Proteome phenotyping of relA mutants in Enterococcus faecalis V583 1

2

Xue Yan1, Aurélie Budin-Verneuil

1, Yanick Auffray

1 and Vianney Pichereau

1,2 3

4

5

1Unité de Recherche Risques Microbiens (U2RM), Equipe Stress Virulence, Université de 6

Caen Basse-Normandie, 14032 Caen, France. 7

2Laboratoire des Sciences de l'Environnement Marin (LEMAR). UMR 6539 8

CNRS/UBO/IRD/Ifremer. Institut Universitaire Européen de la Mer. Université de Bretagne 9

Occidentale, UEB. Technopole Brest Iroise. 29280 Plouzané, France. 10

11

12

Correspondence: 13

14

Vianney PICHEREAU 15

16

Laboratoire des Sciences de l'Environnement Marin (LEMAR) 17

UMR 6539 CNRS/UBO/IRD/Ifremer 18

Institut Universitaire Européen de la Mer 19

Technopole Brest Iroise 29280 Plouzané – France 20

Tel : 02 98 49 86 12 21

Email : vianney.pichereau@univ-brest.fr 22

23

24

ABSTRACT 25

26

The (p)ppGpp synthetase RelA contributes to stress adaptation and virulence in Enterococcus 27

faecalis V583. A 2-DE based proteomic analysis of two relA mutants, ie relA carrying a 28

complete deletion of the relA gene, and relAsp that is deleted from only its 3’ extremity, 29

showed that 31 proteins were deregulated in one or both of these mutants. Mass spectrometry 30

identification of these proteins showed that 10 are related to translation, including 5 ribosomal 31

proteins, 3 proteins involved in translation elongation and 2 proteins in tRNA synthesis; 14 32

proteins are involved in diverse metabolisms and biosynthesis (8 in sugar and energy 33

metabolisms; 2 in fatty acid biosynthesis; 2 in amino acid biosynthesis and 2 in nucleotide 34

metabolism). 5 proteins were relevant to the adaptation to different environmental stresses, ie 35

SodA and a Dps family protein, two cold shock domain proteins, and EF1744, a general stress 36

protein which plays an important role in the response to ethanol stress. The potential role of 37

these proteins in the development of stress phenotypes associated with these mutations are 38

discussed. 39

40

41

Keywords 42

relA mutants – Proteome – Guanine nucleotides– general stress protein 43

44

RESUME 45

La (p)ppGpp synthétase (RelA) contribue à l'adaptation aux contraintes environnementales et 46

la virulence chez Enterococcus faecalis V583. Nous avons analysé par électrophorèse des 47

protéines les modifications du protéome chez deux mutants de délétion du gène relA, l'un 48

(relA) portant une délétion complète du gène relA, l'autre portant une délétion partielle de 49

l'extrémité 3' de ce gène (relAsp). Nous avons identifié par spectrométrie de masse 31 50

protéines dérégulées dans l'un ou l'autre de ces mutants. Dix d'entre elles sont liées au 51

processus de traduction; 14 sont impliquées dans divers métabolismes (8 dans le métabolisme 52

énergétique, 2 dans la biosynthèse des acides gras, 2 dans celle des acides aminés et 2 dans le 53

métabolisme des nucleotides). Cinq protéines apparaissent liées à l'adaptation aux stress 54

environnementaux, dont SodA et une protéine de la famille Dps, deux protéines apparentées 55

aux protéines de choc froid, et EF1744, une protéine générale de stress qui joue un rôle 56

important dans la réponse à l'éthanol. Les rôles potentiels de ces protéines dans le 57

développement des phénotypes associés à ces mutations du gène relA sont discutés. 58

59

60

INTRODUCTION 61

Enterococcus faecalis is a Gram-positive commensal bacterium inhabiting the gastrointestinal 62

tracts of mammals. This bacterium belongs to the Lactic Acid Bacteria, and is the main 63

constituents of some probiotic food supplements (Mercenier et al. 2003, Giraffa 2003, Hugas 64

et al. 2003). It is able to survive to 6.5% NaCl, pH between 4.0 and 9.6, and growth 65

temperatures from 10 to 45°C (Gilmore 2002). In addition, E. faecalis can cause some 66

infections in humans (Schaberg et al. 1991, Noskin et al. 1991), especially in the nosocomial 67

(hospital) environment, where its naturally high level of antibiotic resistance contributes to its 68

pathogenicity. 69

In natural ecosystems, bacteria are subjected to a variety of stress and starvation conditions 70

and have developed highly sophisticated molecular networks to cope with these situations 71

(Pichereau et al. 2000). One crucial component of this adaptational network is the stringent 72

response (Mechold et al. 1996, Chatterji and Ojha 2001) that coordinates the bacterial 73

physiology with the current growth conditions. The stringent response is dependent on the 74

synthesis of the alarmones guanosine 3'-diphosphate -5'-triphosphate (pppGpp) and guanosine 75

