associated 53

with increased virulence, and adapted bacteria may exhibit other functional deficiencies. 54

55

56


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

3

INTRODUCTION 57

Staphylococcus aureus is an important human pathogen that is responsible for a range of 58

hospital and community-acquired infections (10, 15). Successful treatment of such infections 59

is complicated not only by the emergence of meticillin-resistant strains (MRSA) (10) but also, 60

it has been claimed, by the spontaneous generation of small-colony variants (SCVs) (55). 61

62

SCVs are slow growing S. aureus sub-populations that may be recovered from patients with 63

persisting or relapsing infections (44), particularly those undergoing treatment with 64

aminoglycoside antibiotics (13), which exhibit decreased susceptibility to antimicrobials. 65

SCVs display a distinct phenotype, characterised by the formation of pinpoint colonies on 66

agar, low growth rate and reduced pigmentation. Such characteristics may lead to their 67

misidentification in the hospital laboratory (28), potentially complicating diagnosis (43). 68

69

The SCV phenotype has been commonly attributed to defects in the electron transport chain 70

due to mutations in hemB, ctaA or menD which inhibit haemin, haemin A and menadione 71

biosynthesis, respectively (14, 54). Some SCVs result from thymidine auxotrophy, due to 72

mutations in genes encoding thymidine synthesis or transport (9). Such SCVs can survive on 73

exogenous thymine, for instance in the airway secretions of CF patients, which are rich in 74

necrotic cells (9, 24, 25). Thymidine is also required for menadione biosynthesis, suggesting 75

that menadione auxotrophy may be a common factor in SCV formation in menadione and 76

thymidine mutants. However, SCVs isolated from infections can result from mechanisms 77

other than menadione and thymidine auxotrophy (47), suggesting diversity in SCV 78

mechanisms that has not been systematically investigated and that SCVs which emerge from 79

various sources are not necessarily functionally equivalent. 80

81


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

4

Reduced susceptibility to antibiotics in SCVs is likely to be multifactoral, depending on 82

whether the particular SCV is attributable to aminoglycoside-selected auxotrophy or was 83

generated by other means, since electron transport chain defects, result in reduced 84

electrochemical gradients which may impair the uptake of aminoglycosides and other cationic 85

molecules (4, 37). Additionally, lowered growth rate in SCVs could non-specifically reduce 86

antimicrobial susceptibility by non-specific mechanisms associated with reduced expression 87

of pharmacological targets. Alternatively, enhanced survival within the host cells could shield 88

the bacteria from inhibitory concentrations of antibiotics and biocides. The clinical 89

significance of this, however, depends on whether SCVs can revert to full virulence following 90

cessation of treatment (37) and on whether the expression of SCV phenotypes influences the 91

ability to form recalcitrant biofilms on surfaces. 92

93

The generation of the SCV phenotype in S. aureus has reportedly been induced in vitro by 94

exposure to triclosan (7, 49), a trichlorinated diphenyl ether biocide. Effective concentrations 95

of triclosan are highly bactericidal towards susceptible species where the molecule causes 96

membrane damage through direct interaction and by perturbing lipid biosynthesis (36, 46, 97

53). Triclosan is used as an antibacterial in a variety of consumer and clinical applications (5, 98

12, 16) including triclosan wash solutions, which have been used to reduce carriage of 99

methicillin-resistant S. aureus (MRSA) in hospitals (1, 26, 45, 56). The enoyl-acyl carrier 100

protein reductase FabI has been identified as a primary pharmacological target (31, 36) for 101

triclosan. Inhibition renders the cell unable to synthesise fatty acids and growth is therefore 102

inhibited and, depending on concentration and time, bacteria can be permanently inactivated 103

(21). Due to the demonstrable therapeutic efficacy, concern has been expressed that the use of 104

triclosan and other antiseptics should be limited to applications where a clear health benefit 105


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

5

can be demonstrated, due the possibility of reduced susceptibility to the molecule and 106

potentially also to chemically-unrelated compounds (18). 107

108

Reduced susceptibility to triclosan has been shown in a number of in vitro studies but not, to 109

date, in the environment (32, 38). In one study, SCVs generated by the in vitro exposure of 110

bacteria to sub-lethal concentrations of triclosan impregnated in silicone disks exhibited 111

reduced susceptibility to the molecule and altered colony morphology (7), Since triclosan-112

exposed S. aureus has the potential to display a SCV-like phenotype and some clinical SCV 113

isolates appear to be able to cause disease (49), it has been inferred that triclosan use might 114

generate SCVs which could potentially be of clinical significance (7). However, these in vitro 115

studies do not necessarily reflect the clinical impact of triclosan-selected SCVs (48). For 116

example, in a previous report, SCVs generated using triclosan-impregnated disks 117

demonstrated slow growth rate, reduced coagulase and DNAse activity and decreased 118

haemolysis (7), which is suggestive of decreased virulence. 119

120

There is a clear need to determine the ability of triclosan-generated SCVs to initiate infection 121

(48). To our knowledge, there are currently no published reports regarding the impact of 122

triclosan exposure on the pathogenic capability of resulting SCV-like strains. Additionally, 123

whilst antibiotic-selected SCVs may revert to wild-type phenotype in vivo following cessation 124

of treatment, this has not been reported for triclosan-selected SCVs. The present study 125

therefore investigates the growth and virulence of a SCV-like S. aureus strain created by 126

serial exposure to sub-inhibitory concentrations of triclosan. The potential reversion of this 127

strain following subsequent repeated passage on triclosan-free medium was also studied. 128

129

130


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

6

MATERIALS AND METHODS 131

Bacterial strains and growth media. Staphylococcus aureus ATCC 6538 was 132

acquired from the American Type Culture Collection and cultured on Mueller Hinton agar 133

(MHA), Colombia blood agar (CBA) or mannitol salt agar (MSA, Oxoid, UK) aerobically at 134

37˚C. For supplementation experiments, haemin (1, 10 or 100 μM, final concentrations), 135

menadione (0.4, 4 or 40 μM, final concentrations) or thymidine (0.62, 6.2 or 62 μM final 136

concentrations) were added to MHA. Cryogenic stocks of the bacteria were archived at -80˚C. 137

