AAC Accepted Manuscript Posted Online 4 January 2016 Antimicrob. Agents Chemother. doi:10.1128/AAC.02330-15 Copyright © 2016, American Society for Microbiology. All Rights Reserved.

1

An Optimized in vitro Antibiotic Susceptibility Testing Method for Small Colony Variant

2

Staphylococcus aureus

3 4 5

Mimi R. Precit1, Daniel J. Wolter1,2, Adam Griffith2, Julia Emerson1,2, Jane L. Burns1,2, Lucas R.

6

Hoffman1,2#;

7 8

1

University of Washington, Seattle, WA,

9

2

Seattle Children's Hospital, Seattle, WA

10 11

Running Head: S. aureus Small Colony Variant Susceptibility Testing

12 13

#Address correspondence to Lucas Hoffman: [email protected]

14 15 16 17 18 19 20 21 22 23

1

24

Abstract:

25

S. aureus small colony variants (SCVs) emerge frequently during chronic infections and are

26

often associated with worse disease outcomes. There are no standardized methods for SCV

27

antibiotic susceptibility testing (AST) due to poor growth and reversion to normal-colony (NC)

28

phenotypes on standard media. We sought to identify reproducible AST methods for S. aureus

29

SCVs, and to determine whether SCV susceptibilities can be predicted based on treatment

30

history, SCV biochemical type (auxotrophy), or the susceptibilities of isogenic NC co-isolates.

31

We tested growth and stability of SCV isolates on 11 agar media, selecting two for AST that

32

yielded optimal SCV growth and lowest reversion rates to NC phenotypes. We then performed

33

disk diffusion AST on 86 S. aureus SCVs and 28 isogenic NCs, and Etest for a subset of 26

34

SCVs and 24 isogenic NCs. Growth and reversion were optimal on brain-heart infusion agar and

35

Mueller-Hinton agar supplemented with compounds for which most clinical SCVs are

36

auxotrophic: hemin, menadione, and thymidine. SCVs were typically non-susceptible to either

37

trimethoprim-sulfamethoxazole or aminoglycosides in accordance with auxotrophy type. In

38

contrast, SCVs were variably non-susceptible to fluoroquinonolones, macrolides, lincosamides,

39

fusidic acid, and rifampin; mecA-positive SCVs were invariably resistant to cefoxitin. All

40

isolates (both SCVs and NC) were susceptible to quinupristin/dalfopristin, vancomycin,

41

minocycline, linezolid, chloramphenicol, and tigecycline. Analysis of SCV auxotrophy type,

42

isogenic NC antibiograms, and antibiotic treatment history had limited utility in predicting SCV

43

susceptibilities. With clinical correlation, this AST method and these results may prove useful in

44

directing treatment for SCV infections.

45 46

2

47

INTRODUCTION

48

Staphylococcus aureus causes diverse infections, ranging from benign to life threatening.

49

Persistent and relapsing S. aureus infections often occur despite prolonged antimicrobial therapy

50

and have been linked to the emergence of small colony variants (SCVs). S. aureus SCVs have

51

been associated with poor response to antibiotic treatment (1, 2) and are often recovered from

52

many chronic infections, including endocarditis (3, 4), osteomyelitis (5), device infections (3),

53

soft-tissue infections (6), and airway infections (7, 8). SCV airway infections are particularly

54

prevalent among people with the genetic disease cystic fibrosis (CF), affecting between 8-33% of

55

patients, often following prolonged antibiotic treatment (1, 7). In children with CF, SCV

56

respiratory infection was found to be independently associated with worse lung function and

57

faster lung function decline (9).

58

SCVs are typically identified by distinctive phenotypic traits when grown on most agar-based

59

media. SCVs produce colonies approximately one-tenth the size of normal-colony (NC) S.

60

aureus (Figure 1), and they are characteristically non-hemolytic and non-pigmented and have

61

diminished coagulase production (10). SCVs can carry mutational defects in different metabolic

62

pathways, despite sharing similar phenotypic appearances on agar surfaces. The pathways most

63

commonly affected are important for electron transport or DNA biosynthesis, resulting in

64

auxotrophy for specific nutrients. Accordingly, supplementation of growth media with

65

menadione and hemin, cofactors in menaquinone and cytochrome synthesis, respectively,

66

complement electron transport-deficient SCVs unable to synthesize these compounds (11).

67

Thymidine supplementation complements SCVs harboring mutations in thyA, which encodes

68

thymidylate synthase (12). SCVs have been isolated both in vivo and in vitro after antibiotic

69

exposure (7, 13, 14). For example, trimethoprim-sulfamethoxazole (SXT) usage has been

3

70

associated with the recovery of SXT-resistant, thymidine-dependent SCVs from CF airway

71

secretions (15), while hemin- and menadione-dependent SCVs can emerge following

72

aminoglycoside exposure (14, 16).

73

The impaired in vitro growth exhibited by SCVs presents two important challenges to the

74

clinical microbiology laboratory: They are difficult to detect using conventional approaches, and

75

there are no approved methods for SCV antimicrobial susceptibility testing (AST), hindering

76

selection of appropriate treatments. AST was developed for rapidly growing, aerobic bacteria,

77

analyzed within 24h using a standard medium (Mueller-Hinton for S. aureus); however, the

78

growth of SCVs under these conditions is generally insufficient to provide meaningful results.

79

As such, a variety of different methods using diverse media (Table S1) and extended incubation

80

have been utilized for AST of SCVs. These conditions were often chosen empirically for the

81

specific SCVs being characterized. To further complicate testing, SCVs frequently revert to NC

82

phenotypes during in vitro growth, potentially altering susceptibilities and limiting interpretation.

83

The primary goal of this study was to identify in vitro conditions that would adequately support

84

growth, minimize reversion to NC, and allow measurement of in vitro susceptibilities for a

85

diverse collection of S. aureus SCV clinical isolates. We also sought to determine whether the in

86

vitro susceptibilities of SCV isolates from this method could be reliably predicted from (1) SCV

87

auxotrophy type and methicillin resistance (MRSA), (2) susceptibilities of clonally-related NC

88

isolates from the same patients, or (3) the source patient’s antibiotic treatment history, either as

89

complements or alternatives to SCV AST. Our results indicate that this AST method, if

90

validated, could be useful in directing antibiotic treatment for S. aureus SCV infections.

91 92

4

93

MATERIALS AND METHODS

94 95

S. aureus Strains and Isolates

96

S. aureus isolates for this study comprised 85 SCVs and, for each SCV type defined by

97

pulsed-field gel electrophoresis (PFGE), at least 1 preceding or co-isolated NC isolate (n=27)

98

(Table S2). All clinical isolates were obtained from a single-center study of children (n=23) with

99

CF performed from 2008-10; details regarding the source population were described previously

100

(9). Since a menadione-dependent clinical SCV was not available, a menadione-dependent SCV

101

selected in vitro from S. aureus Newman (16) was used. S. aureus ATCC 25923 and ATCC

102

29213 were used as controls in AST, and USA300 MRSA strain JE2 (17) served as a positive

103

control for mecA PCR and β-lactam resistance in AST. This study was approved by Seattle

104

Children’s Hospital Human Subjects IRB.