3',5'-bispyrophosphate (ppGpp) [collectively referred to as (p)ppGpp], which are synthesized 76

by phosphorylation of GDP and GTP, respectively, using ATP as the phosphate donor. 77

Although the stringent response was first described as an adaptation to amino acid starvation, 78

(p)ppGpp were shown to be involved in several stress responses such as thermotolerance 79

(Yang and Ishiguro 2003), adaptation to oxidative stress (Mostertz et al. 2004), osmotic stress 80

(Okada et al. 2002), antibiotics (Wu et al. 2010) and the virulence (Dalebroux et al. 2010). 81

In Gram-positive bacteria, the (p)ppGpp synthetase protein (RelA) was shown to carry both 82

(p)ppGpp degradation and synthesis activities (Mechold et al. 1996, Wendrich and Marahiel 83

1997, Mittenhuber 2001). Two other (p)ppGpp synthetases were also characterized in 84

Streptococcus mutans and Bacillus subtilis, designated RelP and RelQ, that shared very 85

limited homologies with the RelA/SpoT enzymes (Lemos et al. 2007, Nanamiya et al. 2008). 86

The construction and characterization of (p)ppGpp synthetases mutants, ie relA and relQ, 87

allowed determining that RelA is the main system leading to (p)ppGpp production 88

(Abranches et al. 2009, Yan et al. 2009). In this previous study (Yan et al. 2009), we also 89

constructed a mutant (∆relAsp) in which a small part of the relA gene (about 0.7 kbp) 90

encoding the carboxy-terminal domain of the RelA protein was deleted. Both relA deletion 91

mutants displayed physiological phenotypes to sublethal or lethal stresses, as compared to the 92

wild type strain (Yan et al. 2009). For example, both mutants displayed increased resistance 93

to bile salts, ethanol, and acid challenges, but displayed contrasting phenotypes towards 94

oxidative stress (Yan et al. 2009). By using similar mutants, RelA was also recently shown to 95

be implied in the formation of biofilm (Chávez de Paz et al. 2012). 96

Two-dimensional protein electrophoresis (2-DE) is an efficient way to determine proteomic 97

signatures and proteome phenotypes of a given strain (Giard et al. 2004, Budin-Verneuil et al. 98

2007). In this paper, as an attempt to better understand the origin of the phenotypes of the 99

relA mutants, we performed a 2-DE based proteomic analysis of the two mutants. 100

101

102

MATERIALS AND METHODS 103

104

Bacterial strains and growth conditions 105

All strains used in this study were derivatives of the E. faecalis V583 strain. The relA 106

(ppGpp0), relAsp, and relAcomp strains were previously obtained using a double cross-107

over procedure described in Yan et al. (2009). Cells were cultivated with shaking in M17 108

medium supplemented with 0.5% (vol/vol) glucose (GM17) at 37°C. For growth experiments 109

at low temperature, exponentially growing cultures (OD600 = 1) were diluted to a final OD600 110

of 0.1 in fresh GM17 medium, and incubated at the indicated temperature. The growth was 111

monitored spectrophotometrically by measuring the optical density at 600 nm. 112

113

Two-dimensional protein gel electrophoresis 114

10 ml-cultures of E. faecalis V583-,relA- and relAsp- strains were maintained overnight in 115

exponential phase in GM17. They were diluted to a final OD600 of 0.1 in 100 ml of fresh 116

medium and incubated for 24 hours. For proteins extraction, cultures were washed with Tris-117

HCl 50 mM (pH 8), and lysed in a Cell-Disrupter (Constant System Ltd, UK) at 2.7 kbar. 118

After centrifugation (12000 g, 10 min), supernatants were precipitated by adding 4 volumes 119

of acetone, and 500 µg of proteins were solubilised in rehydration buffer as described in 120

Budin-Verneuil et al. (2005). This solution was used to rehydrate Immobiline Drystrips pH 4-121

7 (GE Healthcare, France). Migration parameters for the first dimension were essentially as 122

described in Budin-Verneuil et al. (2005). 123

After the first dimension, strips were incubated 15 min in an equilibration solution (2% SDS, 124

6M Urea, 50 mM Tris-HCl pH 8.8, 30% glycerol (v/v), BBP) containing DTT (1%), and then 125

15 min in the same solution containing iodoacetamide (260 mM). The second dimension was 126

performed as described in Budin-Verneuil et al. (2005). Proteins were revealed by incubating 127

gels in a 0.1% Coomassie blue (Sigma, France) staining solution overnight, followed by 128

washing in a methanol - acetic acid - water (30/7/63 : v/v/v) solution. Gels were scanned, and 129

images analysis was performed by using the Prodigy SameSpot software (Nonlinear 130

