Metabolomic and proteomic responses of Staphylococcus aureus to prolonged cold stress

JO U R N A L OF P ROTE O M ICS 1 21 ( 20 1 5 ) 4 4–55

Available online at www.sciencedirect.com

ScienceDirect www.elsevier.com/locate/jprot

Metabolomic and proteomic responses of Staphylococcus aureus to prolonged cold stress Mousa M. Alreshidia , R. Hugh Dunstana,⁎, Margaret M. Macdonalda , Nathan D. Smithb , Johan Gottfriesc , Tim K. Robertsa a

Pathogenic Microbiology Laboratory, Faculty of Science and Information Technology, School of Environmental and Life Sciences, Department of Biology, University Drive, Callaghan, 2308 NSW, Australia b Analytical and Biomolecular Research Facility (ABRF), University of Newcastle, Callaghan, NSW 2308, Australia c Department of Chemistry, Gothenburg University, Sweden

AR TIC LE I N FO

ABS TR ACT

Article history:

The high pathogenicity of Staphylococcus aureus is thought to be due to its extraordinary

Received 13 September 2014

capacity to rapidly adapt to changes in environmental conditions. This study was carried

Accepted 9 March 2015

out to investigate whether the cytoplasmic profiles of metabolites and proteins of S. aureus

Available online 14 March 2015

were altered in response to prolonged exposure to cold stress. Metabolic profiling and proteomics were used to characterise alterations in cytoplasmic proteins and metabolites

Keywords:

in cells from the mid-exponential phase of growth under ideal conditions at 37 °C and

Metabolomics

compared with equivalent cells exposed to prolonged cold stress for 2 weeks at 4 °C.

Proteomics

Principle component analysis (PCA) of the metabolomic and proteomic data indicated that,

Phenomics

at the mid-exponential phase of growth, prolonged cold stress conditions generated cells

Citric acid

with different metabolite and protein profiles compared with those grown at 37 °C. Nine

Ribosomal proteins

ribosomal proteins and citric acid were substantially elevated in the cytoplasmic fractions

Cold stress

from the cells adapted to cold-stress but most amino acids showed a reduction in their concentration in cold-stressed samples. The data provided strong evidence supporting the hypothesis that specific changes in metabolic homeostasis and protein composition were critical to the adaptive processes required for survival under cold stress. Biological significance Work in our laboratory has shown that prolonged exposure of S. aureus to cold stress can result in the formation of small colony variants (SCVs) associated with significant alterations in the cell wall composition [8]. Further studies revealed that S. aureus altered cell size and cell wall thickness in response to exposure to cold temperatures, alterations in pH and exposure to antibiotics [10]. The current study has utilised the prolonged exposure to cold stress as a model system to explore changes in the proteome and associated metabolic homeostasis following environmental challenges. The study provides an improved understanding of how S. aureus adapts to the changing environment whilst in transition between human hosts. The results indicated an unexpected production of 9 ribosomal proteins and citric acid in response to cold stress suggesting specific survival

⁎ Corresponding author. E-mail address: [email protected] (R.H. Dunstan).

http://dx.doi.org/10.1016/j.jprot.2015.03.010 1874-3919/Crown Copyright © 2015 Published by Elsevier B.V. All rights reserved.

JO U RN A L OF P ROTE O M ICS 1 21 ( 20 1 5 ) 4 4 –55

45

roles for these proteins and citric acid as an adaptation mechanism for empowering survival under these conditions. Crown Copyright © 2015 Published by Elsevier B.V. All rights reserved.

1. Introduction Staphylococcus aureus is an opportunistic pathogen which can cause a wide range of acute and chronic infections leading to high morbidity and mortality worldwide [1,2]. S. aureus has been shown to survive across a wide range of environmental conditions and its potent pathogenicity has been attributed to its ability to rapidly adapt its metabolism and virulence for invasion via newly formed wound sites [3]. During the transfer between hosts in harsh nutrient-limited environments, the bacterium must modify its metabolism and structure to survive on fomites such as medical devices and instruments [4]. Bacterial survival mechanisms would involve metabolic, genomic, proteomic and structural adjustments within the cell to cope with changes in pH, temperature, osmotic pressure, nutrient availability and potential exposures to toxic chemicals. It has been proposed that these responses in bacteria could lead to heterogeneity of phenotypes within the bacterial population maximising chances of survival [5–10]. An obvious example of a phenotypic shift is the formation of small colony variants (SCVs) that can occur following exposure to antibiotics [5] or other environmental stressors such as prolonged cold stress [8]. SCVs have altered metabolism [11], cell structure and cell size characteristics compared with their corresponding wild types. These changes were reversible when SCVs were returned to stress-free, ideal growth conditions [8]. Coagulase negative staphylococci, S. epidermidis and S. lugdunensis, were found to have different biochemical and structural responses to prolonged cold exposure compared with S. aureus. Antibiotics, pH and osmotic stresses also resulted in different types of cellular responses in each of the staphylococci tested [10]. To investigate metabolic processes associated with SCV formation from antibiotic exposure, stable SCV mutants were isolated after exposure to gentamycin [12]. Proteomic investigations of these SCVs revealed extensive alterations in protein profiles suggesting an adaptive response via altered gene expression for survival of the bacterium under those conditions [12]. It has recently been shown that cold stress, as well as exposures to variations in pH and osmotic pressure, can also result in SCV formation [8,10]. It was concluded that several types of alterations in metabolism and cell structure may lead to the formation of SCVs dependent on the type of environmental stimuli involved. Cellular responses and adaptations to various environmental stresses are essential for all organisms in order to survive in their natural habitat [13]. As the proteome and metabolome represent the structural and functional operations of living organisms, proteomic and metabolomic investigations have become fundamental to understanding the dynamics of cellular function and the natural elements that stimulate a unique response in the microbial system [12,14]. When S. aureus was exposed to prolonged cold stress, a significant reduction in cell size and a corresponding increase in cell wall thickness was observed compared with non-stressed controls [10]. The cells exposed to the prolonged cold conditions at 4 °C also had