105 106 107

Media Growth media used included Luria-Bertani broth and agar (LB, Becton Dickinson (BD),

108

Sparks, MD), trypticase soy agar (TSA, BD), TSA with 5% sheep’s blood (BAP, Remel, Lenexa,

109

KS), mannitol salt agar (BD), Mueller-Hinton agar (MHA, Oxoid, Basingstoke, Hampshire,

110

UK), brain heart infusion (BHI, BD), and chocolate agar (Remel). LB and BD-BHI agars were

111

prepared in-house using their respective broth powders and 1.5% w/v Bacto agar (BD).

112

Commercially available BHI agar was also obtained from Hardy Diagnostics (Hardy-BHI, Santa

113

Maria, CA) and Teknova (Teknova-BHI, Hollister, CA). A synthetic CF sputum media,

114

formulated to mimic the nutritional composition of CF sputum, was prepared as described (18),

115

with addition of 0.1 mg/mL herring sperm DNA (Sigma) (SCFMD).

5

116

Some media (TSA, BHI, MHA, and SCFMD) were supplemented as indicated with SCV

117

growth supportive nutrients (denoted by “sup-”): 5 μg/mL of thymidine, 1 μg/mL of hemin, and

118

1 μg/mL of menadione to support SCV growth. Final concentrations were chosen empirically by

119

evaluating the growth of thymidine-, hemin-, and menadione -dependent SCVs on MHA with

120

increasing supplement concentrations (Figure S1). SCVs were inoculated onto agar media from

121

freezer stocks (TSA with 15% glycerol). Supplemented media consisted of the lowest

122

concentration of each supplement that both 1) significantly improved in vitro growth of SCVs

123

auxotrophic for that supplement, and 2) didn’t inhibit growth of NC S. aureus and SCVs of other

124

auxotrophic types. (Figure S1). Previous studies indicated that 5 µg/mL thymidine would

125

approximate the sum of the concentrations of thymidine and its biologically-active metabolite,

126

dTMP, in CF sputum (19, 20). Sup-MHA was also custom-ordered from Teknova and prepared

127

using the same protocol as our in-house sup-MHA, referred to here as THM-MHA (Thymidine-

128

Hemin-Menadione Mueller-Hinton Agar).

129 130 131

Evaluation of SCV Growth Growth of a subset of SCVs (n=10, including 1 menadione-, 2 hemin-, and 7 thymidine-

132

dependent SCVs) and paired, isogenic NCs (n=10) was assessed on 11 agar media. SCVs were

133

cultured directly from frozen stocks (TSA with 15% glycerol) onto 11 different media: LB, TSA,

134

BAP, BHI, mannitol, MHA, SCFMD, sup-TSA, sup-BHI, sup-MHA, and sup-SCFMD. Plates

135

were incubated for 24h at 35°C in both aerobic and microaerobic (5% CO2) conditions. SCVs

136

with colonies similar in size and appearance to NCs (≥1mm in diameter), when grown on the

137

same media, indicated adequate growth. SCV colonies that were ≤1 mm indicated poor growth.

138

6

139 140

Evaluation of SCV Phenotypic Stability SCV stability was evaluated on SCV growth-supportive media in two ways: 1) assigning

141

a media-stability score based on visual inspection of colony morphology variation on each

142

medium, and 2) measuring frequency of reversion of SCVs to stable NC growth. Specifically,

143

SCVs were plated from frozen stocks onto media that adequately supported SCV growth (sup-

144

MHA, BHI, sup-BHI, TSA, sup-TSA), incubated at 35˚C for 24h, and colonies were inspected

145

for consistency. Stability scores reflected number of colony types observed on each medium: a

146

score of 1 indicated a single colony type, 2 indicated 2 colony types, etc. Media with one colony

147

type (sup-MHA, BD-BHI, sup-BD-BHI) for each SCV were chosen for subsequent experiments.

148

The frequency of reversion was determined by sub-culturing 150 randomly-chosen colonies

149

≥1mm from the test media onto LB agar, which identifies SCVs (Figure S2), incubating at 35°C

150

for 24h and measuring colony size. Frequency of reversion was calculated by dividing the

151

number of colonies that were ≥1mm in diameter on LB agar by the total colony number.

152 153

Antimicrobial Susceptibility Testing

154

Disk diffusion was conducted on all SCVs using 24 antibiotics (See Table 1 for list)

155

according to CLSI guidelines (21) except with BHI and sup-MHA substituted for MHA. S.

156

aureus ATCC 25923 and NC isolates were also tested on MHA, BHI, and sup-MHA to compare

157

the effects of media and supplementation on AST. For a subset of isolates (26 SCVs, 13

158

corresponding NCs, and ATCC 25923), disk diffusion was also performed under microaerobic

159

conditions (5% CO2). AST results were interpreted according to CLSI guidelines (21, 22) for all

160

antibiotics except fusidic acid, for which EUCAST guidelines were used (23). CLSI guidelines

7

161

formulated in 2009 (22) were used to interpret β-lactam susceptibilities since these breakpoints

162

are absent from the 2014 guidelines.

163

Minimum inhibitory concentrations (MICs) were determined with Etest for 13 antibiotics

164

(See Table 2 for list) using a subset of isolates (28 SCVs and 24 respective NCs) and ATCC

165

29213 on THM-MHA (Teknova). In parallel, ATCC 29213 MICs were determined on standard

166

MHA from Teknova to compare the effects of supplementation on susceptibility.

167

SCVs, paired isogenic NC isolates, and ATCC control strains were sub-cultured onto

168

chocolate agar plates from frozen stocks and grown aerobically at 35°C overnight. Chocolate

169

agar was validated to support both SCV growth and stability in preliminary, control experiments

170

(not shown). Colonies from individual isolates were suspended in 0.85% saline to 0.5 McFarland

171

(~1x108 cfu/mL), which was swabbed onto agar surfaces of either sup-MHA, BD-BHI, THM-

172

MHA, or MHA (MHA used only for NC and control strains). Antibiotic disks or Etest strips

173

were added as indicated. Plates were incubated aerobically or microaerobically (with 5% CO2) at

174

35°C for 18-24h. All antibiotic disks were obtained from BD except fusidic acid, which was

175

prepared with powder (Sigma, St. Louis, MO) and filter disks (BD) at 10 μg/disk. Etests were

176

obtained from Biomerieux, Durham, NC.

177 178 179

Antimicrobial Susceptibility Testing: Presence of thymidine in AST media Because sup-MHA contains antagonists of trimethoprim-sulfamethoxazole (SXT)

180

activity (e.g. thymidine), the CLSI recommends that slight growth (≤ 20% of lawn growth)

181

around a SXT disk or strip, if present, should be ignored, and measurements made at the margin

182

of heaviest growth (21). SXT susceptible, control organisms (ATCC 25912 and ATCC 29213)

183

display slight growth around SXT disks/strips on sup-MHA and BD-BHI, the latter of which

8

184

presumably contains thymidine, and so we applied the ≤ 20% guideline for both media (Figure

185

S3).

186 187

MRSA Detection

188 189

MRSA were identified using the PBP2’ latex agglutination kit (Oxoid), with confirmation by mecA gene detection using PCR as described (24, 25).

190 191

PFGE

192

Genetic relatedness of clinical isolates was determined by PFGE as described (26).

193

Briefly, chromosomal DNA was digested with 50 U of SmaI (Roche) at room temperature for 3h,

194

and DNA restriction fragments were separated in a 1% SeaKem Gold agarose gel using a CHEF

195

DR-III (Bio-Rad, Hercules, California). Dendrograms were constructed by the UPGMA

196

(Unweighted Pair Group Method with Arithmetic mean) clustering method.