Dynamics, UK) as described in Galland et al. (2013). 131

132

133

Proteins identification 134

Protein spots (1 mm2) were excised from Coomassie stained gels and incubated in 50 mM 135

ammonium bicarbonate (BICAM) (Sigma). Spots were then dehydrated by incubations in 136

acetonitrile (ACN, Sigma), followed by rehydration using 0.4 µg trypsin (Promega, USA) in 137

50 mM BICAM. Digestion was carried out overnight at 37°C. The peptides were recovered 138

by successive incubations of gel pieces in BICAM, ACN, and 0.2% formic acid as described 139

in Benachour et al. (2009). 140

To analyse peptides, we used an electrospray ion trap mass spectrometer (LCQ DECA XP, 141

Thermo Finnigan, France) coupled on line with an HPLC (Surveyor LC, Thermo Finnigan). 142

Peptides were separated and analysed mainly as described in Benachour et al. (2009). 143

Recorded MS/MS spectra were compared to theoretical fragmentations of a trypsinolysed 144

database deduced from the E. faecalis V583 genome, using the Sequest software (version 2, 145

Thermo Finnigan). 146

147

RESULTS AND DISCUSSION 148

Proteins deregulated in the relA mutants 149

We previously obtained and characterized two mutants carrying either a total deletion of the 150

relA gene encoding the (p)ppGpp synthetase, (relA mutant), or a partial deletion leading to 151

the synthesis of a C-terminus truncated protein (relAsp mutant), that displayed differential 152

phenotypes towards various stressing conditions, and in virulence assays (Yan et al. 2009). 153

Indeed, both mutants displayed increased resistance to bile salts, ethanol, and acid challenges, 154

but displayed contrasting phenotypes towards H2O2 and in virulence assays (Yan et al. 2009). 155

To determine the impact of these mutations at a proteomic level, proteins were extracted from 156

24 h- cultures (late exponential phase) grown in GM17 with shaking, a culture condition that 157

did not have any effect on the growth of mutants (Yan et al. 2009). 158

After 2-DE separation of proteins, the electrophoregrams (Fig. 1) were analyzed using the 159

Prodigy SameSpot software. In all, we selected 31 proteins spots displaying changes above 160

1.5 (increased protein abundance) or below 0.75 (decreased protein abundance) in at least one 161

mutant (Table 1). The corresponding protein spots were excised and identified by mass 162

spectrometry. 163

The modifications of protein profiles were very similar in both mutants, as we did not observe 164

proteins displaying very different accumulation profiles between the relA- and the relAsp- 165

mutants. Among the 31 proteins, 14 were accumulated, but only 9 with fold factors above 2 166

(Table 1). Seventeen were down-accumulated (only 10 with factors below 0.5). Two protein 167

spots were identified as the same protein, ie RplJ (Ef2716), while displaying both different 168

molecular weights and isoelectric points (pI). 169

We performed network analysis on the obtained dataset, using the STRING v9.1 program 170

(Franceschini et al. 2013). This revealed linkage between 24 among the 30 unique proteins 171

evidenced in our study (Fig. 2). The whole image showed two major clusters, corresponding 172

to the proteins related to translation and to metabolic processes. This was clearly confirmed 173

by enrichment analyses, that gave p-values below 10-2

for translation (GO biological 174

processes) and glycolysis/gluconeogenesis metabolic pathways (KEGG pathway). All these 175

proteins were classified according to their potential role and accumulation profiles. We 176

discuss later the potential impact of the observed modifications on the physiology of the 177

mutants. 178

179

1. Rel mutations impact the translation apparatus 180

The stringent response is known to have a major impact on the translation process. As 181

examples, downregulation of some genes encoding ribosomal proteins was observed in rel-182

deficient strains of Corynebacterium glutamicum (Brockmann-Gretza and Kalinowski 2006) 183

and in wild type and relA mutant strains of Escherichia coli (Durfee et al. 2008). Ten 184

proteins which intracellular amounts were modified in relA and relAsp mutants were 185

identified as proteins related to translation (Table 1). Five of them corresponded to ribosomal 186

proteins, three to translation elongation factors and two were related to tRNA synthesis. All of 187

these proteins displayed similar accumulation patterns in both mutants. 188

189

2. Metabolism and biosynthesis 190

Our proteomic approach strongly suggested a deep metabolic remodelling in the rel mutants. 191

As evidenced by the STRING analysis (Franceschini et al. 2013), the most widely modified 192

metabolism was glycolysis. Indeed, the abundances of several glycolytic enzymes revealed 193

modified. For example, Pgi (phospho-glucose isomerase), Gap-1, Gap-2 appeared down-194

accumulated in both mutants. Concomitantly, we observed a decrease in the intracellular 195

amounts of Ldh-1 and AdhE, two enzymes involved in the NADH2-dependent reduction of 196

pyruvate to lactate and ethanol, respectively. These results strongly suggested that the 197

glycolytic activity, and lactic acid fermentation, should be reduced in the mutants. In a recent 198

transcriptomic analysis of a (p)ppGpp0 E. faecalis mutant derived from the OG1RF strain, 199