significantly different amino acid compositions of the cell wall and associated proteins [8]. These data suggested significant phenotypic variations resulting from the exposure to prolonged cold stress. The current study therefore sets out to characterise any changes in the cytoplasmic metabolic and proteomic profiles that could be associated with phenotypic shifting in S. aureus as a result of exposure to prolonged cold stress. It was hypothesized that S. aureus would alter its cytoplasmic metabolites and proteins to overcome cold stress and facilitate survival.

2. Materials and methods 2.1. Bacterial growth condition The bacterial strain used in this study was a clinical isolate of S. aureus from patients that had been suffering from chronic muscle pain [15]. This isolate had been used in subsequent investigations to monitor metabolic responses to environmental stresses [8,10]. The isolate has been maintained as culture stock on horse blood agar (HBA) and preserved appropriately on sterile glass beads at −80 °C with a regular sub-culturing to maintain viability. The identity of the isolate was checked regularly using API™ Staph biochemistry and through PCR [16]. Overnight cultures (50 ml) of S. aureus were grown for 16 h in Tryptic Soy Broth (TSB) at 37 °C with constant agitation (120 rpm). Eight flasks containing 95 ml TSB culture media were inoculated with 5 ml of overnight culture in 500 ml conical flasks which were then incubated at 37 °C with constant agitation (120 rpm) for 3 h. Four replicate cultures were harvested at the mid-exponential phase of growth (3 h) and processed for analyses to represent the baseline comparison as reference control samples. The remaining four cultures were stored at 4 °C for 2 weeks. Cell viability was checked by the plating method and the identity was confirmed with API. The reference control samples and the cold-treated samples were harvested by centrifugation at 6500 ×g for 25 min at 4 °C. Harvested cells were then washed three times with phosphate buffered saline (PBS) to ensure the removal of all residual TSB. The washed cells were immediately quenched using liquid nitrogen for lyophilisation and subsequent extraction for metabolomic and proteomic analysis.

2.2. Proteome analysis Proteins were extracted from the cytoplasm of dried cells from both the reference control and cold stressed samples. The cells were resuspended in 500 μl of SDS Lysis Buffer containing 2% SDS, 0.375 M Tris pH 6.8, 3.4 M sucrose (Sigma-Aldrich) and 1 tablet of protease inhibitor (complete Mini, Roche Diagnostics), thoroughly mixed and heated at 100 °C for 6 min. Cell debris was removed by centrifugation at 14,000 rpm for 25 min. The

46

JO U R N A L OF P ROTE O M ICS 1 21 ( 20 1 5 ) 4 4–55

supernatant containing the extracted proteins was carefully removed and stored at −20 °C until further investigation. Protein concentrations in each of the eight replicate treatment samples were determined using the BCA™ assay (Bio-Rad) following the manufacturer's instructions and Bovine Serum Albumin (BSA) was used as the reference standard. Aliquots from each sample containing 15 μg protein were precipitated using sample/methanol/chloroform in the ratio of 1:1:0.5 (v/v/v). The mixture was vortexed and centrifuged at 14,000 rpm for 15 min. The upper layer was discarded and 75% of the original volume of methanol was added to each replicate and centrifuged at 14,000 rpm for 20 min. The supernatant was removed without disrupting the protein pellet and was air dried for 10 min. Protein precipitate was redissolved with 15 μl of SDS-PAGE loading buffer SDS Lysis Buffer containing bromophenol blue and proteins were then denatured by adding 0.6 μl of 2-mercaptoethanol to each replicate and heating for 2 min. The samples were then run into 12% nUView™ Tris–Glycine NG Precast Gels, NuSep®, the gel visualised by UV ultraviolet (UV) light (Kodak System DC290) to determine any large scale differences in protein expression between the control and cold stressed cells. The gel evaluation provided a mechanism to check consistency between extractions of samples within treatments and to ensure that the protein concentrations had been accurately assayed prior to subsequent digestion and proteomic analysis by LC–MS/MS. Aliquots of 150 μg protein from each sample were precipitated as above, digested by adding trypsin (Promega) in a ratio of 50:1 (protein:trypsin) in 25 mM ammonium bicarbonate and incubated overnight at 37 °C with constant shaking. Trypsin digestion was inactivated by adding formic acid to 1% (v/v) final concentration and samples centrifuged at 14,000 rpm for 40 min immediately prior to proteomic analyses. The peptides yielded from the tryptic digestion of cytoplasmic proteins were analysed using liquid chromatography coupled with electrospray ionization tandem mass spectrometry (LC/ESI–MS/MS). The peptides were separated by nanoscale reverse phase high performance liquid chromatography (RP-HPLC) (Dionex Ultimate 3000 RSLC nano LC System) equipped with online electrospray ionization (ESI). The samples were initially loaded onto the system via a trap column (Dionex, Acclaim PepMap 100, C18 3 μm bead 100 A°, ID 75 μm × 2 cm) at flow rate of 5 μl/min in Solvent A (2% acetonitrile, 0.1% triflouroacetic acid in HPLC grade water) to wash the samples before passing onto the analytical column (Dionex, Acclaim PepMap C18 2 μm bead, 75 μm × 15 cm) where the peptides were resolved using a gradient of 2–40% solvent B (0.1% formic acid/80% acetonitrile) over 120 min at flow rate of 400 nl/min. The temperature of the column was set to 35 °C. As the peptides eluted from the system, they were directly sprayed into the nanoflow ESI source of a Bruker AmaZon ETD, 3D-ion trap mass spectrometer (Bruker Daltonics Bremen, Germany).