197 198 199

Correlation of Susceptibilities with Antibiotic Exposure of Source Patients The relationship between antibiotic exposure during the 90 days prior to isolation of an

200

SCV and susceptibility to that antibiotic was determined using antibiotic treatment data from the

201

source study (9). Antibiotic exposure was identified if the end date for treatment with that

202

antibiotic occurred within 90 days prior to the culture date. Antibiotic end dates were missing for

203

15 of 407 antibiotic courses identified; in those cases, end dates were assigned a value equal to

204

(start date +28 days) for inhaled tobramycin, (start date + 180 days) for maintenance oral

205

azithromycin, or (start date + 14 days) for other antibiotics. Logistic regression models

206

accounted for multiple isolates per subject and used robust variance estimates. The odds ratio

9

207

with 95% confidence interval was calculated as the measure of association between the predictor

208

(antibiotic exposure within 90 days prior to culture date) and the outcome (isolate not susceptible

209

to the antibiotic tested). Models could not be fit when the antibiotic exposure perfectly predicted

210

isolate status as susceptible or not susceptible (these observations were dropped from the model

211

leaving no variability in the predictor variable). This analysis excluded 3 replicate SCVs from

212

the same culture as another SCV and 11 isolates for which antibiotic treatment data were

213

incomplete, leaving 71 isolates from 21 subjects as the analyzed dataset.

214 215

RESULTS

216 217 218

SCV Growth and Phenotypic Stability on Various Media All SCVs grew poorly (menadione-dependent and hemin-dependent SCVs) or not at all

219

(thymidine-dependent SCVs) on MHA, the Clinical Laboratory Standards Institute (CLSI)-

220

recommended AST medium (21) (Figure 2). By comparison, all SCVs grew similar to isogenic

221

NC isolates on 5 media: supplemented (sup) and unsupplemented TSA and BD-BHI agar, and

222

sup-MHA. Media supplementation indicates the addition of SCV-growth supportive nutrients:

223

thymidine, hemin, and menadione (Figure S1). Surprisingly, thymidine-dependent SCVs did not

224

grow on multiple lots each of BHI agar from two other companies (Hardy-BHI and Teknova-

225

BHI), indicating manufacturing differences. The medium formulated to mimic CF sputum

226

(SCFMD) did not support growth of all SCVs, even when supplemented with thymidine, hemin,

227

and menadione (Figure 2). SCV growth on the tested media was comparable under aerobic and

228

microaerobic atmospheric conditions (not shown).

10

229

Improved SCV growth may either be due to presence of growth-supportive nutrients or to

230

genetic reversion. In the first case, in vitro susceptibilities of SCVs on a medium would be

231

measured under chemically complementary conditions. In the latter case, results would reflect

232

the susceptibilities of in vitro-selected NC revertants, rather than SCVs. SCVs grew with

233

substantial morphological variability (stability score > 1) on TSA and sup-TSA, indicative of

234

genetic instability and reversion, while other media supported uniform growth (e.g., sup-MHA)

235

(Figure 3A). To determine whether uniform colony sizes indicated SCV stability, we compared

236

reversion frequencies for each SCV on the media that supported uniform growth and on blood

237

agar (BAP), a medium previously used in SCV AST (7, 19). Reversion frequencies were lowest

238

on sup-MHA under both aerobic and microaerobic conditions (Figure 3B). BAP yielded the

239

highest reversion rates under both atmospheric conditions and, on this medium, thymidine-

240

dependent SCVs exhibited a distinctive, anucleate cellular morphology upon Gram’s staining

241

(27), indicating poor growth (not shown). Sup-MHA and BD-BHI were therefore chosen for

242

AST as they best supported growth and maintained SCV phenotypic stability. While SCVs grow

243

to approximately NC sizes on these media, these isolates retain SCV phenotypes upon sub-

244

culturing to LB, and for simplicity we will refer to these isolates as SCVs even when cultivated

245

on growth-complementing media.

246 247 248

AST: Aerobic Disk Diffusion To determine how closely disk diffusion zone diameters on our test media would overlap with

249

the acceptable range-limits defined by the CLSI, we tested the susceptibility of control strain

250

ATCC 25923 on MHA (CLSI-recommended), sup-MHA, and BD-BHI. The average zone

251

diameters and standard deviations for ATCC 25923 fell within CLSI-accepted range-limits for

11

252

each antibiotic on all media with a few exceptions on BD-BHI (Figure 4). Susceptibilities of a

253

subset of 14 NC clinical isolates (8 MSSA and 6 MRSA) were also identical on MHA, sup-

254

MHA, and BD-BHI (not shown). Therefore, supplementation of MHA with thymidine, hemin,

255

and menadione did not significantly alter antibiotic susceptibilities of control isolates, suggesting

256

this medium is reliable for S. aureus AST. Although control strain results fell outside acceptable

257

limits for several antibiotics on BD-BHI, susceptibility testing was conducted with both BD-BHI

258

and sup-MHA for the complete set of isolates for comparison.

259

AST was performed on 112 clinical isolates (including SCVs and isogenic NCs), a

260

menadione-dependent SCV selected in vitro from S. aureus strain Newman (16), and S. aureus

261

Newman (Table S2). The clinical isolates belonged to 15 distinct PFGE groups and included

262

both MSSA (n=71) and MRSA (n=43). MSSA isolates belonged to 9 different PFGE groups, and

263

MRSA isolates comprised 6 distinct PFGE groups with just over half (n=25) belonging to the

264

USA100 pulsotype (Figure S4). A single isogenic MRSA NC/SCV pair belonged to the USA700

265

pulsotype.

266

All isolates grew on BD-BHI and sup-MHA, and zone diameters were measurable after

267

18-24h. No discrepancies in AST results were observed between BD-BHI and sup-MHA for

268

thymidine-dependent SCVs (MSSA and MRSA). All thymidine-dependent SCVs were resistant

269

to trimethoprim-sulfmethoxazole, while NC isolates were susceptible (Table 1). Collectively,

270

thymidine-dependent SCVs tended to be non-susceptible to azithromycin. Small percentages

271

(~12-30%) of thymidine-dependent SCV-MSSA were non-susceptible to fluoroquinolones,

272

aminoglycosides, clindamycin, telithromycin, and rifampin. Fewer thymidine-dependent SCV-

273

MRSA were susceptible to these antibiotics than thymidine-dependent SCV-MSSA (Table 1).

12

274

In contrast with thymidine-dependent SCVs, hemin- and menadione-dependent SCV-

275

MSSA exhibited different susceptibilities on BD-BHI and sup-MHA. While these SCVs were

276

generally susceptible to most antibiotics, they were all non-susceptible to the aminoglycosides

277

tobramycin and amikacin on BD-BHI (Table 1). However, these SCVs exhibited slightly larger

278

zone diameters (by 2-8 mm) on sup-MHA that, in some cases, rendered them aminoglycoside-

279

susceptible. Hemin-dependent SCV-MRSA were uniformly non-susceptible to both

280

aminoglycosides (Table 1). An SCV of undetermined auxotrophy was resistant only to

281

trimethoprim-sulfamethoxazole (Table 1).