Gaca et al. (2013) showed that this mutant displayed a decreased lactate production, and in 200

fact turned its metabolism towards heterolactic fermentation, mainly producing acetoin. The 201

only protein highly accumulated in this category corresponded to a IIAB component of a 202

mannose specific PTS transport system, which was one of the two largely accumulated 203

protein in both mutants, with an accumulation factor above three. 204

The proteomes of the mutants also revealed modifications in other metabolisms, such as fatty 205

acids- (FabI and FabF-1 were reduced in the two mutants) and amino-acids- biosynthesis 206

(with the reduction of GlyA in both mutants). It should be noted that interactions between the 207

(p)ppGpp production and lipids biosynthesis were already shown in other bacteria. For 208

example, the acyl carrier protein ACP, an essential component in fatty acid metabolism, 209

physically interacts with SpoT in E. coli (Battesti and Bouveret 2006). This interaction is 210

required for the accumulation of (p)ppGpp in response to fatty acid starvation by shifting the 211

balance between the synthetic and hydrolytic activities of SpoT. 212

Nucleotides metabolism was also affected, as we observed the negative deregulation of ThyA, 213

and the large accumulation (increased above 2.7 fold in both mutants) of GuaB, an IMP 214

dehydrogenase. This suggested the deregulation of purine metabolism, which was quite 215

expected considering the chemical structure of (p)ppGpp. 216

217

3. Stress responses 218

We previously studied the responses of the relA- and relAsp- mutants to different stress 219

conditions and in an animal virulence model (Yan et al. 2009), and observed both convergent 220

and different phenotypes for the relA and relAsp mutants. As discussed in the next sections, 221

this proteomic approach gave more insights to the molecular mechanisms implied in the 222

development of these phenotypes. 223

224

Oxidative stress 225

The Dps family of proteins is a group of bacterial stress-inducible polypeptides that bind 226

DNA, and likely confer resistance to peroxide damages during periods of oxidative stress and 227

long term nutrient limitation (Almirón et al. 1992, Ishikawa et al. 2003). The SodA protein 228

provides a defence against oxidative stress by dismuting superoxide radicals in oxygen and 229

hydrogen peroxide (Poyart et al. 2000). Interestingly, upregulation of the sodA gene in relA 230

null mutants in E. coli has been previously described (Argaman et al. 2012). Here we showed 231

that the amounts of SodA and of Dps were increased in both relA- and in the relAsp- 232

mutant. However, the induction factors of both SodA and Dps were higher in the relAsp 233

mutant compared to the relA one. 234

We previously showed that the relA and relAsp mutants displayed very contrasting 235

phenotypes towards oxidative stress (Yan et al. 2009). Indeed, the relA mutant was highly 236

sensitive to oxidative stress, while the relAsp one was highly resistant to this stress. This 237

proteome phenotyping experimental strategy aimed at better understanding the different stress 238

phenotypes of the mutants. With this end in view, the differential expression of SodA and Dps 239

could be related to the contrasted phenotypes of the mutants. However, one could expect that 240

such a different sensitivity should be correlated to proteins that would be both down- and up- 241

accumulated in the two strains. In fact, only one protein spot displayed such an accumulation 242

profile, that was identified as Ef2923, a small (70 amino acids) conserved hypothetical protein. 243

Determining what is the actual role of this protein should be of particular importance in this 244

context. 245

246

Cold stress 247

Both relA mutants displayed increased intracellular amounts of the cold-shock domain family 248

proteins Ef1367 and Ef2925. It should be noted that shifting an E. coli relA-spoT mutant, 249

unable to synthesize (p)ppGpp, from 24 to 10 °C resulted in a greater induction of cold shock 250

proteins than in the wild type strain (Jones et al. 1992), suggesting interactions between 251

(p)ppGpp accumulation and the resistance to cold stress. Therefore, we tested the ability of 252

our mutants to proliferate at low temperature. The relAsp mutant displayed a reduced growth 253

at 14°C, while the relA mutant was even more sensitive to this condition than the wild type 254

V583 strain (Fig. 3). Thus, although these mutants accumulated high amounts of Ef1367 and 255

Ef2925 cold shock domain proteins, they were sensitive to cold stress. 256

257

Ethanol stress 258

We previously showed that both relA mutants were more resistant to ethanol than the wild 259

type V583 strain (Yan et al. 2009). Molecular bases of bacterial resistance/sensitivity to 260

ethanol is not clearly understood. In a recent proteomic study on the lactic acid bacterium 261