Selection of isolates with reduced susceptibility to triclosan. Reproducible 138

concentration gradients of triclosan were created on Mueller Hinton agar by depositing stock 139

solutions of triclosan (100 μg/ml or 1 mg/ml) with a Wasp II spiral plater (Don Whitley, 140

Shipley, UK) (30, 35). Plates were dried for 1 hour at room temperature prior to radial 141

deposition of an overnight suspension of S. aureus and incubated for 4 days aerobically at 142

37˚C. Growth observed at the highest triclosan concentration was aseptically removed and 143

used to inoculate further gradient plates. This process was repeated for ten passages. A further 144

ten passages were performed on triclosan-free MHA. Wild-type bacteria (P0), those passaged 145

ten times on triclosan (P10) and those passaged a further ten times on triclosan-free MHA 146

(x10) were archived at -80 for subsequent MIC and MBC determination. 147

Determination of bacterial minimum inhibitory concentrations (MIC). Overnight 148

cultures of S. aureus were diluted 1:100 in sterile Mueller Hinton broth (MHB) and aliquots 149

(120 μl) were delivered to wells of polystyrene 96-well plates (Corning Ltd, Corning, USA). 150

Stock solutions (1.16 mg/ml) of triclosan were prepared in 25% ethanol, filter sterilised and 151

stored at –80˚C. Doubling dilutions of triclosan were added to the diluted overnight cultures 152

(final concentrations 232 μg/ml to 0.23 μg/ml). Sterile and triclosan-free controls were also 153

included. Plates were incubated for 24 h aerobically at 37 ˚C with shaking at 100 rpm. MICs 154

were determined as the lowest concentration of triclosan showing no turbidity in comparison 155


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

7

to a sterile negative control. Three technical and two biological replicates were conducted. 156

Previous validation studies with staphylococci and other genera showed that the triclosan 157

solvent has no effect on the outcome of MIC or MBC determinations (30). 158

Determination of bacterial minimum bactericidal concentrations (MBC). 159

Aliquots (10 μl) taken from MIC plates (above) were spot-plated onto MHA in triplicate and 160

incubated aerobically at 37˚C. MBCs were determined as the lowest concentration of triclosan 161

at which no growth was observed after 4 days of incubation. 162

Planktonic growth rate measurement. Overnight suspensions of S. aureus P0, P10 163

and x10 were diluted 1:100 in MHB (30 ml) and incubated at 37 oC with shaking at 150 rpm. 164

Samples were regularly removed and diluted as appropriate. Optical density was measured at 165

600 nm using a Helios spectrophotometer (Pye Unicam Ltd., Cambridge, UK). 166

Determination of biofilm growth rates. Overnight suspensions of S. aureus P0, P10 167

and x10 were diluted 1 in 40 in MHB and aliquots (200μl) delivered to wells of polystyrene 168

96 well plates. Sterile controls were also included. Plates were incubated statically aerobically 169

or in an anaerobic workstation (Don Whitley, Shipley, UK) at 37 ˚C in humid containers to 170

minimise evaporation. At 6 time points up to 50 h, planktonic and biofilm growth was 171

measured. Planktonic suspensions were transferred into semi-micro cuvettes and cell density 172

was measured spectrophotometrically at 600 nm. The remaining biofilm was stained for 60 173

seconds with 1% (w/v) crystal violet (200 μl) which was then removed and wells were 174

washed three times with distilled water before allowing plates to air-dry. Biofilm-bound CV 175

was eluted in 100 % ethanol (200 μl) and absorbance was measured spectrophotometrically at 176

600 nm. Biofilm units’ (arbitrary units) were calculated by dividing biofilm absorbances by 177

their corresponding planktonic OD. 178

Determination of biofilm architecture and viability. Overnight suspensions of S. 179

aureus P0 and P10 were diluted 1 in 40 in MHB (40 ml) in a Duran bottle containing a 180


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

8

partially-submerged sterile glass microscope slide. Bottles were incubated at 37 oC statically 181

for 24, 48 or 72 h before gentle rinsing in PBS (50 ml). Biofilm viability and structure was 182

visualised by fluorescence microscopy. Briefly, a working solution of BacLight LIVE/DEAD 183

stain (Invitrogen Ltd, Paisley, UK) was prepared by adding 1 μl each of SYTO 9 (component 184

A) and propidium iodide (component B) to 98 μl distilled water. This solution (20 μl) was 185

applied directly to the biofilm and covered with a glass cover slip. Slides were incubated at 186

room temperature in the dark for 15 minutes, according to the standard BacLight staining 187

protocol. Biofilms were visualised with an Axioskop 2 fluorescence microscope with a 10 x 188

objective lens (Carl Zeiss Ltd, Rugby, UK) and images captured using a digital microscope 189

eyepiece (Cosmos Biomedical, Derbyshire, UK) and exported as JPEG files. Bacterial cells 190

incubated in the presence of both stains fluoresce either green (viable) or red (dead). The 191

excitation and emission maxima for these dyes are 480 and 500nm for SYTO 9 and 490 and 192

635nm for propidium iodide. The percentage viable biomass was determined by calculating 193

the proportion of red fluorescence as a percentage of total fluorescence. Biofilm three-194

dimensional structure was visualised with an LSM Confocor 2 confocal microscope with a 10 195

x objective lens (Carl Zeiss Ltd, Rugby, UK). Images were processed, surface-rendered and 196

quantified using Imaris (Bitplane AG, Zurich, Switzerland). 197

Determination of haemolysin activity. This method was adapted from Hathaway and 198

Marshall (19). Overnight suspensions of S. aureus P0, P10 and x10 were diluted 1:50 in MHB 199

(50 ml) in a conical flask and incubated with shaking at 37 oC to an optical density at 600 nm 200

of 0.3. Whole defibrinated horse blood (Oxoid, UK) was added to a final concentration of 5% 201

(v/v) before incubating for a further 3 h. Negative and positive controls (sterile MHB and 202

distilled water, respectively) were included. Aliquots (1 ml) were centrifuged at 16,000 x g 203

(1-14 microfuge, Sigma, UK) for 4 min at room temperature and the absorbance of the 204

supernatant was measured spectrophotometrically at 540 nm. Colony counts were performed 205