2.2.1. Protein identification and quantification Bruker raw format (.baf) files were converted into mascot generic format using Data Analysis 4.1 (Bruker Daltonics Bremen, Germany). They were then imported into Bruker's Proteinscape platform and searched against the Uniprot database (538,013 sequences), Firmicutes taxonomy (Gram positive

bacteria) using Mascot 2.3.02 (Matrix Science) search algorithm. Search parameters were set as following: variable modifications (Oxidation of Methionine M; and Deamidation of Asparagine N and Glutamine Q), Trypsin was selected as the enzyme with up to 2 missed cleavages allowed. Peptide mass tolerance was set to 1 Da and 0.7 Da for fragment ion mass tolerance. Peptide thresholds were set requiring False Positive Rate less than 0.05% and an individual peptide MASCOT score greater than 40. Those spectra meeting these criteria were validated by manual inspection to ensure accurate y- and b-ion detection with overlapping sequence coverage. AmaZon ETD raw format files were converted into MZXML format using Compassxport (Bruker Daltonics, Bremen, Germany) to be imported to the commercial label-free quantitation software Decyder™ MS 2.0 PepDetect module (GE Healthcare, UK) for peptide detection, and subsequently for the relative quantitation of peptides between control and cold stressed samples. Peptides in each of the reference control and cold replicate LCMS runs were detected and cross-matched according to retention time and mass to charge ratio (m/z) using the Decyder PepDetect and PepMatch modules respectively. Background subtraction modelling was set to smooth surface to model local variations in background intensity to account for varying background due to the acetonitrile gradient. Charge state assignment was set to always require a charge state assignment with three isotopic peaks required. LC peak shape tolerance, mass to charge shift and mz shape tolerance were set to 20%, 0.1 u and 5% respectively. The intensity of detected peptides was calculated by integrating the area of the peptides' Extracted Ion Chromatogram and converted into a log2 value. Processed intensity maps were then submitted to the Decyder PepMatch module. Each analysis was assigned as either a control or cold stressed experimental group and time aligned according to Base-peak Ion Chromatogram. Time aligned intensity maps were matched and cross detected according to retention time and m/z tolerances (within 1 minute and 0.5 Da respectively). PepMatched data was assessed manually for accuracy and normalized in software by measured intensity distribution (assumes that a majority of peptides between samples do not vary in intensity). Peptides with a t-test probability of < 0.05 were manually validated by comparing Extracted Ion Chromatograms (EIC) from the raw data. The Matched data generated by Decyder™ MS was exported to an Excel® (Microsoft®) spreadsheet file for further data analysis using Principal Component Analysis (PCA) (SIMCA-p+ (12.0, Umetrics Sweden).

2.3. Metabolic profiling analyses Approximately 10–12 mg of lyophilised cells were resuspended with 10 ml of 1:1(v/v) of cold methanol/water stored at −20 °C and mixed thoroughly. The methanol/water lyophilised cell slurries were snap-frozen in liquid nitrogen and placed at −20 °C for 30 min for a process of slow thawing. Metabolites were separated from the cell debris by centrifugation at 6500 ×g for 25 min. The supernatants that contained the metabolites were dried using a vacuum concentrator (CentriVap, LABCORNCO, VWR) and the dried metabolites were resuspended with 450 μl of sterile Milli-Q water.