Lactobacillus plantarum ST4, Lee et al. (2012) showed that among 28 deregulated proteins, 262

ethanol stress induced the accumulation of the translation elongation factors FusA and Tuf, 263

while lactate dehydrogenase appeared down-regulated. Thus, these proteins displayed the 264

same accumulation pattern as in our mutants. In addition, we previously showed that the 265

general stress protein EF1744 responded a variety of environmental stresses, including 266

ethanol (Giard et al. 2004). In our study, this protein was among the two most deregulated, 267

with accumulation factors higher than three. Recently, we constructed a null mutant of ef1744, 268

and analysed its stress phenotypes (manuscript in preparation). This mutant was highly 269

sensitive (32 fold) to 22% ethanol, as compared to the wild type. Therefore, we infer that the 270

accumulation of this protein in both relA mutants should be related to their ethanol stress 271

phenotypes. 272

273

CONCLUDING REMARKS 274

275

In this study, we examined the protein content of two relA mutants, ie the relA- and the 276

relAsp- mutant, which display different (p)ppGpp accumulation profiles and different stress 277

and virulence phenotypes (Yan et al. 2009). Our proteomic data suggested that both mutants 278

modulated the expression of metabolic enzymes, and of stress proteins. A number of the 279

modifications we observed could be correlated to the stress phenotypes we described 280

previously for these mutants (Yan et al. 2009). Indeed, we observed the induction of oxidative 281

stress proteins (ie, SodA and Dps) higher in the relAsp mutant, and of proteins related to 282

ethanol resistance (Ef1744) in both mutants. We also observed the accumulation of proteins 283

related to cold stress, and showed that both mutants displayed sensitivity to cold stress. In all, 284

this study gave the first elements of the (p)ppGpp modulon in E. faecalis, and allowed (i) 285

better understanding some stress phenotypes previously observed in the relA mutants, and (ii) 286

infering a role of (p)ppGpp on cold stress adaptation of E. faecalis. 287

288

289

290

ACKNOWLEDGEMENTS 291

X.Y. was the recipient of a doctoral fellowship from INRA (MICA Department) and the 292

Conseil Régional de Basse-Normandie, France. The authors greatly thank Marie-Jeanne Pigny 293

for expert technical assistance, and Nicolas Verneuil for critical reading of the manuscript. 294

295

REFERENCES 296

Abranches, J., Martinez, A.R., Kajfasz, J.K., Chávez, V., Garsin, D.A., and Lemos, J.A. 2009. 297

The molecular alarmone (p)ppGpp mediates stress responses, vancomycin tolerance, and 298

virulence in Enterococcus faecalis. J. Bacteriol. 191: 2248–2256. 299

Almirón, M., Link, A.J., Furlong, D., and Kolter, R. 1992. A novel DNA-binding protein with 300

regulatory and protective roles in starved Escherichia coli. Genes Dev. 6: 2646–2654. 301

Argaman, L., Elgrably-Weiss, M., Hershko, T., Vogel, J., and Altuvia, S. 2012. RelA protein 302

stimulates the activity of RyhB small RNA by acting on RNA-binding protein Hfq. Proc. 303

Natl. Acad. Sci. U. S. A. 109: 4621–4626. 304

Battesti, A., and Bouveret, E. 2006. Acyl carrier protein/SpoT interaction, the switch linking 305

SpoT-dependent stress response to fatty acid metabolism. Mol. Microbiol. 62: 1048–306

1063. 307

Benachour, A., Morin, T., Hébert, L., Budin-Verneuil, A., Le Jeune, A., Auffray, Y., and 308

Pichereau, V. 2009. Identification of secreted and surface proteins from Enterococcus 309

faecalis. Can. J. Microbiol. 55: 967–974. 310

Brockmann-Gretza, O., and Kalinowski, J. 2006. Global gene expression during stringent 311

response in Corynebacterium glutamicum in presence and absence of the rel gene 312

encoding (p)ppGpp synthase. BMC Genomics 7: 230. 313

Budin-Verneuil, A., Pichereau, V., Auffray, Y., Ehrlich, D., and Maguin, E. 2007. Proteome 314

phenotyping of acid stress-resistant mutants of Lactococcus lactis MG1363. Proteomics 315

7: 2038–2046. 316

Budin-Verneuil, A., Pichereau, V., Auffray, Y., Ehrlich, D.S., and Maguin, E. 2005. 317

Proteomic characterization of the acid tolerance response in Lactococcus lactis MG1363. 318