282

MRSA control strain JE2 and all S. aureus isolates were screened for methicillin

283

resistance using oxacillin and cefoxitin disks. JE2 was resistant to both antibiotics, with zone

284

diameters comparable (±1 mm) between MHA, BD-BHI, and sup-MHA. Cefoxitin resistance

285

was detected for all NC mecA-positive isolates on these media. Oxacillin resistance was detected

286

in 67% and 77% of these isolates on BHI and sup-MHA respectively. Similarly, 97% of mecA-

287

positive thymidine-dependent SCVs were resistant to cefoxitin, and 67.7% to oxacillin, on both

288

media. A single mecA-positive, thymidine-dependent SCV was susceptible to both antibiotics on

289

BD-BHI, sup-MHA, and BAP. Similar results were observed for hemin-dependent SCV-MRSA

290

(Table 1). The mecA-negative MSSA-SCVs were β-lactam susceptible. Therefore, MRSA status

291

correlated well with cefoxitin resistance for NCs and SCVs.

292

All isolates were susceptible to minocycline, chloramphenicol, quinupristin/dalfopristin,

293

linezolid, and tigecycline (Table 1). All but 3 MRSA isolates (1-NC, 2-SCVs) were susceptible

294

to fusidic acid. While MRSA status correlated well with β-lactam resistance, and thymidine

295

auxotrophy correlated well with trimethoprim-sulfamethoxazole resistance, susceptibility to

13

296

other clinically-useful classes (including aminoglycosides, fluoroquinolones, macrolides,

297

clindamycin, and rifampin) could not be reliably predicted from those two characteristics.

298 299 300

AST: Microaerobic Disk Diffusion The CF lung has areas with limited oxygen (28), and SCVs frequently exhibit metabolic

301

characteristics (29) that could impact growth under different atmospheric conditions. Therefore,

302

we compared AST results performed under microaerobic versus aerobic conditions. ATCC

303

25923 gave similar results under both conditions on BD-BHI. On MHA and sup-MHA, zone

304

sizes for cefoxitin, chloramphenicol, oxacillin, and tigecycline fell out of CLSI ranges only under

305

microaerobic conditions (Figure 4 and Figure S5).

306

Microaerobic susceptibilities for the vast majority of clinical isolates tested (26 SCVs and

307

13 paired isogenic NCs) were identical to those from aerobic conditions, with one exception.

308

Four clonally-related, thymidine-dependent, mecA-positive isolates were non-susceptible to

309

cefoxitin and oxacillin under aerobic conditions, but were susceptible to both under microaerobic

310

conditions (not shown).

311 312 313

Comparison of disk diffusion AST on in-house and commercially-prepared sup-MHA To this point, method development was completed with in-house prepared sup-MHA. To

314

test whether commercially-prepared sup-MHA would perform similarly to our own, we asked

315

Teknova to prepare sup-MHA (referred to here as THM-MHA) using our supplementation

316

protocol. As a quality control, disk-diffusion antibiograms for a subset of 5 clinical SCVs,

317

corresponding isogenic NC isolates and Control Strain ATCC 25923 were determined on

14

318

Teknova-MHA (NC and the control strain only) and THM-MHA (all isolates). Susceptibilities

319

for all tested strain and isolates were identical between sup-MHA and THM-MHA (not shown).

320 321 322

AST: Etest Etest was performed using THM-MHA to obtain MICs. As a quality-control step, we

323

compared susceptibilities of S. aureus ATCC 29213 on THM-MHA and Teknova-MHA. MICs

324

fell within the CLSI acceptable range limits for all drugs tested (Figure 5) on both media. Thus,

325

supplementation of the CLSI-recommended base medium did not influence susceptibilities of the

326

S. aureus control strain.

327

We performed Etest on a subset of the isolates (24-NC and 26-SCVs) previous analyzed by

328

disk diffusion. All isolates grew on THM-MHA, and MICs were measurable after 24h. As

329

cefazolin and meropenem Etest strips are not available, cephalothin and imipenem (respectively)

330

were used as surrogates. Vancomycin AST was performed only by Etest because there are no

331

established vancomycin disk diffusion breakpoints; all isolates were vancomycin-susceptible

332

(Table 2). No discrepancies between disk diffusion and Etest results were observed.

333 334

NC Susceptibilities as a Predictor of SCV Susceptibilities

335

We evaluated whether in vitro susceptibilities of SCVs could be predicted from

336

antibiograms of isogenic NC S. aureus isolated before or concurrently with SCVs from the same

337

patients. As stated in the introduction, exposure to either trimethoprim-sulfamethoxazole or

338

aminoglycosides (amikacin or tobramycin) has been linked to SCV emergence, and thus

339

NC/SCV susceptibility differences were expected for these drugs. However, some SCV/NC pairs

340

also differed in susceptibility to other antibiotics (Table 3), and susceptibilities of some

15

341

genetically-related SCVs from the same patient differed as well (Table S3). Therefore, the

342

susceptibilities of SCVs cannot be consistently predicted by isogenic isolates from the same

343

patients, indicating that in vitro susceptibilities for an SCV can only be reliably defined by

344

testing that isolate.

345 346

Antibiotic Exposure as a Predictor of SCV Susceptibilities

347

To determine whether recent exposure history to a specific antibiotic reliably predicts

348

SCV resistance to that antibiotic (or others in its class), we analyzed the relationship between

349

SCV in vitro susceptibilities on sup-MHA and antibiotic treatment data from the parent study

350

(9). As shown in Table 4, SCVs were frequently, but not always, non-susceptible to antibiotics to

351

which the source patient was recently exposed. Exposures to tobramycin and trimethoprim-

352

sulfamethoxazole were each associated with significantly higher odds that subsequent SCVs

353

would be non-susceptible to those antibiotics on logistic regression (OR (95% CI) 18.1 (3.2,

354

100.9) and 2.5 (0.3, 20.6), respectively). However, the correlation was not 100%. While

355

treatment with other antibiotics was relatively less frequent prior to SCV isolation, SCVs isolated

356

after azithromycin, clindamycin, or rifampin exposure were invariably resistant to those drugs;

357

by contrast, SCVs isolated after fluoroquinolone treatment were uniformly susceptible to

358

levofloxacin. There were no other antibiotic exposures identified in SCV-positive subjects.

359

Therefore, history of exposure to all above antibiotics but fluoroquinolones may provide

360

clinically meaningful rationale for avoiding those drugs when treating SCV infections, assuming

361

in vitro susceptibilities accurately predict clinical response. Nevertheless, these results indicate

362

that treatment history, as for other candidate predictors, has limited utility in predicting in vitro

363

susceptibility results.

16

364

DISCUSSION

365

SCVs have been selected in vitro following antibiotic exposure, and antibiotic treatment

366

has been associated with the subsequent SCV detection (10, 30). Therefore, antibiotic selection

367

plays key roles in SCV emergence and subsequent SCV susceptibilities. While AST is standard

368

for identifying appropriate therapies, the growth characteristics of SCVs render them

369

incompatible with recommended AST methods. Several studies have reported susceptibilities for

370

limited numbers of SCVs using various in vitro methods (Table S1), but media vary substantially

371

in their abilities to support SCV growth and minimize reversion (including the same medium

372

from different manufacturers, as shown here). We found that two media, BD-BHI and sup-MHA

373

(commercially, THM-MHA) allowed for reproducible AST determination within 24h while

374

minimizing reversion. The resulting in vitro susceptibilities could not be reliably predicted from

375

clinical or biochemical characteristics, nor from the susceptibilities of genetically-related NC

376

isolates from within patients.