JO U RN A L OF P ROTE O M ICS 1 21 ( 20 1 5 ) 4 4 –55

The amino acid metabolites were analysed using a commercial analytical kit (Phenomenex® EZ: faast™). The method was conducted according to the manufacturer's instructions. The derivatised amino acids were then separated by an auto-sampler Agilent gas chromatography (Hewlett Packard HP 6890 series) coupled with flame ionization detector which was calibrated to measure more than 40 amino acid metabolites as previously described [8]. The injection volume was 2 μl with splitless mode and flow rate of the carrier gas (Helium) was 0.5 ml/min. Nor-valine was used as an internal standard to calculate the concentrations of amino acids present in the sample as nmol/mg cell dry weight. Additional cytoplasmic metabolites including organic acids, carbohydrates, purines and pyrimidines were evaluated by forming the methoxy-amine-trimethyl-silyl (TMS) derivatives by reacting the dry extracts with Methoxyamine-HCL (MOX) and Bis(trimethylsily)trifluoraacetamide (BSTFA). Lyophilised metabolites were mixed with 50 μl of MOX, vortexed, and then heated at 60 °C for 30 min. Samples were allowed to cool and then 150 μl of BSTFA was added to each sample, vortexed and heated at 100 °C for 60 min. Derivatised samples were analysed by gas chromatography (Agilent, Hewlett-Packed 5973) coupled with mass spectrometry. The injection volume was 1 μl/sample and flow rate was 0.5 ml/min. Metabolites in the chromatogram were identified on the basis of matching their mass spectra and retention time indices with data in user-generated mass spectral libraries generated from reference standards.

2.3.1. Metabolic profile data processing and analysis Four replicates of the reference control and the cells exposed to prolonged cold stress were used in this experiment to investigate the responses of the cytoplasmic metabolites and proteome. The acquired amino acid data obtained from GC-FID were imported to STATISTICA 6 (ANOVA) to find the amino acids that were significantly altered after being exposed to cold stress. Principal Component Analysis (PCA) was then preformed utilizing SIMCA-p+ (12.0, Umetrics Sweden) [8]. The data were subjected to mean centring and unit variance scaling before PCA calculations. The model complexity and validity were assessed by leave one out cross-validation as implemented in the software.

3. Results 3.1. Proteome analysis Eight broth cultures of a clinical isolate of S. aureus were grown to the mid-exponential phase of growth at 37 °C under optimal growth conditions. Four cultures were harvested for extraction to represent the reference control group for comparison with the remaining four cultures which were subjected to a prolonged cold treatment for 2 weeks at 4 °C prior to harvesting. The cells were checked for viability after the cold treatment and their identities were confirmed. Protein analysis by SDS-PAGE showed that the cytoplasmic proteins in both control and cold-stressed samples had similar profile compositions with some distinct variations apparent in the cold stressed cells. The

47

intensity of the two bands with molecular weights of ≈14 kDa and ≈15 kDa were relatively less intense in all cold-stressed samples compared with the reference controls as shown in Fig. 1. It was evident that the cold stress samples consistently had more intense bands at ≈12 kDa, ≈19 kDa, ≈25 kDa and ≈33 kDa. The SDS-PAGE results indicated that the protein extractions for the control and treatment samples were consistent between replicates and provided evidence that S. aureus altered its proteomic homeostasis after being exposed to prolonged cold stress. Proteins from the eight replicates (150 μg per sample) were digested using tryptic enzyme to investigate the types of specific protein alterations inferred by SDS-PAGE. The resulting digests were subjected to liquid chromatography mass spectrometry (LC–MS/MS) to analyse peptides for the identification of those cytoplasmic proteins which altered in composition during the prolonged cold treatment compared with the reference control samples [17]. The data revealed that 17 proteins were up-regulated and 11 proteins were down-regulated in response to prolonged cold stress at 4 °C (Table 1). The major functional group of proteins that displayed an increase during the cold exposure were translation proteins including ten ribosomal proteins S1, S7, S8, L3, L5, L10, L17, L24, L25 and L30 as well as elongation factor G (Table 1). A total of 32 ribosomal proteins were identified in the extracts but not all ribosomal proteins altered their abundance in response to the prolonged cold stress. In Fig. 2 the data have been summarised for the ribosomal proteins L18, L15 and S13 representing those proteins which did not display an altered abundance following the prolonged cold treatment compared with L10, L30 and S8 which displayed significant increases in cytoplasmic concentration following prolonged exposure to cold stress (P < 0.01). The only translation protein to be down-regulated was elongation factor Tu. Proteins associated with post-translational modification (trigger factor), transcription regulation (transcriptional regulator sarA) and cell division (cell division protein) were up-regulated. Alkaline shock protein 23 was also up-regulated as a potential response to stress but the other stress-related protein, peptide methionine sulfoxide reductase msrB, was down-regulated. Two proteins involved in energy metabolism, ATP synthase subunit alpha and enolase, were up-regulated whereas all other responses associated with carbohydrate metabolism, amino acid biosynthesis, cell adhesion and cell biosynthesis were down-regulated as might be expected under the prolonged cold conditions. Principle components analysis (PCA) for the peptides that showed a significant alteration was carried out to further investigate proteome profiles rendering a two component model as validated by cross validation (CV). The PCA scores for each replicate revealed 2 clusters well separated by t1 scores where the controls had negative t1 scores and the cells exposed to prolonged cold stress had positive t1 scores as shown in Fig. 3. The first principal component (PC1) explained more than 90% of the data variation (i.e. R2 = 0.91 and CV by Q2 = 0.87) and the Eigenvalue was 7.3 as compared with 0.7 for PC2. The large difference between the Eigenvalues, i.e. more than ten-fold, indicated that most of the explained data variation related to clustering of the cold treated bacteria from un-treated samples. Furthermore the second component scores displayed a near