Proteomics 5: 4794–4807. 319

Chatterji, D., and Ojha, A.K. 2001. Revisiting the stringent response, ppGpp and starvation 320

signaling. Curr. Opin. Microbiol. 4: 160–165. 321

Chávez de Paz, L.E., Lemos, J.A., Wickström, C., and Sedgley, C.M. 2012. Role of (p)ppGpp 322

in biofilm formation by Enterococcus faecalis. Appl. Environ. Microbiol. 78: 1627–1630. 323

doi: 10.1128/AEM.07036-11. 324

Dalebroux, Z.D., Svensson, S.L., Gaynor, E.C., and Swanson, M.S. 2010. ppGpp conjures 325

bacterial virulence. Microbiol. Mol. Biol. Rev. 74: 171–199. 326

Durfee, T., Hansen, A.-M., Zhi, H., Blattner, F.R., and Jin, D.J. 2008. Transcription profiling 327

of the stringent response in Escherichia coli. J. Bacteriol. 190: 1084–1096. 328

Franceschini, A., Szklarczyk, D., Frankild, S., Kuhn, M., Simonovic, M., Roth, A., Lin, J., 329

Minguez, P., Bork, P., von Mering, C., and Jensen, L.J. 2013. STRING v9.1: protein-330

protein interaction networks, with increased coverage and integration. Nucleic Acids Res. 331

41: D808–815. 332

Gaca, A.O., Kajfasz, J.K., Miller, J.H., Liu, K., Wang, J.D., Abranches, J., and Lemos, J.A. 333

2013. Basal levels of (p)ppGpp in Enterococcus faecalis: the magic beyond the stringent 334

response. MBio 4: e00646–13. 335

Galland, C., Dupuy, C., Capitaine, C., Auffret, M., Quiniou, L., Laroche, J., and Pichereau, V. 336

2013. Comparisons of liver proteomes in the European flounder Platichthys flesus from 337

three contrasted estuaries. J. Sea Res. 75: 135–141. 338

Giard, J.-C., Auffray, Y., Benachour, A., Hartke, A., Laplace, J.-M., Rince, A., Verneuil, N., 339

and Pichereau, V. 2004. Proteomics Analysis: A Powerful Tool to Identify Proteome 340

Phenotype and Proteome Signature in Enterococcus faecalis. Curr. Proteomics 1: 273–341

282. 342

Gilmore, M.S. 2002. The Enterococci: Pathogenesis, Molecular Biology, and Antibiotic 343

Resistance. Washington, DC: American Society for Microbiology Press. 344

Giraffa, G. 2003. Functionality of enterococci in dairy products. Int. J. Food Microbiol. 88: 345

215–222. 346

Hugas, M., Garriga, M., and Aymerich, M.T. 2003. Functionality of enterococci in meat 347

products. Int. J. Food Microbiol. 88: 223–233. 348

Ishikawa, T., Mizunoe, Y., Kawabata, S., Takade, A., Harada, M., Wai, S.N., and Yoshida, S. 349

2003. The iron-binding protein Dps confers hydrogen peroxide stress resistance to 350

Campylobacter jejuni. J. Bacteriol. 185: 1010–1017. 351

Jones, P.G., Cashel, M., Glaser, G., and Neidhardt, F.C. 1992. Function of a relaxed-like state 352

following temperature downshifts in Escherichia coli. J. Bacteriol. 174: 3903–3914. 353

Lee, S.G., Lee, K.W., Park, T.H., Park, J.Y., Han, N.S., and Kim, J.H. 2012. Proteomic 354

analysis of proteins increased or reduced by ethanol of Lactobacillus plantarum ST4 355

isolated from Makgeolli, traditional Korean rice wine. J. Microbiol. Biotechnol. 22: 516–356

25. 357

Lemos, J.A., Lin, V.K., Nascimento, M.M., Abranches, J., and Burne, R.A. 2007. Three gene 358

products govern (p)ppGpp production by Streptococcus mutans. Mol. Microbiol. 65: 359

1568–1581. 360

Mechold, U., Cashel, M., Steiner, K., Gentry, D., and Malke, H. 1996. Functional analysis of 361

a relA/spoT gene homolog from Streptococcus equisimilis. J. Bacteriol. 178: 1401–1411. 362

Mercenier, A., Pavan, S., and Pot, B. 2003. Probiotics as biotherapeutic agents: present 363

knowledge and future prospects. Curr. Pharm. Des. 9: 175–191. 364

Mittenhuber, G. 2001. Comparative genomics and evolution of genes encoding bacterial 365

(p)ppGpp synthetases/hydrolases (the Rel, RelA and SpoT proteins). J. Mol. Microbiol. 366

Biotechnol. 3: 585–600. 367

Mostertz, J., Scharf, C., Hecker, M., and Homuth, G. 2004. Transcriptome and proteome 368

analysis of Bacillus subtilis gene expression in response to superoxide and peroxide 369

stress. Microbiology 150: 497–512. 370

Nanamiya, H., Kasai, K., Nozawa, A., Yun, C.-S., Narisawa, T., Murakami, K., Natori, Y., 371