377

As expected, all thymidine-dependent SCVs were resistant to trimethoprim-sulfamethoxazole

378

on both media, and hemin- and menadione-dependent SCVs were non-susceptible to

379

aminoglycosides (at least on BD-BHI). Similarly, MRSA isolates were generally resistant to

380

β−lactams. Therefore, our results indicate that specific AST elements may be reliably predicted

381

from either SCV auxotrophy or MRSA status. However, discrepancies existed even for these

382

simple comparisons (discussed in detail below). SCV and genetically-related NC antibiograms

383

were similar, but not always, indicating that NC isolate susceptibilities may be used as a rough

384

guide for SCV treatment at best. Our results also indicate that SCVs from within patients can

385

exhibit diverse susceptibilities (Table S3). Similar diversity within CF patients has been

386

identified for P. aeruginosa (31, 32). While we did not determine whether isolates exhibited

17

387

hypermutability, which has been found in CF S. aureus isolates (34), this characteristic could

388

impact the tempo of this diversification and adaptation, including the development of resistance

389

and SCV emergence. Similarly, genotyping our isolates with multilocus sequence typing would

390

allow for defining clonal complexes, an additional, portable candidate predictor of

391

susceptibilities. These analyses are therefore of interest for future studies.

392

While azithromycin resistance was common among SCVs regardless of auxotrophy, prior

393

work found no association between azithromycin usage and SCV detection (33), indicating

394

SCVs were not likely selected by azithromycin. Rather, these results probably reflect common

395

macrolide usage in the CF population (34). Similarly, SCV-MRSA were more often multidrug-

396

resistant compared to SCV-MSSA, likely reflecting higher antibiotic exposure intensities among

397

MRSA-positive CF patients (35, 36).

398

Identifying MRSA is an important function of the clinical laboratory, as these isolates are

399

difficult to treat and may be associated with worse patient outcomes (36). Disk diffusion on BD-

400

BHI and sup-MHA correctly identified SCV-MRSA using cefoxitin disks. False negatives were

401

encountered using oxacillin. During manuscript preparation, another study reported detection of

402

thymidine-dependent SCV-MRSA through agar media supplementation with 10 μg/mL

403

thymidine (37). Our study supports this approach (albeit with 5 µg/mL thymidine) and extends it

404

to other auxotrophic types by supplementing menadione and hemin. Our results further show that

405

AST with supplemented media is reproducible and useful beyond detecting methicillin

406

resistance.

407

While SCVs generally grew well on BD-BHI and sup-MHA with minimal reversion, each

408

medium had notable limitations. The control strain, ATCC 29523, did not always exhibit

409

susceptibilities within the CLSI-acceptable ranges for certain antibiotics on BHI, including

18

410

tigecycline, and this strain was often more susceptible to specific drugs on BD-BHI than sup-

411

MHA. Thymidine-dependent SCVs grew on BHI agar from one manufacturer but not from two

412

others, suggesting formula variations impacted growth. The media developed here (sup-MHA

413

and THM-MHA) used the CLSI-recommended base medium, MHA, supplemented with

414

compounds required by SCVs for optimal growth. Concentrations were chosen based on the best

415

available knowledge about their abundance in CF sputum (19, 38) as well as their empirically-

416

determined abilities to support SCV growth and stability (Figure S1). This medium eliminated

417

manufacturer-to-manufacturer differences. Most importantly, supplementation of MHA did not

418

impact susceptibilities of ATCC 29523 (disk diffusion) or ATCC 29213 (Etest); zone diameters

419

and MICs fell within CLSI-acceptable ranges and were comparable to unsupplemented MHA

420

(Figure 4,5). AST results were comparable on MHA and sup-MHA for NC isolates tested. The

421

only susceptibility differences observed between BD-BHI and sup-MHA was for

422

aminoglycosides among hemin- and menadione-dependent SCVs, both of which have defects in

423

electron transport and, consequently, in aminoglycoside uptake (29). Therefore, these

424

susceptibility differences likely reflect different hemin and menadione concentrations.

425

One could argue that stimulating the growth of SCVs through media supplementation

426

masks true susceptibilities. However, SCVs reach sputum densities in CF children similar to

427

those of NC S. aureus (9), and thymidine concentrations have been detected in CF sputum at

428

levels similar to those in sup-MHA (19, 38). In contrast with thymidine, the concentrations of

429

hemin and the naturally-occurring menadione analogs (menaquinones) have not been defined for

430

CF sputum, but they may be sufficient in CF airways to support SCV growth. Thus, while the in

431

vivo growth characteristics and antibiotic susceptibilities of SCVs are not known, these behaviors

432

could be approximated by growth on supplemented media.

19

433

While the AST method developed here was applied to CF respiratory SCVs, this

434

methodology could extend to diverse S. aureus SCV infections. This method may or may not

435

reflect in vivo antibiotic response of SCVs, but we identified 100% susceptibility among a large

436

collection of clinical SCVs to several antibiotics (linezolid, tigecycline, chloramphenicol,

437

minocycline, and quinupristin/dalfopristin) that may prove useful as empirical choices for

438

treating SCV infections in the absence of AST. We suggest that antibiotic treatment of S. aureus

439

SCV infection using the results (for testing empirical choices) and methods (for testing AST-

440

guided choices) described here should form the basis of prospective clinical trials, preferably

441

measuring both clinical and microbiological outcomes.

442 443 444

FUNDING INFORMATION:

445

This work was supported by the NIH (K02HL105543-01, ESR01A29422 and P30DK089507),

446

CFF (HOFFMA07P0), ATS (CF-07-003), and a Seattle Children’s Hospital Ignition Award, and

447

an NSF fellowship.

448 449

ACKNOWLEDGEMENTS:

450

We thank S. Swanzy, A.M. Buccat and X. Qin for advice, reagents, and protocols, and L.

451

McDougal for patterns used in PFGE analysis.

452 453 454 455

20

456

REFERENCES

457

1.

Proctor RA, van Langevelde P, Kristjansson M, Maslow JN, Arbeit RD. 1995.

458

Persistent and relapsing infections associated with small-colony variants of

459

Staphylococcus aureus. Clin Infect Dis 20:95–102.

460

2.

Besier S, Smaczny C, Mallinckrodt von C, Krahl A, Ackermann H, Brade V,

461

Wichelhaus TA. 2007. Prevalence and clinical significance of Staphylococcus aureus

462

small-colony variants in cystic fibrosis lung disease. J Clin Microbiol 45:168–172.

463

3.

Maduka-Ezeh A, Seville MT, Kusne S, Vikram HR, Blair JE, Greenwood-

464

Quaintance K, Arabia F, Patel R. 2012. Thymidine auxotrophic Staphylococcus aureus

465

small-colony variant endocarditis and left ventricular assist device infection. J Clin

466

Microbiol 50:1102–1105.

467

4.

468 469

Baddour LM, Christensen GD. 1987. Prosthetic valve endocarditis due to small-colony staphylococcal variants. Rev Infect Dis 9:1168–1174.

5.

Eiff von C, Bettin D, Proctor RA, Rolauffs B, Lindner N, Winkelmann W, Peters G.

470

1997. Recovery of small colony variants of Staphylococcus aureus following gentamicin

471

bead placement for osteomyelitis. Clin Infect Dis 25:1250–1251.

472

6.

Abele-Horn M, Schupfner B, Emmerling P, Waldner H, Göring H. 2000. Persistent

473

wound infection after herniotomy associated with small-colony variants of Staphylococcus

474

aureus. Infection 28:53–54.

475 476

7.