48

JO U R N A L OF P ROTE O M ICS 1 21 ( 20 1 5 ) 4 4–55

Lanes 7-11 Prolonged cold

Lanes 2-6 Reference controls 1

2

3

4

5

6

7

8

9

10

11

Fig. 1 – SDS-PAGE analyses of cytoplasmic protein extracts from cells grown to the mid-exponential phase of growth at 37 °C under optimal conditions (lanes 2–6) compared with corresponding extracts from cells subjected to prolonged cold stress (lanes 7–11). Lane 1 indicates protein molecular weight markers for 10–250 kDa. The symbol “+” and “−” indicate those bands deemed to be respectively higher and lower in intensity for the cold stressed samples relative to control reference samples.

orthogonal distribution to the first component, which certifies high accuracy estimates of the individual peptide leverages via the PC1 loadings. In corollary, these analyses showed that the replicate reference control samples of S. aureus were clustered separately from the replicates exposed to the prolonged cold treatment representing the presence of different proteome profiles in the two groups (Fig. 3).

3.2. Metabolic profile analyses Cytoplasmic fractions were also extracted and analysed for determining the amino acid composition profiles. These results indicated that all the identified amino acids, except glutamine, underwent substantial alterations in cytoplasmic composition relative to the reference controls following exposure to the prolonged cold treatment (Table 2). The total abundance of cytoplasmic amino acids in the reference control cells was 332 nmol mg−1 dry mass which was drastically reduced to 82.7 nmol mg−1 dry mass in the cells following prolonged exposure to 4 °C. This reduction in total abundance was driven by diminished levels of 15 out of 19 amino acid components detected, but there were 4 notable exceptions including glycine, tyrosine, tryptophan and ß-aminoisobutyric acid, which displayed significant increases in cytoplasmic concentrations. The most abundant amino acids present in the control cells included glutamic acid, aspartic acid, alanine and proline which were 1–3 orders of magnitude higher than other amino components. Following the exposure of the cells to prolonged cold stress, the glutamic acid was reduced by a factor of 30 and the other major components by factors of 1.8–13.6 representing considerable alterations in cytoplasmic homeostasis.

The PCA analysis of the amino acid data revealed a similar clustering and separation of control and cold treated cells as shown for the peptide data. The analysis revealed that all included amino acid variation, as estimated by PCA loadings p1 (see Fig. 4B), contributed to the cluster separation with one exception, i.e. glutamine. This model interpretation was corroborated by the confidence intervals, as generated by the cross validation, indicating solid correlations between the clustering and the amino acid variations along the first component due to the high Q2 (i.e. >0.9). Moreover the within cluster variation was modelled as orthogonal, by PC2, to the clustering, indicating un-biased estimates of the p1-loadings. These analyses showed that the replicate reference control samples of S. aureus were clustered separately from the replicates exposed to the prolonged cold treatment representing the presence of different amino acid profiles in the two groups (Fig. 4A). Cell extracts were also derivatised to form the trimethylsilyl/ methoxy amine derivatives (TMS and TMS-MOX) to investigate changes amongst the other non-amino acid cytoplasmic components. Twenty additional metabolites were identified by comparisons to their reference mass spectra in the user-generated database and their retention time indices. The range of metabolites identified included organic acids, sugars, sugar alcohols, purine and pyrimidine bases, nucleosides and nucleotides. There were several minor alterations in the profiles of these metabolites but the major feature of these results was the very large increase in the abundance of citric acid in the cytoplasm following cold stress where it accounted for a 100-fold increase compared with the reference control as shown in Fig. 5.

Table 1 – Proteins altered in response to prolonged cold stress (P < 0.01). Function Translation

Protein name

Accession

30S ribosomal protein S1 30S ribosomal protein S7 30S ribosomal protein S8

RS1_STAAC RS7_STAAB RS8_STAAS

50S ribosomal protein 50S ribosomal protein 50S ribosomal protein 50S ribosomal protein 50S ribosomal protein 50S ribosomal protein Elongation factor G

RL3_STAA3 RL5_STAA3 RL10_STAAS RL24_STAA1 RL25_STAA3 RL30_STAAC EFG_STAAB

L3 L5 L10 L24 L25 L30

Peptides

MASCOT score

Regulation

MW KDa

K.ATLPNEDVVESDPSTTK.A R.ILYSAFNLVEQR.S K.LELPASNIK.K M.TMTDPIADMLTR.V K.EADSGIFYYQNAK.K K.LISVSLPR.V K.SGVMEGNVITAEEVK.T R.VVVEGVNIMK.K R.TVEVPVQLVGEAVGAK.E K.TNSSVVVEDNPAIR.G K.LYDGSYHDVDSSEMAFK.I K.SIAEDIIK.K K.LLDYAEAGDNIGALLR.G K.VGEEVEIIGLHDTSK.T R.MVSEFAQR.I