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

9

by plating appropriate dilutions on MSA. To avoid variation caused by differences in cell 206

number, specific haemolysis was calculated as ∆A540/cfu. 207

Determination of DNAse activity. This method was adapted from the Health 208

Protection Agency National Standard Thermonuclease Test Method (20). Overnight 209

suspensions of S. aureus P0, P10 and x10 were diluted 1 in 50 in MHB (50 ml) in a conical 210

flask and incubated with shaking at 37 oC to an optical density at 600 nm of 0.3. Aliquots 211

(1ml) were incubated at 100 oC for 15 min, centrifuged at 13,000 rpm for 5 min and 212

supernatants retained. Wells were cut into plates containing DNAse test agar (Oxoid, UK) and 213

filled with supernatant (40 μl). Plates were incubated overnight at 37 oC and then overlaid 214

with 1M HCl. Polymerised DNA makes the agar opaque, and zones of clearing surrounding 215

the wells indicate levels of DNase activity. Zone diameters were recorded and colony counts 216

were performed by plating appropriate dilutions of unheated cultures on MSA to account for 217

variations caused by differences in cell number. 218

Coagulase assay. Suspensions of S. aureus P0, P10 and x10 in Mueller Hinton Broth 219

(1 ml, OD600 = 0.4) were added to rabbit plasma with EDTA (Bactident coagulase, Merck, 220

Darmstadt, Germany, 3 ml), in triplicate, and incubated at 37 oC in a water bath. Tubes were 221

monitored for signs of coagulation over 3 h and scored on a five-point scale according to 222

manufacturers’ instructions. 223

Galleria mellonella pathogenesis assay. The pathogenesis model was adapted from 224

Peleg et al. (42). Final larval-stage G. mellonella (Live Foods Direct, Sheffield, UK) were 225

stored in the dark at 4 oC for less than 7 days before randomly assigning 16 to each treatment 226

group and incubating at 37 oC for 30 min. Overnight suspensions of S. aureus strains P0, P10 227

and x10 were washed twice in PBS and diluted appropriately in PBS to achieve an optical 228

density at 600 nm of 0.1 (5-8 x 105 cfu/ml, as confirmed by colony counts on MSA) Aliquots 229

of each suspension (5 μl) were injected into the hemocoel of each larva via the last left proleg 230


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

10

using a Hamilton syringe (2.5-4 x103 cfu per individual). Larvae were incubated in plastic 231

petri dishes at 37 oC and the number of surviving individuals was recorded daily (dead larvae 232

were unresponsive to touch and appeared black). An untreated group and a group injected 233

with sterile PBS were used as negative controls. The experiments were terminated when at 234

least 2 individuals in a control group had died. Three independent replicates were performed 235

and significance was calculated using the Log-Rank test. Data were plotted as survival curves 236

and representative data are presented. Dead individuals were homogenised in sterile PBS, 237

diluted appropriately and aliquots were spread on MSA to calculate bacterial load per 238

individual at death. 239

240

RESULTS 241

Triclosan exposure selects for isolates with reduced triclosan susceptibility. The 242

minimum inhibitory concentration (MIC) of triclosan to Staphylococcus aureus ATCC 6538 243

(P0) was 0.23 μg/ml. After 5 passages (P5), this increased to 9.67 μg/ml and after 10 244

exposures (P10), to 29 μg/ml (a 126-fold increase compared to the parent strain; Fig. 1.). 245

Colonies of P10 were markedly smaller than those of their parent strain (Fig. 2.). Further 246

passaging on triclosan-free medium resulted in a partial reversion of susceptibility and colony 247

size; following 5 passages on MHA, the MIC was 7.3 μg/ml, and after 10 passages, 1.1 μg/ml. 248

Minimum bactericidal concentration (MBC) data showed a similar trend (Fig. 1.). 249

Triclosan-adapted S. aureus grows more slowly than its parent strain in 250

planktonic and biofilm modes. The growth kinetics of triclosan-adapted S. aureus (P10), its 251

parent strain (P0) and the strain further passaged on triclosan-free medium (x10) were 252

compared in shaken broth cultures (planktonic growth). As shown in Figure 3, P10 underwent 253

an extended lag phase and entered stationary phase later than P0 and x10. The planktonic 254

growth kinetics of P0 and x10 were not significantly different. The exponential phase growth 255


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

11

rate of P10 (0.19/h) was less than half that of P0 (0.43/h) and X10 (0.42/h), Biofilm growth 256

was measured over 50 h using the well-characterised microplate assay (41). Formation of 257

biofilm by P10 and x10, relative to planktonic growth, was significantly lower than that of P0 258

(Fig. 4.). Strain x10 formed less biofilm mass than P10 for the first 15 h of growth, after 259

which levels were not significantly different. Strains P10 and x10 were also significantly 260

biofilm defective when grown anaerobically (data not shown). Biofilms formed on glass 261

slides by P0, P10 and x10 differed markedly with respect to aerial coverage (Fig. 5.). 262

Additionally, epifluorescence and confocal microscopy showed that P0 formed relatively 263

thick three-dimensional microcolony structures whereas strains P10 and x10 formed thinner, 264

less structurally complex biofilms (Fig. 6 A.). Analysis of representative images indicated that 265

the density of biomass in P0 biofilms was markedly higher than in those of P10 or x10 over 266

72 h (Fig. 6 B.). 267

The in vitro haemolytic activity of triclosan-adapted S. aureus is lower than that 268

of the parent strain. The ability of planktonic P0, P10 and x10 populations to lyse 269

erythrocytes was investigated. Higher absorbance readings are indicative of lysis and release 270

of haemoglobin into the supernatant. Haemolysis by P10 was, on average, 2% of the P0 value, 271

a statistically significant difference (Fig. 7. p<0.01). Although x10 showed a partial reversion 272

of haemolytic activity, it was still significantly lower than the parent strain (5% of the P0 273

value). 274

The in vitro DNAse activity of triclosan-adapted S. aureus is lower than that of 275

the parent strain. The levels of thermostable DNase produced by planktonic P0, P10 and 276

x10 populations was investigated with an agar diffusion test. Zones of clearing after addition 277

of heated culture supernatant indicate deoxyribonuclease activity. Activity (mean zone 278

diameter 9.47 ± 0.53 cm) was only observed surrounding wells filled with culture supernatant 279

from the P0 strain. No DNAse activity was observed surrounding wells filled with culture 280


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

12

supernatants from the P10 or x10 strains (Table 2.). Cell density in the P0 cultures (4.6 x 108 281

cfu/ml) was not significantly different from those in the P10 or x10 populations (p=0.2). 282