Kahl B, Herrmann M, Everding AS, Koch HG, Becker K, Harms E, Proctor RA, Peters G. 1998. Persistent infection with small colony variant strains of Staphylococcus

21

477 478

aureus in patients with cystic fibrosis. J Infect Dis 177:1023–1029. 8.

Besier S, Zander J, Siegel E, Saum SH, Hunfeld K-P, Ehrhart A, Brade V,

479

Wichelhaus TA. 2008. Thymidine-dependent Staphylococcus aureus small-colony

480

variants: human pathogens that are relevant not only in cases of cystic fibrosis lung

481

disease. J Clin Microbiol 46:3829–3832.

482

9.

Wolter DJ, Emerson JC, McNamara S, Buccat AM, Qin X, Cochrane E, Houston LS,

483

Rogers GB, Marsh P, Prehar K, Pope CE, Blackledge M, Déziel E, Bruce KD,

484

Ramsey BW, Gibson RL, Burns JL, Hoffman LR. 2013. Staphylococcus aureus small-

485

colony variants are independently associated with worse lung disease in children with

486

cystic fibrosis. Clin Infect Dis 57:384–391.

487

10.

Proctor RA, Eiff von C, Kahl BC, Becker K, McNamara P, Herrmann M, Peters G.

488

2006. Small colony variants: a pathogenic form of bacteria that facilitates persistent and

489

recurrent infections. Nat Rev Microbiol 4:295–305.

490

11.

Proctor RA, Kahl B, Eiff von C, Vaudaux PE, Lew DP, Peters G. 1998.

491

Staphylococcal small colony variants have novel mechanisms for antibiotic resistance.

492

Clin Infect Dis 27 Suppl 1:S58–74.

493

12.

Chatterjee I, Kriegeskorte A, Fischer A, Deiwick S, Theimann N, Proctor RA, Peters

494

G, Herrmann M, Kahl BC. 2008. In vivo mutations of thymidylate synthase (encoded by

495

thyA) are responsible for thymidine dependency in clinical small-colony variants of

496

Staphylococcus aureus. J Bacteriol 190:834–842.

497

13.

Schaaff F, Bierbaum G, Baumert N, Bartmann P, Sahl H-G. 2003. Mutations are

22

498

involved in emergence of aminoglycoside-induced small colony variants of

499

Staphylococcus aureus. Int J Med Microbiol 293:427–435.

500

14.

Balwit JM, van Langevelde P, Vann JM, Proctor RA. 1994. Gentamicin-resistant

501

menadione and hemin auxotrophic Staphylococcus aureus persist within cultured

502

endothelial cells. J Infect Dis 170:1033–1037.

503

15.

Kahl BC, Belling G, Reichelt R, Herrmann M, Proctor RA, Peters G. 2003.

504

Thymidine-dependent small-colony variants of Staphylococcus aureus exhibit gross

505

morphological and ultrastructural changes consistent with impaired cell separation. J Clin

506

Microbiol 41:410–413.

507

16.

Hoffman LR, Déziel E, D'Argenio DA, Lépine F, Emerson J, McNamara S, Gibson

508

RL, Ramsey BW, Miller SI. 2006. Selection for Staphylococcus aureus small-colony

509

variants due to growth in the presence of Pseudomonas aeruginosa. Proc Natl Acad Sci

510

USA 103:19890–19895.

511

17.

Diep BA, Gill SR, Chang RF, Phan TH, Chen JH, Davidson MG, Lin F, Lin J,

512

Carleton HA, Mongodin EF, Sensabaugh GF, Perdreau-Remington F. 2006.

513

Complete genome sequence of USA300, an epidemic clone of community-acquired

514

meticillin-resistant Staphylococcus aureus. Lancet 367:731–739.

515

18.

516 517 518

Palmer KL, Aye LM, Whiteley M. 2007. Nutritional cues control Pseudomonas aeruginosa multicellular behavior in cystic fibrosis sputum. J Bacteriol 189:8079–8087.

19.

Besier S, Ludwig A, Ohlsen K, Brade V, Wichelhaus TA. 2007. Molecular analysis of the thymidine-auxotrophic small colony variant phenotype of Staphylococcus aureus. Int J

23

519 520

Med Microbiol 297:217–225. 20.

Zander J, Besier S, Saum SH, Dehghani F, Loitsch S, Brade V, Wichelhaus TA.

521

2008. Influence of dTMP on the phenotypic appearance and intracellular persistence of

522

Staphylococcus aureus. Infect Immun 76:1333–1339.

523

21.

Clinical and Laboratory Standards Institute. 2014. Performance standards for

524

antimicrobial susceptibility testing: 24th informational supplement. Document M100-S24.

525

Wayne, PA.

526

22.

Clinical and Laboratory Standards Institute 2009. Performance Standards for

527

Antimicrobial Susceptibility Testing: 19th informational supplement. Document M100-

528

S19. Wayne, PA

529

23.

Matuschek E, Brown DFJ, Kahlmeter G. 2014. Development of the EUCAST disk

530

diffusion antimicrobial susceptibility testing method and its implementation in routine

531

microbiology laboratories. Clin Microbiol Infect 20:O255–66.

532

24.

Mantion B, Cavalié L, Prère M-F. 2015. Evaluation of an immunochromatographic

533

assay for detection of PBP2a on Staphylococcus aureus clinical isolates. J Microbiol

534

Methods 112:46–48.

535

25.

Adaleti R, Nakipoglu Y, Karahan ZC, Tasdemir C, Kaya F. 2008. Comparison of

536

polymerase chain reaction and conventional methods in detecting methicillin-resistant

537

Staphylococcus aureus. J Infect Dev Ctries 2:46–50.

538

24

539

26.

McDougal LK, Steward CD, Killgore GE, Chaitram JM, McAllister SK, Tenover

540

FC. 2003. Pulsed-field gel electrophoresis typing of oxacillin-resistant Staphylococcus

541

aureus isolates from the United States: establishing a national database. J Clin Microbiol

542

41:5113–5120.

543

27.

Osawa K, Kataoka N, Maruo T. 2011. Comparison of the quick Gram stain method to

544

the B&M modified and favor methods. Rinsho Biseibutshu Jinsoku Shindan Kenkyukai

545

Shi 22:23–25.

546

28.

Worlitzsch D, Tarran R, Ulrich M, Schwab U, Cekici A. 2002. Reduced oxygen

547

concentrations in airway mucus contribute to the early and late pathogenesis of

548

Pseudomonas aeruginosa cystic fibrosis airway infection. J Clin Invest.

549

29.

Proctor RA, Kriegeskorte A, Kahl BC, Becker K, Löffler B, Peters G. 2014.

550

Staphylococcus aureus Small Colony Variants (SCVs): a road map for the metabolic

551

pathways involved in persistent infections. Front Cell Infect Microbiol 4:99.

552

30.

553 554

Kahl BC. 2014. Small colony variants (SCVs) of Staphylococcus aureus--a bacterial survival strategy. Infect Genet Evol 21:515–522.

31.

Foweraker JE, Laughton CR, Brown DFJ, Bilton D. 2005. Phenotypic variability of

555

Pseudomonas aeruginosa in sputa from patients with acute infective exacerbation of

556

cystic fibrosis and its impact on the validity of antimicrobial susceptibility testing. J

557

Antimicrob Chemother 55:921–927.

558 559

32.