−1.26 2.02 −1.46 −1.28 −0.96 −0.97 −1.32 −0.90 −1.14 −1.61 −0.40 −0.61 0.66 0.59 −1.30

84.5 80.4 45.68 100.06 60.9 45.14 95.56 54.35 109.5 72.76 71.14 43.46 98.8 44.53 42.77

Up-regulated Up-regulated Up-regulated

43.2 17.8 14.8

Up-regulated Up-regulated Up-regulated Up-regulated Up-regulated Up-regulated Up-regulated

22.5 20.3 17.7 11.5 23.8 6.5 76.6

Down-regulated

43.1

Up-regulated

48.6

K.ILSQEDYFDK.K K.VIGVGGGGNNAVNR.M K.VILEYGESAPK.I K.QQEQNQEPQFK.N K.ALDDDEIIELVDK.S R.AVFVLDADNK.V R.LSNGNTAGATGSSAAQIMAQR.T R.LSNGNTAGATGSSAAQIMAQR.T R.LSNGNTAGATGSSAAQIMAQR.T R.IMEVPVGEELIGR.V R.IAQVNAVDLDEVLNR.K

−0.74 −1.5 −1.70 −1.59 0.76 0.80 3.40 3.23 2.79 −1.2 1.00

68.23 87.5 56.3 76.34 132.65 104.75 74.24 132.65 100.43 60.64 129.43

Up-regulated Up-regulated Up-regulated

14.7 41.0 19.2

Down-regulated Down-regulated Down-regulated

16.3 18.0 24.2

Up-regulated Down-regulated

54.5 34.0

−0.81 0.99 0.96 1.36 1.39 0.86 0.61

80.33 70.87 114.73 69.33 61.28 104.75 43.99

Up-regulated Down-regulated

47.1 33.0

Down-regulated Down-regulated

34.4 35.2

EFTU_STAAC

Trigger factor

TIG_STAAB

Transcriptional regulator sarA Cell division protein Alkaline shock protein 23

SARA_STAAB FTSZ_STAAC ASP23_STAAC

Peptide methionine sulfoxide reductase msrB Probable thiol peroxidase Immunodominant staphylococcal antigen A precursor

MSRB_STAAM TPX_STAAC ISAA_STAAC

ATP synthase subunit alpha Probable manganese-dependent inorganic pyrophosphatase Enolase Fructose-bisphosphate aldolase class 1

ATPA_STAAB PPAC_STAAB

LDH2_STAAM ODPB_STAAC ILVE_STAAC

R.LEMPQVDEAELLEGLK.Q

0.6

72.33

Down-regulated

40.1

Cell adhesion

L-lactate dehydrogenase 2 Pyruvate dehydrogenase E1 component subunit beta Probable branched-chain-amino-acid aminotransferase Serine-aspartate repeat-containing protein D

R.GLETAVGDEGGFAPK.F R.VVTSPSFSPDK.I K.GFIAALDQSGGSTPK.A K.EYGVNEDQYSNEDEMFQLVHDMR.T K.GLVPIIEPEVNINAK.D K.IYEMPLSAEEQALFDK.S R.VAAADTIYPFTQAENVWLPNK.N

SDRD_STAA8

ATL_STAAM

74.95 65.3 43.45 50.37 64.68 53.08 77.89

146.0

Bifunctional autolysin precursor [Includes: N-acetylmuramoyl-L-alanine amidase

2.02 3.05 2.80 2.47 2.58 1.34 2.30

Down-regulated

Cell wall biogenesis and degradation

K.YSLGDYVWYDSNK.D K.QTIYVNPSENSLTNAK.L K.NIGDIKDPNNGETIATAK.H K.TTANIQYPDYVVNEK.N K.QTISNQEALQPDLQENK.S K.AFNEQPFAVVK.E K.NPTQNISGTQVYQDPAIVQPK.T

Down-regulated

136.7

Posttranslational modification Transcription regulation Cell division Stress related proteins, protein turnover chaperones

Pathogenicity

Energy production and conversion

Amino acid biosynthesis

ENO_STAAB ALF1_STAAC

49

Elongation factor Tu

JO U RN A L OF P ROTE O M ICS 1 21 ( 20 1 5 ) 4 4 –55

Average difference (2Log)

50

26 Affected ribosomal proteins

A

25 24 23

Control

22

Cold

21

Relative Abundance

Relative Abundance

JO U R N A L OF P ROTE O M ICS 1 21 ( 20 1 5 ) 4 4–55

30

Unaffected ribosomal proteins

B

25 20 15

Control

10

Cold

5 0

20

L30 S8 L10 Ribosomal proteins

L18 S13 L15 Ribosomal proteins

Fig. 2 – The response of some ribosomal proteins to prolonged cold stress. A: Shows examples of ribosomal proteins that were significantly altered in response to prolonged cold stress (P < 0.05). B: Shows examples of ribosomal proteins that were not significantly altered in response to prolonged cold stress.