The in vitro coagulase activity of triclosan-adapted S. aureus is lower than that of 283

the parent strain. The levels of coagulase produced by planktonic P0, P10 and x10 284

populations were investigated using a standard tube coagulase test. Coagulase activity was 285

observed in all strains. Although strain P0 displayed a positive result after 30 minutes, 286

significant coagulation was delayed in strains P10 and x10, both showing a positive result 287

after 2 h (Table 1). 288

Supplementation with haemin, menadione or thymidine does not restore a non-289

SCV phenotype. To ascertain whether the SCV phenotype was caused by mutations affecting 290

biosynthesis of haemin, menadione or thymidine, MHA containing these molecules was 291

inoculated with P10. The concentrations used were based on the previous reports that 292

maximal restoration of normal phenotypes occurred with 1.0 µM haemin, 0.375 µM 293

menadione or 6.2 µM of thymidine (9, 56). Further plates were inoculated with P0 and x10 to 294

control any growth-inhibitory effects of medium supplementation. No supplementation 295

affected the colony morphologies of any strain (Table 2), suggesting that the phenotype of 296

P10 is not attributable to defects in haemin, menadione or thymidine biosynthesis. 297

Triclosan-adapted S. aureus is less virulent in an invertebrate model. The Galleria 298

mellonella pathogenesis assay was used to assess the relative virulence of the test strains. 299

Groups of larvae were injected with P0, P10 or x10 and incubated at 37oC. Dead individuals 300

were recorded daily and data were plotted as survival curves (Figure 8A). Over 6 days of 301

incubation, the P10 treatment group suffered the fewest deaths whereas survival in the group 302

inoculated with the parent strain P0 fell rapidly, levelling out at 5 individuals in the example 303

shown. In comparison, x10 displayed moderate virulence, suggesting a partial reversion in 304

phenotype. A Log Rank test including all replicates showed that all three treatment groups 305


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

13

were significantly different (n=48). Within 24 h of death, the bacterial load of each individual 306

was measured. Although numbers varied markedly between individuals (107 – 1011), there 307

was no significant difference between treatment groups (Figure 8B). Colony morphologies of 308

bacteria recovered from dead larvae were similar to the inoculum in all groups, suggesting 309

that the SCV-like strain did not revert to wild-type, or vice versa, during infection. Results 310

have been summarized in Table 2. 311

312

DISCUSSION 313

S. aureus is known to generate phenotypically-distinct sub-populations termed small colony 314

variants (SCVs), which have occasionally been isolated from patients with chronic and 315

relapsing infections. Clinically-occurring SCVs commonly result from haemin, menadione or 316

thymidine auxotrophy, but may also result from mutations in other genes (47). SCV-like 317

variants can arise in response to sub-lethal exposure to aminoglycoside antibiotics (13) and 318

the biocide triclosan. Thus, concerns have been expressed that exposure of S. aureus to 319

triclosan might lead to the emergence of S. aureus SCVs with reduced susceptibility to 320

antimicrobials (7). However, care must be taken when extrapolating clinical relevance from in 321

vitro data. The fact that triclosan is normally applied topically at concentrations considerably 322

higher than those used to generate SCVs should be taken into account (48), as should the fact 323

that although elevated (0.0029% w/v), the MICs for triclosan in the adapted strains described 324

in the current study remained orders of magnitude lower than the triclosan concentrations in 325

washes used to control MRSA in hospitals (0.3 – 1.0 % w/v) (11, 57). Little is known about 326

the physiology and pathogenicity of triclosan-induced SCVs and doubts have been raised 327

concerning their ability to initiate infection in healthy individuals (48). 328

329


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

14

There have been conflicting reports concerning the virulence of SCVs, whether of clinical 330

origin or created in the laboratory. For example, a hemB mutant was more virulent than its 331

parent strain in a murine septic arthritis model (23), but demonstrated no significant 332

difference in virulence in a rabbit endocarditis model (6). Another report suggested that hemB 333

and menD mutants and SCVs of clinical origin were reduced in virulence (50). Clinically-334

isolated SCVs may however be physiologically distinct from hem and men mutants generated 335

in the laboratory and in this respect, recent proteomic analyses revealed significant 336

differences in protein expression between a hemB mutant, a gentamycin-induced SCV and a 337

clinical SCV isolate (29). Markedly increased expression of glycolytic and pyruvate 338

dehydrogenase enzymes was detected in the laboratory mutant and gentamicin-induced 339

strains, compared to the clinical isolate. However, there has previously been little information 340

available to indicate whether strains selected by triclosan exposure vary in virulence 341

compared to the progenitor strains. 342

343

The current investigation has demonstrated that repeated exposure of S. aureus to triclosan 344

can result in a phenotype with marked similarities to SCVs, but that are attenuated in biofilm 345

formation capacity, haemolysis, DNAase, coagulase activities and pathogenesis in an animal 346

model (Table 2). This suggests that whilst repeated exposure to triclosan may result in an 347

SCV-like phenotype, this is not necessarily associated with increases in variables associated 348

with pathogencity. 349

350

Serial exposure of bacteria to concentration gradients of antimicrobials in agar is a highly 351

selective procedure that has been validated as an effective method for selecting isolates with 352

decreased susceptibility (35). The G. mellonella infection model has been previously used to 353

study bacterial pathogenesis (5, 22, 39), including antimicrobial insusceptibility in S. aureus 354


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

15

(42). Data generated with this model in the current study showed a clear distinction between 355

the virulence of strains P0, P10 and x10. Benefits of G. mellonella over other non-mammalian 356

systems such as C. Elegans, include the fact that individual animals may be injected with a 357

defined inoculum, individual animals can be maintained at human body temperature, and that 358

the system has been shown to broadly reflect virulence patterns seen in mammalian systems 359

(22). 360

361

With respect to specific virulence factors, DNAse activity was markedly reduced following 362

triclosan exposure. It has been suggested that staphylococcal DNAse interferes with the 363

antimicrobial activity of neutrophil-produced extracellular traps (NETs) through the break 364

down of the chromatin backbone. DNAse-mediated NET degradation has been demonstrated 365

for S. aureus in a murine model, in which mice infected with wild-type S. aureus exhibited a 366

significantly higher mortaility rate than those infected with the nuclease-deficient mutant (8). 367