Mayer-Hamblett N, Rosenfeld M, Gibson RL, Ramsey BW, Kulasekara HD, RetschBogart GZ, Morgan W, Wolter DJ, Pope CE, Houston LS, Kulasekara BR, Khan U,

25

560

Burns JL, Miller SI, Hoffman LR. 2014. Pseudomonas aeruginosa In Vitro Phenotypes

561

Distinguish Cystic Fibrosis Infection Stages and Outcomes. American Journal of

562

Respiratory and Critical Care Medicine 140617081504001.

563

33.

Green N, Burns JL, Mayer-Hamblett N, Kloster M, Lands LC, Anstead M, Ratjen F,

564

Saiman L. 2011. Lack of association of small-colony-variant Staphylococcus aureus

565

strains with long-term use of azithromycin in patients with cystic fibrosis. J Clin

566

Microbiol 49:2772–2773.

567

34.

Prunier A-L, Malbruny B, Laurans M, Brouard J, Duhamel J-F, Leclercq R. 2003.

568

High Rate of Macrolide Resistance in Staphylococcus aureus Strains from Patients with

569

Cystic Fibrosis Reveals High Proportions of Hypermutable Strains. J Infect Dis 187:1709–

570

1716.

571

35.

Ren CL, Morgan WJ, Konstan MW, Schechter MS, Wagener JS, Fisher KA,

572

Regelmann WE. 2007. Presence of methicillin resistant Staphylococcus aureus in

573

respiratory cultures from cystic fibrosis patients is associated with lower lung function.

574

Pediatric Pulmonology 42:513–518.

575

36.

Sawicki GS, Rasouliyan L, Ren CL. 2012. The Impact of MRSA on Lung Function in

576

Patients with Cystic Fibrosis. American Journal of Respiratory and Critical Care Medicine

577

179:734–735.

578

37.

Horiuchi K, Matsumoto T, Ota Y, Kasuga E, Negishi T, Yaguchi T, Sugano M,

579

Honda T. 2015. Addition of thymidine to culture media for accurate examination of

580

thymidine-dependent small-colony variants of methicillin-resistant Staphylococcus

26

581 582

aureus: A pilot study. J Microbiol Methods 110:40–44. 38.

Zander J, Besier S, Faetke S, Saum SH, Müller V, Wichelhaus TA. 2010.

583

Antimicrobial activities of trimethoprim/sulfamethoxazole, 5-iodo-2'-deoxyuridine and

584

rifampicin against Staphylococcus aureus. Int J Antimicrob Agents 36:562–565.

585

27

586

FIGURE LEGENDS

587 588

Figure 1. Distinguishing phenotypic features of SCVs. In vitro growth comparison of a NC

589

isolate (left) and a genetically related SCV (right) on blood agar showing characteristics that

590

commonly distinguish SCVs, including reduced hemolysis, small colony morphology, and

591

decreased pigmentation.

592 593

Figure 2. Quality of SCV growth as determined by visual inspection of colony size on agar

594

based media. (A) 10 SCVs were characterized by colony size as an indicator of adequate growth

595

on various media. *SCV and NC colony growth were tested at a later date on THM-MHA

596

(Thymidine, Hemin, Menadione-MHA, which is chemically identical to sup-MHA, but

597

manufactured by Teknova) and Teknova manufactured unsupplemented MHA (Teknova-MHA)

598

to confirm that isolates grew comparably on commercially available and in-house prepared sup-

599

and unsupplemented MHA. Abbreviations: Luria-Bertani (LB), trypticase soy agar + 5% sheep

600

blood (BAP), Mueller-Hinton Agar (MHA), trypticase soy agar (TSA), brain heart infusion

601

(BHI), synthetic CF sputum media supplemented herring sperm DNA (SCFMD). Media

602

designated with “sup-” indicates base media (MHA, TSA, BHI, SCFMD) supplemented with

603

SCV growth supportive nutrients thymidine, hemin, and menadione.

604 605

Figure 3. Morphological variability and quantification of reversion frequency as measures

606

of SCV stability. (A) TD-SCVs grown on TSA (left), sup-TSA (middle), and sup-MHA (right)

607

displayed variable colony morphologies on TSA and sup-TSA, indicative of reversion. However,

608

on sup-MHA a single uniform colony was observed (a single, representative TD-SCV is shown).

609

All SCVs tested showed comparable, uniform growth on sup-MHA, sup-BHI, and BHI (only 28

610

sup-MHA shown). (B) SCV isolates were grown on media listed on the x-axis overnight after

611

which colonies selected at random were sub-cultured onto LB (discriminatory medium) to

612

distinguish between reversion to NC growth and complementation by medium chemical content.

613

The percentage of colonies exhibiting reversion is shown on the y-axis. *Commercially prepared

614

sup-MHA (THM-MHA) was validated at a later date.

615 616

Figure 4. Average zone diameters for ATCC 25923 in air. The average zone size for the

617

control strain ATCC 25923 fell within CLSI-accepted ranges (horizontal red lines) for MHA

618

(black circles) and sup-MHA (light grey squares) for all antibiotics tested, and for the majority of

619

antibiotics tested on BD-BHI (dark grey triangles). Error bars depict the standard deviation for

620

15 replicates. The average zone diameter for tigecycline and the standard deviations for

621

levofloxacin, cefoxitin, and oxacillin fell outside of the CLSI-defined acceptable range-limits on

622

BHI. Fusidic acid does not have CLSI-defined range-limits; therefore, no red lines are shown.

623

Although a range limit for ATCC 25923 has not been reported for fusidic acid, average zone

624

diameters and standard deviations for this antibiotic were comparable on all three media and well

625

above the EUCAST breakpoint (≥ 24 mm).

626 627 628

Figure 5. Median MICs for ATCC 29213. The median MICs as determined by Etest for

629

control strain ATCC 29213 fell within CLSI acceptable ranges (horizontal red lines) for MHA

630

(black circles) and sup-MHA (grey squares) for all the antibiotics tested. Errors bars depict the

631

lower and upper range-limits detected for 4 replicates.

632

29

LB BA Te P kn ov MH a- A M H su A* p TH -M M HA -M H BD A* su p- -BH BD I H - BH ar Te dy I kn BH ov I aB SC HI su FM pS D M CF an M ni D to lS al t TS su A pTS A

Percent of SCVs (N=10) 100%

90% Large Colonies

80%

70% Small Colonies

60% No Growth

50%

40%

30%

20%

10%

0%

Medium

A

TSA

sup-TSA

sup-MHA

Medium

B

50% 45% 40%

Aerobic

Microaerobic

Percent Reversion

35% 30% 25% 20% 15% 10% 5% 0% BAP

BD-BHI

sup-BD-BHI

Medium

sup-MHA

THM-MHA*

40

40

35

35

30

30

25

25

20

20

15

15

10

10

5

5

00

0

Antibiotic

A

C

L

P

X

A

S

C

x

C

N

T

N

A

N

F

I

D

X

C

Z

le in in m cin ate line cin in m ne in in te in in in in ol in id lin id tin y y an yc zol myc xac zol ene ycli xac ris amp myc xazo kac lana oxit nic myc Ac acil acta c i a ho lo ine rop oc iflo lfop if u om vul gecy com om efa da ic Ox zob ef phe sfo r f m r r v d R b t o C n n e C lin A la L a x th la Ti a si v m Fo ith To ame M Mi Mo n/D Va /T Teli n/C /C C ra Fu Le Az in i in lo i lf l t l l l h l s u i i il C -S ici ac pr rc er ox im ca nu i r p i i m T P op A Qu eth m i Tr