4. Discussion The proteomic and metabolic profiling approaches applied in this study provided evidence that S. aureus responded to prolonged exposure to cold stress at 4 °C by altering its cytoplasmic proteins and metabolites to maintain cellular homeostasis and hence facilitate survival. SDS-PAGE analyses of cytoplasmic protein extracts from the cells revealed clear differences in protein composition between the reference control and cells exposed to prolonged cold stress at 4 °C. Further proteome analysis demonstrated that 28 proteins were significantly altered in cold treated samples (Table 1), 17 of which were up-regulated and 11 down-regulated. Ribosomal proteins were the main group of proteins that were significantly up-regulated in the cold-treated samples. These ribosomal proteins were normally thought to be associated

with the synthesis of proteins but it has been suggested that some ribosomal proteins may have additional or alternative roles including acting as a temperature sensor to acclimatize to cold stress and facilitate appropriate cellular function [18]. In a similar manner, an extensive up-regulation in ribosomal proteins was observed in clinically-derived small colony variants of S. aureus isolated from an osteomyelitis patient who was comprehensively treated with gentamicin [12]. Listeria monocytogenes has also shown a significant upregulation of ribosomal genes in response to cold stress and osmotic stress [19]. This up-regulation in cytoplasmic ribosomal proteins might reflect a requirement to synthesise more proteins to overcome the cold-stress conditions. In the present study, only a small number of other proteins were observed to increase in abundance following the prolonged exposure to 4 °C including trigger factor, transcriptional regulator sarA, cell division protein, ATP synthase subunit

Fig. 3 – Principle component analysis (PCA) scores (t1 versus t2) plotted from S. aureus proteomic peptide data. The S. aureus cultures were grown under ideal conditions at 37 °C (control) or exposed to prolonged cold stress at 4 °C for 2 weeks (cold) before extraction of proteins and analyses by LC–MS/MS.

51

JO U RN A L OF P ROTE O M ICS 1 21 ( 20 1 5 ) 4 4 –55

Table 2 – The concentrations of cytoplasmic amino acids extracted from cells grown to the mid-exponential phase of growth at 37 °C (reference control) compared with equivalent sets of cells that were subsequently exposed to 2 weeks of cold stress at 4 °C (Means + SD). Amino acid concentrations expressed as nmol mg−1 dry cell weight (Mean + SD, P < 0.05). Amino acids

Amino acids significantly decreased

Alanine Aminobutyric acid Valine Isoleucine Proline Asparagine Aspartic acid Methionine Hydroxyproline Glutamic acid Phenylalanine Cystathionine Glycine–proline (dipeptide) Lysine Histidine

Amino acid concentration in reference control samples n=4 nmol mg−1 dry cell mass

Amino acid concentration in cold stressed samples n=4 nmol mg−1 dry cell mass

38.2 3.8 3.8 0.5 31.5 2.4 81.8 2.6 0.8 150 2.6 0.6 1.0 2.8 8.4

± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

1.06 0.06 0.14 0.01 1.01 0.06 3.1 0.10 0.01 1 2.9 0.08 0.02 0.04 0.06 0.5

2.8 0 1.0 0 6.7 0 46 1.09 0 5.0 2.0 0 0 1.8 5.4

± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

0.05 0 0.06 0 0.04 0 1.4 0.01 0 0.13 0.03 0 0 0.05 0.3

0 0.6 0.17 4.05 332

± ± ± ± ±

0 0.04 0.004 0.1 27.6

0.9 0.9 0.8 12.7 82.7

± ± ± ± ±

0.01 0.04 0.03 0.1 13.0

Amino acids significantly increased Glycine Tyrosine Tryptophan ß-Aminoisobutyric acid Total amino acids measured for controls and cold treatments

alpha and enolase. All other significant alterations in protein composition involved a reduction in abundances and included proteins associated with stress, pathogenicity, carbohydrate metabolism, amino acid biosynthesis, cell adhesion and cell wall biogenesis. It would be expected that, under the prolonged cold conditions, bacteria would reduce growth and downregulate their metabolic rate to survive and the observed proteomic response was consistent with this strategy. Since there was no evidence of substantial protein synthesis to accommodate the response to the cold exposure, the up-regulation of nine ribosomal proteins when the other ribosomal proteins remained constant suggested that these proteins may play specific roles in the adaptation of these bacteria to the prolonged exposure to 4 °C. It has already been established that some ribosomal proteins are not important in the translation apparatus as some such as S1, S21, S22, S31e, and L25 do not exist in all bacteria [20]. It has also been established that the knock-out of some of these ribosomal genes did not affect the viability of the cells [21]. Alternatively, several ribosomal proteins have been found to have an extra-ribosomal role by functioning as regulatory proteins (e.g. S1 regulating transcription efficacy) or forming complexes with other cellular constituents. S1 is the largest ribosomal protein and interacts with ribosomes and mRNA and thus can be directly involved in the initiation of the translation process [20,22,23]. Ribosomal proteins have also been found amongst the surface and secreted proteins of bacteria [23–25] but the functions of these secreted ribosomal proteins remain unclear. It is therefore proposed that certain members of the suite of ribosomal proteins may be secreted to the surface of the cell or into the external environment as a defensive mechanism in response to external challenges from host immune system, antibiotics and changing environmental conditions. This proposal is further supported