Host extracellular DNA in G. mellonella reportedly plays a similar role in the immune 368

response as it does in humans (3), suggesting that a reduction in DNase activity might 369

partially account for the reduced virulence of the S. aureus SCV observed in our G.mellonella 370

assay. 371

372

Importantly, S. aureus became markedly defective in its ability to form biofilms following 373

triclosan adaptation. Strains P10 and x10 formed biofilm less readily than the parent strain on 374

polystyrene over 50 h incubations and biofilms grown on glass were less dense, exhibiting 375

simpler architecture than the parent strain (Figs. 4-6). To our knowledge, there are currently 376

no data in the literature concerning the ability of triclosan-selected SCVs to form biofilms, 377

and data on other SCVs otherwise generated have not been conclusive; for example, S. aureus 378


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

16

SCVs created by exposure to amikacin formed greater biofilm mass on polystyrene (52) but 379

not on glass (51) and a Staphylococcus epidermidis SCV formed by a mutation in hemB was 380

almost entirely biofilm-defective at 24 h but, contrary to findings presented in the current 381

study, reverted to levels approximating the wild-type at 48 h (2). However, since the authors 382

did not measure biofilm relative to planktonic growth, this may be a reflection of the 383

inherently slow growth of the SCV phenotype, rather than a specific measure of biofilm-384

forming capability. In the current study, biofilm formation on polystyrene was measured 385

quantitatively over 50 h and these data were then converted to relative biofilm-forming units 386

(absorbance of the biofilm-bound crystal violet divided by the corresponding planktonic OD) 387

and therefore reflect biofilm formation, irrespective of planktonic bacterial productivity. Since 388

initial attachment does not appear to be impaired and biofilm deficiency is not due solely to 389

reduced growth rate, strains P10 and x10 appear to be specifically deficient in biofilm 390

maturation. This is potentially a consequence of repression of the intercellular 391

polysaccharides or proteins PIA or Aap, respectively. The staphylococcal accessory regulator 392

gene sarA, which controls several virulence determinants, including biofilm formation, 393

haemolysins and DNAse, may also be implicated. Furthermore, recent analysis has shown 394

that an increase in DNAse production resulting from a mutation in the stress response factor 395

sigB inhibited biofilm growth by breaking down extracellular DNA (eDNA) (27). Although 396

this appears to be in contrast to data presented in the current investigation, there is much to 397

learn about the role of eDNA in such a multi-factoral process as biofilm formation. 398

Importantly, since S. aureus infections related to medical devices are strongly associated with 399

biofilm formation (17), the selection for biofilm-deficient strains may be clinically 400

advantageous since biofilms are markedly less susceptible to antimicrobial therapy than their 401

planktonic counterparts (34, 40), a phenomenon that is due to biofilm-specific physiology as 402

well as diffusion-resistance (33). Therefore, the inability to form biofilm could potentially 403


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

17

outweigh changes in susceptibility related to the SCV phenotype. Further, despite a raised 404

MIC to a single biocide, an inability to form biofilm may leave such SCVs vulnerable to a 405

wide range of antimicrobial and immunological challenges. 406

Although the phenotype of the triclosan-adapted strain (P10) was practically identical to 407

previously-described SCVs in terms of colony size and planktonic growth, initial 408

supplementation data indicated that P10 was not an electron transport chain-defective SCV. It 409

is however possible that mechanisms other than auxotrophy to haemin, menadione or 410

thymidine were responsible for the induction of a SCV-like response in this strain. 411

S. aureus generated from 10 further passages on triclosan-free medium (x10) showed partial 412

reversion in planktonic growth rate and, to a lesser extent, in colony diameter and triclosan 413

susceptibility. Only minor reversions in haemolytic activity and biofilm formation were 414

observed and like P10, no DNAse activity was detected. Strain x10 was significantly less 415

pathogenic than the parent strain in the G. mellonella model and exhibited a marked increase 416

in triclosan susceptibility. Furthermore, P10 did not revert to normal colony size in vivo 417

during the course of the infection model (data not shown). 418

In conclusion, we have shown that whilst repeated exposure to triclosan might select for S. 419

aureus strains with decreased susceptibility to triclosan, this may be associated with 420

deficiencies in growth and virulence. 421

422


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

18

REFERENCES 423

1. Al-Mukhtar, A., V. K. Wong, H. Z. Malik, M. Abu-Hilal, M. Denton, M. Wilcox, 424 J. P. Lodge, G. J. Toogood, and K. R. Prasad. 2009. A simple prophylaxis regimen 425 for MRSA: its impact on the incidence of infection in patients undergoing liver 426 resection. Ann R Coll Surg Engl 91:35-8. 427

2. Al Laham, N., H. Rohde, G. Sander, A. Fischer, M. Hussain, C. Heilmann, D. 428 Mack, R. Proctor, G. Peters, K. Becker, and C. von Eiff. 2007. Augmented 429 expression of polysaccharide intercellular adhesin in a defined Staphylococcus 430 epidermidis mutant with the small-colony-variant phenotype. J Bacteriol 189:4494-431 501. 432

3. Altincicek, B., S. Stotzel, M. Wygrecka, K. T. Preissner, and A. Vilcinskas. 2008. 433 Host-derived extracellular nucleic acids enhance innate immune responses, induce 434 coagulation, and prolong survival upon infection in insects. J Immunol 181:2705-12. 435

4. Aperis, G., B. B. Fuchs, C. A. Anderson, J. E. Warner, S. B. Calderwood, and E. 436 Mylonakis. 2007. Galleria mellonella as a model host to study infection by the 437 Francisella tularensis live vaccine strain. Microbes Infect 9:729-34. 438

5. Barbolt, T. A. 2002. Chemistry and safety of triclosan, and its use as an antimicrobial 439 coating on Coated VICRYL* Plus Antibacterial Suture (coated polyglactin 910 suture 440 with triclosan). Surg Infect (Larchmt) 3 Suppl 1:S45-53. 441

6. Bates, D. M., C. von Eiff, P. J. McNamara, G. Peters, M. R. Yeaman, A. S. Bayer, 442 and R. A. Proctor. 2003. Staphylococcus aureus menD and hemB mutants are as 443 infective as the parent strains, but the menadione biosynthetic mutant persists within 444 the kidney. J Infect Dis 187:1654-61. 445