Zone Diameter (mm)

Zone Diameter (mm)

Average Zone Size for ATCC 25923

16.0 16

8.0 8

88.0

4.0 4

44.0

2.0 2

2.0 2

1.0 1

1.0 1

0.5 0.5 0.25 .25

0.5 0. 0. 0.25 I

T

e n l in in m ne lid tin ate em ol hi cin yc yc ico cta cli zo ot an en az ris n xa a m l l m e y p x p e a a o b u i o n l o oc br lfo ph zo ph nc Li of Im lav in eth in m To Ta ev Da Va Ce a / /C m M / r Cl L n n a o f in li li ul ist Ch cil cil -S pr ra xi u e m o p ri in Pi Am op Qu eth im Tr Antibiotic in yc m da

MIC mg/L

MIC mg/L

16 16.0

Table 1. Susceptibility profiles of NC and SCV clinical isolates determined by disk diffusion.

a

These isolates displayed identical percent susceptibilities on both sup-MHA (specified in table as sMHA) and BHI;

therefore percent susceptibilities are shown as a single column representing both media. Note. Discrepancies between media for the same group of isolates is noted by bold boxes and discussed in the text. ND-SCV refers to an SCV with a Non-Defined auxotrophy. NT in the rows corresponding to the β-lactams indicates that these drugs were Not Tested for the corresponding MRSA isolates.

Table 2. Susceptibility profiles of NC and SCV clinical isolates determined by Etest. % Susceptible (Etest) NC MSSA (n=17)

TD-SCV MSSA (n=15)

94.1

80.0

100.0

100.0

57.1

33.3

50.0

Trimethoprim/Sulfamethoxazole 100.0

0.0

50.0

100.0

100.0

0.0

100.0

Antibiotic Clindamycin

HD-SCV MD-SCV NC TD-SCV HD-SCV MSSA MSSA MRSA MRSA MRSA (n=2) (n=1) (n=7) (n=6) (n=2)

Levofloxacin

88.2

80.0

100.0

100.0

14.3

0.0

50.0

Chloramphenicol

100.0

100.0

100.0

100.0

100.0

100.0

100.0

Linezolid

100.0

100.0

100.0

100.0

100.0

100.0

100.0

Vancomycin

100.0

100.0

100.0

100.0

100.0

100.0

100.0

Minocycline

100.0

100.0

100.0

100.0

100.0

100.0

100.0

Quinupristin/Dalfopristin

100.0

100.0

100.0

100.0

100.0

100.0

100.0

Cephalothin

100.0

100.0

100.0

100.0

NT

NT

NT

Amoxicillin/Calvulanate

100.0

100.0

100.0

100.0

NT

NT

NT

Imipenem

100.0

100.0

100.0

100.0

NT

NT

NT

Piperacillin Tazobactam

100.0

100.0

100.0

100.0

NT

NT

NT

Tobramycin

64.7

73.3

50.0

100.0

0.0

0.0

0.0

Note. NT in the rows corresponding to the β-lactams indicates that these drugs were Not Tested for the corresponding MRSA isolates. All testing was completed on THM-MHA manufactured by Teknova.

Table 3. Comparison of SCV and NC susceptibility profiles. Percent of SCVs with same susceptibility profile as paired NC isolate Antibiotic

TD-MSSAa (n=48)

TD-MRSAa (n=31)

HD/MD-MSSA (n=3)

HD-MRSA (n=2)

Cefoxitin

100

97

100

100

Levofloxacin

100

97

100

100

Moxifloxacin

100

97

100

100

96

100

100 (sup-MHA)

33 (BHI)

100

Amikacinb

90

97

0 (sup-MHA)

33 (BHI)

100

TrimethoprimSulfamethoxazole

0

0

67

100

Minocycline

100

100

100

100

Chloramphenicol

100

100

100

100

Azithromycin

88

100

100

50

Clindamycin

92

84

100

100

Quinupristin/Dalfopristin

100

100

100

100

Telithromycin

89

84

100

100

Linezolid

100

100

100

100

Rifampin

100

90

100

100

Tigecycline

100

100

100

100

Fusidic Acid

100

96

100

100

Tobramycin

a

b

The isolate collection was primarily composed of thymidine-dependent (TD) SCVs. Hemin- and menadione-

dependent (HD/MD) SCVs made up the minority of clinical isolates. b

Exposure to these drugs has been linked to the emergence of SCVs (NN and AN with HD/MD-SCVs and SXT

with TD-SCVs), and therefore, discrepancies between NC and SCV susceptibilities are not surprising for these drugs. Note. Shading highlights any value(s) less than 100% similarity between SCV and NC susceptibilities to an antibiotic.

Table 4: Relationships between antibiotic exposure and in vitro susceptibilities All antibiotics to which subjects culture-positive for SCVs during the study reported exposure within 90 days prior to cultures are included. Antibiotic tested

Amikacin

Azithromycin

Clindamycin

Levofloxacin Rifampin Tobramycin TrimethoprimSulfamethoxazole a

Susceptibility findings for 71 isolates in the analysis Antibiotic exposure within 90 days prior to culture datea Logistic model resultsb dataset 29 susceptible Amikacin – No instances of amikacin exposure Not applicable 42 not susceptible OR=19.9 Tobramycin – 27 isolates with prior tobramycin exposure; 25 isolates were not susceptible (95% CI 3.5, 111.9; p=0.001) 13 susceptible Clarithromycin – No instances of clarithromycin exposure Not applicable Azithromycin – 8 isolates (from 5 subjects) with prior azithromycin 58 not susceptible Unable to fit exposure; all 8 isolates were not susceptible Azithromycin or Clindamycin – 11 isolates (from 6 subjects) with prior azithromycin or clindamycin exposure; all 11 isolates were Unable to fit not susceptible Clindamycin – 3 isolates (from 1 subject) with prior clindamycin 49 susceptible Unable to fit exposure; all 3 isolates were not susceptible 22 not susceptible Azithromycin or Clindamycin – 11 isolates with prior azithromycin OR=5.2 or clindamycin exposure; 7 isolates were not susceptible (95% CI 0.8, 34.0; p=0.08) 38 susceptible OR=0.14 Quinolones (ciprofloxacin, levofloxacin, or moxifloxacin)– 8 isolates with prior quinolone exposure; 7 isolates were susceptible (95% CI 0.02, 1.1; p=0.06) 33 not susceptible 41 susceptible Rifampin – 12 isolates (from 4 subjects) with prior rifampin Unable to fit exposure; all 12 isolates were not susceptible 30 not susceptible 28 susceptible OR=18.1 Tobramycin – 27 isolates with prior tobramycin exposure; 25 isolates were not susceptible 43 not susceptible (95% CI 3.2, 100.9; p=0.001) 4 susceptible OR=2.5 Trimethoprim/Sulfamethoxazole (SXT) – 50 isolates with prior SXT exposure; 48 isolates were not susceptible 67 not susceptible (95% CI 0.3, 20.6; p=0.39)

Previous antibiotic exposure was determined by evaluating antibiotic start and end dates in relation to the culture date. Antibiotic exposure was defined as

present if the antibiotic was used within the 90 days prior to the culture date. b Logistic regression models accounted for multiple isolates per subject and used robust variance estimates. Logistic models were unable to be fit when presence of prior antibiotic exposure perfectly predicted isolate susceptibility category.