Kawamura, F., and Tozawa, Y. 2008. Identification and functional analysis of novel 372

(p)ppGpp synthetase genes in Bacillus subtilis. Mol. Microbiol. 67: 291–304. doi: 373

10.1111/j.1365-2958.2007.06018.x. 374

Noskin, G.A., Till, M., Patterson, B.K., Clarke, J.T., and Warren, J.R. 1991. High-level 375

gentamicin resistance in Enterococcus faecalis bacteremia. J. Infect. Dis. 164: 1212–376

1215. 377

Okada, Y., Makino, S., Tobe, T., Okada, N., and Yamazaki, S. 2002. Cloning of rel from 378

Listeria monocytogenes as an osmotolerance involvement gene. Appl. Environ. 379

Microbiol. 68: 1541–1547. 380

Pichereau, V., Hartke, A., and Auffray, Y. 2000. Starvation and osmotic stress induced 381

multiresistances. Influence of extracellular compounds. Int. J. Food Microbiol. 55: 19–25. 382

Poyart, C., Quesnes, G., and Trieu-Cuot, P. 2000. Sequencing the gene encoding manganese-383

dependent superoxide dismutase for rapid species identification of enterococci. J. Clin. 384

Microbiol. 38: 415–418. 385

Schaberg, D.R., Culver, D.H., and Gaynes, R.P. 1991. Major trends in the microbial etiology 386

of nosocomial infection. Am. J. Med. 91: 72S–75S. 387

Wendrich, T.M., and Marahiel, M.A. 1997. Cloning and characterization of a relA/spoT 388

homologue from Bacillus subtilis. Mol. Microbiol. 26: 65–79. 389

Wu, J., Long, Q., and Xie, J. 2010. (p)ppGpp and drug resistance. J. Cell. Physiol. 224: 300–390

304. 391

Yan, X., Zhao, C., Budin-Verneuil, A., Hartke, A., Rincé, A., Gilmore, M.S., Auffray, Y., and 392

Pichereau, V. 2009. The (p)ppGpp synthetase RelA contributes to stress adaptation and 393

virulence in Enterococcus faecalis V583. Microbiology 155: 3226–3237. 394

Yang, X., and Ishiguro, E.E. 2003. Temperature-sensitive growth and decreased 395

thermotolerance associated with relA mutations in Escherichia coli. J. Bacteriol. 185: 396

5765–5771. 397

398

FIGURES LEGENDS 399

400

401

402

Figure 1: Proteome phenotypes of Enterococcus faecalis V583 (A), relA (B) and relAsp (B). Proteins were extracted from WT and relA 403

mutants after 24 hours culture in GM17 medium with shaking at 37°C. Identified proteins are arrowed. 404

405

Figure 2: Network analysis of proteins shown to be deregulated in this study (built using STRING v9.1 (Franceschini et al. 2013). 406

407

Figure 3: Growth at low temperature (14°C). The growth of the V583- (closed circles), relA- (closed squared), relAsp- (closed triangles), 408

and relAcomp- (open squared) strains in GM17 with shaking. 409

410

Table 1. Identification of proteins displaying modified abundances in relA- and relAsp- mutants.

Locus Name of gene and/or function MM (Da) pI IF relA1 IF relAsp1 Coverage

(%)2

Nb of

peptides2 Best Sp2

Translation

ef0007 (RpsF) ribosomal protein S6 11605.2 5.01 1.58 1.62 48.0 6 2087

ef0200 (FusA) translation elongation factor G 76678.7 4.80 1.13 1.49 19.3 14 1625

ef0201 (Tuf) translation elongation factor Tu 43387.9 4.73 2.47 2.88 29.6 21 2522

ef0633 (TyrS-1) tyrosyl-tRNA synthetase 47260.6 5.09 0.65 0.60 23.7 8 2892

ef0725 (GatA) glutamyl-tRNA(Gln) amidotransferase, A subunit 52619.6 4.97 0.57 0.64 19.6 8 1885

ef1171 (RpmE) ribosomal protein L31 10125.1 5.57 0.47 0.30 65.2 8 1526

ef2397 (Tsf) translation elongation factor Ts 32133.8 4.87 0.55 0.46 63.1 24 1882

ef2715 (RplL) ribosomal protein L7-L12 12396.2 4.49 1.41 1.47 26.2 4 1510

ef2716 (RplJ) ribosomal protein L10 17622.4 5.39 0.65 0.42 50.0 13 1912

ef2716 (RplJ) ribosomal protein L10 17622.4 5.39 1.98 2.31 72.3 15 1785

Sugar and energy metabolism

ef0020 PTS system. mannose-specific IIAB component 35523.9 5.11 3.75 3.27 20.6 6 1058