7. Bayston, R., W. Ashraf, and T. Smith. 2007. Triclosan resistance in methicillin-446 resistant Staphylococcus aureus expressed as small colony variants: a novel mode of 447 evasion of susceptibility to antiseptics. J Antimicrob Chemother 59:848-53. 448

8. Berends, E. T., A. R. Horswill, N. M. Haste, M. Monestier, V. Nizet, and M. von 449 Kockritz-Blickwede. Nuclease expression by Staphylococcus aureus facilitates 450 escape from neutrophil extracellular traps. J Innate Immun 2:576-86. 451

9. Besier, S., A. Ludwig, K. Ohlsen, V. Brade, and T. A. Wichelhaus. 2007. 452 Molecular analysis of the thymidine-auxotrophic small colony variant phenotype of 453 Staphylococcus aureus. Int J Med Microbiol 297:217-25. 454

10. Boucher, H., L. G. Miller, and R. R. Razonable. 2010. Serious infections caused by 455 methicillin-resistant Staphylococcus aureus. Clin Infect Dis 51 Suppl 2:S183-97. 456

11. Brady, L. M., M. Thomson, M. A. Palmer, and J. L. Harkness. 1990. Successful 457 control of endemic MRSA in a cardiothoracic surgical unit. Med J Aust 152:240-5. 458

12. Braid, J. J., and M. C. Wale. 2002. The antibacterial activity of triclosan-459 impregnated storage boxes against Staphylococcus aureus, Escherichia coli, 460 Pseudomonas aeruginosa, Bacillus cereus and Shewanella putrefaciens in conditions 461 simulating domestic use. J Antimicrob Chemother 49:87-94. 462

13. Brouillette, E., A. Martinez, B. J. Boyll, N. E. Allen, and F. Malouin. 2004. 463 Persistence of a Staphylococcus aureus small-colony variant under antibiotic pressure 464 in vivo. FEMS Immunol Med Microbiol 41:35-41. 465

14. Clements, M. O., S. P. Watson, R. K. Poole, and S. J. Foster. 1999. CtaA of 466 Staphylococcus aureus is required for starvation survival, recovery, and cytochrome 467 biosynthesis. J Bacteriol 181:501-7. 468

15. Deleo, F. R., M. Otto, B. N. Kreiswirth, and H. F. Chambers. 2010. Community-469 associated meticillin-resistant Staphylococcus aureus. Lancet 375:1557-68. 470


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

19

16. DeSalva, S. J., B. M. Kong, and Y. J. Lin. 1989. Triclosan: a safety profile. Am J 471 Dent 2 Spec No:185-96. 472

17. Francolini, I., and G. Donelli. 2010. Prevention and control of biofilm-based 473 medical-device-related infections. FEMS Immunol Med Microbiol 59:227-38. 474

18. Gilbert, P., and A. J. McBain. 2003. Potential impact of increased use of biocides in 475 consumer products on prevalence of antibiotic resistance. Clin Microbiol Rev 16:189-476 208. 477

19. Hathaway, S. C., and R. B. Marshall. 1980. Haemolysis as a means of 478 distinguishing between Leptospira interrogans serovars balcanica and hardjo. J Med 479 Microbiol 13:477-81. 480

20. Health Protection Agency. 2010. Thermonuclease Test. National Standard Method 481 BSOP TP 34, UK. 482

21. Heath, R. J., and C. O. Rock. 1995. Enoyl-acyl carrier protein reductase (fabI) plays 483 a determinant role in completing cycles of fatty acid elongation in Escherichia coli. J 484 Biol Chem 270:26538-42. 485

22. Jander, G., L. G. Rahme, and F. M. Ausubel. 2000. Positive correlation between 486 virulence of Pseudomonas aeruginosa mutants in mice and insects. J Bacteriol 487 182:3843-5. 488

23. Jonsson, I. M., C. von Eiff, R. A. Proctor, G. Peters, C. Ryden, and A. Tarkowski. 489 2003. Virulence of a hemB mutant displaying the phenotype of a Staphylococcus 490 aureus small colony variant in a murine model of septic arthritis. Microb Pathog 491 34:73-9. 492

24. Kahl, B., M. Herrmann, A. S. Everding, H. G. Koch, K. Becker, E. Harms, R. A. 493 Proctor, and G. Peters. 1998. Persistent infection with small colony variant strains of 494 Staphylococcus aureus in patients with cystic fibrosis. J Infect Dis 177:1023-9. 495

25. Kahl, B. C., G. Belling, R. Reichelt, M. Herrmann, R. A. Proctor, and G. Peters. 496 2003. Thymidine-dependent small-colony variants of Staphylococcus aureus exhibit 497 gross morphological and ultrastructural changes consistent with impaired cell 498 separation. J Clin Microbiol 41:410-3. 499

26. Kampf, G., and A. Kramer. 2004. Epidemiologic background of hand hygiene and 500 evaluation of the most important agents for scrubs and rubs. Clin Microbiol Rev 501 17:863-93, table of contents. 502

27. Kiedrowski, M. R., J. S. Kavanaugh, C. L. Malone, J. M. Mootz, J. M. Voyich, 503 M. S. Smeltzer, K. W. Bayles, and A. R. Horswill. Nuclease modulates biofilm 504 formation in community-associated methicillin-resistant Staphylococcus aureus. PLoS 505 One 6:e26714. 506

28. Kipp, F., B. C. Kahl, K. Becker, E. J. Baron, R. A. Proctor, G. Peters, and C. von 507 Eiff. 2005. Evaluation of two chromogenic agar media for recovery and identification 508 of Staphylococcus aureus small-colony variants. J Clin Microbiol 43:1956-9. 509

29. Kriegeskorte, A., S. Konig, G. Sander, A. Pirkl, E. Mahabir, R. A. Proctor, C. 510 von Eiff, G. Peters, and K. Becker. 2011. Small colony variants of Staphylococcus 511 aureus reveal distinct protein profiles. Proteomics 11:2476-90. 512

30. Ledder, R. G., P. Gilbert, C. Willis, and A. J. McBain. 2006. Effects of chronic 513 triclosan exposure upon the antimicrobial susceptibility of 40 ex-situ environmental 514 and human isolates. J Appl Microbiol 100:1132-40. 515

31. Levy, C. W., A. Roujeinikova, S. Sedelnikova, P. J. Baker, A. R. Stuitje, A. R. 516 Slabas, D. W. Rice, and J. B. Rafferty. 1999. Molecular basis of triclosan activity. 517 Nature 398:383-4. 518


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

20

32. Levy, S. B. 2001. Antibacterial household products: cause for concern. Emerg Infect 519 Dis 7:512-5. 520

33. Lewis, K. 2007. Persister cells, dormancy and infectious disease. Nat Rev Microbiol 521 5:48-56. 522

34. Lewis, K. 2001. Riddle of biofilm resistance. Antimicrob Agents Chemother 45:999-523 1007. 524

35. McBain, A. J., R. G. Ledder, P. Sreenivasan, and P. Gilbert. 2004. Selection for 525 high-level resistance by chronic triclosan exposure is not universal. J Antimicrob 526 Chemother 53:772-7. 527

36. McMurry, L. M., M. Oethinger, and S. B. Levy. 1998. Triclosan targets lipid 528 synthesis. Nature 394:531-2. 529

37. Melter, O., and B. Radojevic. Small colony variants of Staphylococcus aureus--530 review. Folia Microbiol (Praha) 55:548-58. 531

38. Meyer, B., and B. Cookson. 2010. Does microbial resistance or adaptation to 532 biocides create a hazard in infection prevention and control? J Hosp Infect 76:200-5. 533

39. Mylonakis, E., R. Moreno, J. B. El Khoury, A. Idnurm, J. Heitman, S. B. 534 Calderwood, F. M. Ausubel, and A. Diener. 2005. Galleria mellonella as a model 535 system to study Cryptococcus neoformans pathogenesis. Infect Immun 73:3842-50. 536

40. Nickel, J. C., I. Ruseska, J. B. Wright, and J. W. Costerton. 1985. Tobramycin 537 resistance of Pseudomonas aeruginosa cells growing as a biofilm on urinary catheter 538 material. Antimicrob Agents Chemother 27:619-24. 539

41. O'Toole, G. A., and R. Kolter. 1998. Initiation of biofilm formation in Pseudomonas 540 fluorescens WCS365 proceeds via multiple, convergent signalling pathways: a genetic 541 analysis. Mol Microbiol 28:449-61. 542

42. Peleg, A. Y., D. Monga, S. Pillai, E. Mylonakis, R. C. Moellering, Jr., and G. M. 543 Eliopoulos. 2009. Reduced susceptibility to vancomycin influences pathogenicity in 544 Staphylococcus aureus infection. J Infect Dis 199:532-6. 545

43. Proctor, R. A., and G. Peters. 1998. Small colony variants in staphylococcal 546 infections: diagnostic and therapeutic implications. Clin Infect Dis 27:419-22. 547

44. Proctor, R. A., C. von Eiff, B. C. Kahl, K. Becker, P. McNamara, M. Herrmann, 548 and G. Peters. 2006. Small colony variants: a pathogenic form of bacteria that 549 facilitates persistent and recurrent infections. Nat Rev Microbiol 4:295-305. 550

45. Rashid, A., L. K. Solomon, H. G. Lewis, and K. Khan. 2006. Outbreak of epidemic 551 methicillin-resistant Staphylococcus aureus in a regional burns unit: management and 552 implications. Burns 32:452-7. 553

46. Russell, A. D. 2004. Whither triclosan? J Antimicrob Chemother 53:693-5. 554 47. Schaaff, F., G. Bierbaum, N. Baumert, P. Bartmann, and H. G. Sahl. 2003. 555

Mutations are involved in emergence of aminoglycoside-induced small colony 556 variants of Staphylococcus aureus. Int J Med Microbiol 293:427-35. 557

48. Seaman, P. F., D. Ochs, and M. J. Day. 2007. Comment on: Triclosan resistance in 558 methicillin-resistant Staphylococcus aureus expressed as small colony variants: a 559 novel mode of evasion of susceptibility to antiseptics. J Antimicrob Chemother 560 60:175-6; author reply 176-7. 561

49. Seifert, H., H. Wisplinghoff, P. Schnabel, and C. von Eiff. 2003. Small colony 562 variants of Staphylococcus aureus and pacemaker-related infection. Emerg Infect Dis 563 9:1316-8. 564

50. Sifri, C. D., A. Baresch-Bernal, S. B. Calderwood, and C. von Eiff. 2006. 565 Virulence of Staphylococcus aureus small colony variants in the Caenorhabditis 566 elegans infection model. Infect Immun 74:1091-6. 567


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

21

51. Singh, R., P. Ray, A. Das, and M. Sharma. 2010. Enhanced production of 568 exopolysaccharide matrix and biofilm by a menadione-auxotrophic Staphylococcus 569 aureus small-colony variant. J Med Microbiol 59:521-7. 570

52. Singh, R., P. Ray, A. Das, and M. Sharma. 2009. Role of persisters and small-571 colony variants in antibiotic resistance of planktonic and biofilm-associated 572 Staphylococcus aureus: an in vitro study. J Med Microbiol 58:1067-73. 573

53. Villalain, J., C. R. Mateo, F. J. Aranda, S. Shapiro, and V. Micol. 2001. 574 Membranotropic effects of the antibacterial agent Triclosan. Arch Biochem Biophys 575 390:128-36. 576

54. von Eiff, C., C. Heilmann, R. A. Proctor, C. Woltz, G. Peters, and F. Gotz. 1997. 577 A site-directed Staphylococcus aureus hemB mutant is a small-colony variant which 578 persists intracellularly. J Bacteriol 179:4706-12. 579

55. von Eiff, C., R. A. Proctor, and G. Peters. 2000. Staphylococcus aureus small 580 colony variants: formation and clinical impact. Int J Clin Pract Suppl:44-9. 581

56. Webster, J., J. L. Faoagali, and D. Cartwright. 1994. Elimination of methicillin-582 resistant Staphylococcus aureus from a neonatal intensive care unit after hand washing 583 with triclosan. J Paediatr Child Health 30:59-64. 584

57. Zafar, A. B., R. C. Butler, D. J. Reese, L. A. Gaydos, and P. A. Mennonna. 1995. 585 Use of 0.3% triclosan (Bacti-Stat) to eradicate an outbreak of methicillin-resistant 586 Staphylococcus aureus in a neonatal nursery. Am J Infect Control 23:200-8. 587