ef0255 (Ldh-1) L-lactate dehydrogenase 35487.1 4.77 0.37 0.42 05.5 2 1586

ef0710 (PtsI) phosphoenol pyruvate protein phosphotransferase enzyme I 63178.1 4.68 0.17 0.15 12.5 5 1297

ef0900 (AdhE) aldehyde-alcohol dehydrogenase 92365.2 5.70 0.21 0.28 11.3 8 2096

ef0949 (Pta) phosphotransacetylase 35558.6 4.97 0.76 0.88 42.5 13 1501

ef1416 (Pgi) glucose-6-phosphate isomerase 49734.2 4.96 0.64 0.67 46.3 13 1916

ef1526 (Gap-1) glyceraldehyde 3-phosphate dehydrogenase 36208.9 4.87 0.31 0.44 05.6 2 1054

ef1964 (Gap-2) glyceraldehyde 3-phosphate dehydrogenase 35771.5 5.03 0.47 0.56 51.1 23 2571

Fatty acid biosynthesis

ef0282 (FabI) enoyl-(acyl-carrier-protein) reductase 26766.7 5.29 0.46 0.51 35.6 11 2476

ef0283 (FabF-1) 3-oxoacyl-(acyl-carrier-protein) synthase II 43196.2 5.11 0.64 0.60 17.5 7 2026

Amino acid biosynthesis

ef2550 (GlyA) serine hydroxymethyltransferase 44538.4 5.47 0.18 0.25 35.5 15 2385

ef3037 (PepA) glutamyl-aminopeptidase 39204.3 5.68 0.59 0.89 34.9 9 1830

Nucleotide metabolism

ef1576 (ThyA) Thymidylate synthase 36344.1 4.97 0.76 0.89 05.1 2 2002

ef3293 (GuaB) inosine-5-monophosphate dehydrogenase 52872.7 5.70 2.99 2.76 13.4 3 1076

Stress

ef0463 (SodA) superoxide dismutase. Mn 22697.4 4.99 1.33 1.58 30.7 9 2097

ef1367 cold-shock domain family protein 7185.9 4.53 1.65 1.40 87.9 6 2163

ef1744 general stress protein. putative 20544.7 4.61 3.56 3.36 35.4 5 1774

ef2925 cold-shock domain family protein 7299.9 4.35 1.47 2.38 80.3 6 1408

ef3233 Dps family protein 17888.3 4.56 1.69 2.19 51.6 16 1734

Unknown function

ef2923 conserved hypothetical protein 8536.5 4.63 2.96 0.80 70.0 6 1601

ef3115 conserved hypothetical protein 12983.8 5.40 1.94 1.64 70.0 7 1503 1 IF for a given spot corresponds to the ratio between the normalized volume measured for the mutant and that measured for the wild type strain.

2 protein coverage is expressed as the percentage of amino acids effectively sequenced by LC-MS/MS. Sp is the score given by the SEQUEST

software for a given sequenced peptide.

Fig. 1

Ef0900

Ef0725

Ef1416Ef0283Ef0633

Ef1526

Ef1964

Ef0255

Ef0949Ef1576

Ef2550

Ef3037

Ef2397

Ef0282

ef0228 Ef0200

Ef0463

Ef2716a

Ef3233

Ef2634

Ef3293

Ef3115

Ef0007

Ef0201

Ef1367

Ef2925

Ef0020

Ef1744

Ef0201

Ef1171

Ef2716b

Ef0710

Ef2715

Ef2923

AEf0900

Ef0725

Ef1416Ef0283Ef0633

Ef1526

Ef1964

Ef0255

Ef0949Ef1576

Ef2550

Ef3037

Ef2397

Ef0282

ef0228 Ef0200

Ef0463

Ef2716a

Ef3233

Ef2634

Ef3293

Ef3115

Ef0007

Ef0201

Ef1367

Ef2925

Ef0020

Ef1744

Ef0201

Ef1171

Ef2716b

Ef0710

Ef2715

Ef2923

B CEf0900

Ef0725

Ef1416Ef0283Ef0633

Ef1526

Ef1964

Ef0255

Ef0949Ef1576

Ef2550

Ef3037

Ef2397

Ef0282

ef0228Ef0200

Ef0463

Ef2716a

Ef3233

Ef2634

Ef3293

Ef3115

Ef0007

Ef0201

Ef1367

Ef2925

Ef0020

Ef1744

Ef0201

Ef1171

Ef2716b

Ef0710

Ef2715

Ef2923

Fig. 2

Fig. 3

V583

relA

relAsp

relAcomp

V583

relA

relAsp

relAcomp

Time (h)

OD

600

0.1

1

10

0 10 20 30 40 50

Time (h)

OD

600

0.1

1

10

0 10 20 30 40 500.1

1

10

0 10 20 30 40 50