588 589


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

22

590 FIGURE LEGENDS 591 592 Fig. 1. Minimum inhibitory concentrations (MICs, white bars) and minimum bactericidal 593 concentrations (MBCs, black bars) of triclosan to S. aureus ATCC 6538 before exposure 594 (P0), after 5 or 10 passages in the presence of triclosan (P5 and P10, respectively) and after 5 595 or 10 subsequent passages on triclosan-free medium (X5 and x10 respectively). Error bars 596 show standard deviation (n=6). 597 598 Fig. 2. Colonies formed on Colombia blood agar by S. aureus parent strain (P0, A), triclosan-599 adapted strain (P10, B) and one passaged a further ten times on triclosan-free medium (x10, 600 C) after aerobic incubation at 37 oC for 48 h. Colonies formed by the triclosan-adapted strain 601 were markedly smaller and more heterogeneous than those formed by the parent strain. 602 603 Fig. 3. Planktonic growth of S. aureus parent strain (P0, triangles), triclosan-exposed strain 604 (P10, squares) and one passaged a further ten times on triclosan-free medium (x10, circles). 605 P10 exhibited an extended lag phase, slower growth and a delayed stationary phase when 606 compared to P0. Error bars show standard deviation (n=3). 607 608 Fig. 4. Biofilm growth of S. aureus parent strain (P0, triangles), triclosan-exposed strain (P10, 609 squares) and one passaged a further ten times on triclosan-free medium (x10, circles) on 610 polystyrene. Data are shown as ‘biofilm units’. A biofilm unit is defined as the absorbance of 611 the biofilm-bound crystal violet divided by the corresponding planktonic OD, and corrects the 612 data to give a representation of biofilm formation irrespective of planktonic mass. Biofilm 613 formation by P10 was markedly lower than P0. This difference was significant at all time 614 points from 6 h (p<0.001, n=12). 615 616 Fig. 5. Biofilm growth on glass of S. aureus parent strain (P0), triclosan-exposed strain (P10) 617 and one passaged a further ten times on triclosan-free medium (x10) as viewed by 618 epifluorescence microscopy. Green fluorescence represents viable cells and red fluorescence 619 represents non-viable cells. Representative fields of view at 24 h are shown. 620 621 Fig. 6. Biofilm growth on glass of S. aureus parent strain (P0), triclosan-exposed strain (P10) 622 and one passaged a further ten times on triclosan-free medium (x10) as viewed by confocal 623 microscopy. Specialist software was used to surface-render the images, allowing comparison 624 of relative density and structure (A), and measurement of biomass density (B). Representative 625 images and data shown (P0, black bars; P10, white bars; x10 striped bars). 626 627 Fig. 7. Haemolytic activity of S. aureus parent strain (P0, black bar), triclosan-exposed strain 628 (P10, white bar) and one passaged a further ten times on triclosan-free medium (x10, striped 629 bar). Whole blood (final concentration 5% v/v) was added to pre-stationary phase cultures and 630 free haemoglobin from lysed erythroctes was measured spectrophotometrically at 450 nm. 631 Data are expressed as mean percentages of the mean P0 value. Error bars show standard 632 deviation (n=3). 633 634 Fig. 8. Example survival curve showing virulence of S. aureus parent strain (P0, green fine-635 dash curve), triclosan-exposed strain (P10, red dot-dash curve) and one passaged a further ten 636 times on triclosan-free medium (x10, blue coarse-dash curve) in a G. mellonella survival 637 assay (A, n=16) and bacterial load at death (B, each point corresponds to the bacterial load in 638


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

23

an individual animal). P0 caused significantly higher death rates than both other strains, 639 whereas the virulence of x10 was not fully restored. Untreated and PBS-treated control data re 640 also shown. Although the bacterial load recovered from dead larva varied, there was no 641 significant difference between treatment groups. Example images of treatment groups are also 642 shown.643


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

24

Table 1. Coagulase activity of S. aureus parent strain (P0), triclosan-adapted strain (P10) and 644 one passaged a further ten times on triclosan-free medium (x10). Tubes were monitored for 645 signs of coagulation over 3 h and scored on a five-point scale according to manufacturers’ 646 instructions. 647

648

Strain

P0 P10 X10

0.5 h ++++ - -

1 h ++++ - +

2 h ++++ +++ +++

3 h ++++ ++++ ++++

Negative symbol: no coagulase detected; +: small separate clots (negative); ++: small joined 649 clots (negative); +++: extensively coagulated clots (positive); ++++: complete coagulation 650 (positive) (Bactident coagulase, Merck, Darmstadt, Germany).651


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

25

652 Table 2. Summary indicating presence (+) or absence (-) of tested characteristics in strains P0, 653 P10 and x10 654

Strain

P0 P10 X10

MIC to triclosan (μg/ml) 0.23 29 1.1

Slow planktonic growth - + -

Impaired biofilm growth - + +

Impaired DNAse activity - + +

Impaired haemolytic activity - + +

Delayed coagulase activity - + +

Impaired virulence - + +

Reversal by hemin supplementation - - -

Reversal by menadione supplementation - - -

Reversal by thymidine supplementation - - -

655


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

26

656

FIG. 1. Latimer et al. 657

658

0

10

20

30

40

50

60

70

PO P5 P10 X5 X10

Strain

Tri

clo

san

co

nce

ntr

atio

n (μ

g/m

l)

659

660


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

27

661


663

664

665

A

B C


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

28


667

668

669

670

0

1

2

3

4

5

6

7

0 5 10 15 20 25 30 35Time (h)

OD

63

0


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

29


672

673

0

0.2

0.4

0.6

0.8

1

1.2

0 5 10 15 20 25 30 35 40 45

Time (h)

Bio

film

Uni

ts (

Bio

film

A 6

30/P

lank

toni

c O

D 6

30)


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

30


675

676


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

31


678

679

680

A

B


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

32


0

10

20

30

40

50

60

70

80

90

100

Ha

em

oly

sis

(%

of

wild

-ty

pe

va

lue

)

682


http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

33


684

685

686

687

688

Untreated

PBS

P10

x10

P0

A

B on February 13, 2018 by guest

http://aac.asm.org/

Dow

nloaded from

http://aac.asm.org/

attenuated virulence and biofilm formation in staphylococcus

Documents