with the finding that the 50S ribosomal protein L25, a general stress protein, was found to have been secreted together with other proteins after exposure of S. aureus to antimicrobial silver ions [25]. Significant alterations in four proteins associated with glycolysis were identified in cells undergoing prolonged exposure at 4 °C. Enolase, a catalyst in the formation of phosphoenolpyruvate from 2-glycerophosphate in glycolysis, was elevated in the cells exposed to the cold stress. Enolase has been classified as a moonlighting protein since it has been found to have multiple roles in the cell; in addition to glycolysis, it has been suggested to play important roles in tRNA confirmation as well as functioning as a plasminogen receptor on the cell surface [26–29]. These additional roles may form part of the response for survival under prolonged conditions at 4 °C. Conversely, three proteins that were also associated with glycolysis were reduced; fructose-bisphosphate aldolase class 1, L-pyruvate dehydrogenase E1 component subunit-beta and lactate dehydrogenase 2. These reductions in cytoplasmic levels may be linked to a potential down-regulation of energy metabolism to facilitate survival at 4 °C. There was no evidence of reduced levels of other glycolytic proteins suggesting that alterations in these 3 proteins may represent an efficient mechanism of maintaining lower metabolic rates whilst preserving a capacity to rapidly power up the cell in return to favourable conditions by restoring the levels of these 3 components. Previous research in our laboratory has shown that prolonged cold stress of S. aureus at 4 °C leads to the formation of small colony variants (SCVs) and that the proportions of SCVs in the population increase with exposure time with concomitant alterations in the cell wall structure and composition [8]. This current study has provided a similar stress regime to broth cultures to generate sufficient cell material for

52

JO U R N A L OF P ROTE O M ICS 1 21 ( 20 1 5 ) 4 4–55

A

B

Fig. 4 – A: Principle component analysis (PCA) scores (t1 versus t2) scatter plotted from S. aureus profiles of cytoplasmic amino acid data. The S. aureus cultures were grown under ideal conditions at 37 °C (control) or exposed to prolonged cold stress at 4 °C for 2 weeks (cold) before extraction of amino acids for analyses by GC-FID. B: Loadings (p1) generated by principle component analysis (PCA) from S. aureus profiles of cytoplasmic amino acid data, indicating that all measured AAs, except GLN, contributed significantly to the obtained model according to the cross validated first component confidence intervals.

analyses of the proteome and the metabolic profiles in response to the prolonged exposures at 4 °C. Prior investigations of S. aureus SCVs have shown that these phenotypes do not have the usual array of virulence factors but instead have a range of factors that facilitate adhesion to host cells and intracellular colonisation in the host cells [9,11,30]. SCVs produce an impressive array of adhesive factors that are associated with intracellular colonisation and survival such as fibronectin-binding proteins that facilitate strong adhesion to endothelial cells therefore facilitating rapid internalization [31]. This increased endocytic uptake of the hemB mutant was correlated with an increased expression of the fibrinogen adhesion gene cflA and was found to occur independently of the agr gene that mainly regulates virulence factors in S. aureus [9,32,33]. The temperature of 4 °C was selected as this is the standard temperature regime for cold storage and allows an assessment of the remarkable adaptability and survival capacity of this pathogen. The knowledge of how this organism can survive at these temperatures may provide insight to better mechanisms of aseptic preparation and storage. Clinical SCV isolates from patients following long term treatment with gentamicin showed enhanced expressions of alkaline shock protein 23, trigger factor, transcriptional regulator sarA, and cell division protein [12] similar to the responses observed in S. aureus in this investigation. In the absence of host cells as an external epigenetic trigger in the

present study under in vitro conditions at 4 °C, S. aureus may have elicited a different kind of response by down-regulating proteins associated with cell adhesion and pathogenicity with a view to minimising the metabolic rate, preserving cellular integrity and being ready to reactivate rapidly if an appropriate opportunity was presented. Our data has not shown any unexpected production of CspB and CspC proteins. These proteins might be an instant response to cold shock but it would appear, that once acclimated, these proteins were not relevant to the new homeostasis established for the ongoing survival for 2 weeks at 4 °C. However, a study which exposed S. aureus CECT 976 to 12 °C did not show significant up-regulations of CspC and CspB [34]. The promoter of CspC gene was shown to be expressed greatly in response to exposure to antibiotics as well as to H2O2 more strongly than cold [35]. The cold shock proteins have been found to be higher in methicillin-resistant S. aureus than in methicillin-sensitive S. aureus, suggesting that these proteins may have a role in the antibiotic resistance associated with virulence rather than in cold stress [36]. Our experimental model was based on conditions that can generate SCVs which have reduced virulence factors. This may also explain why these proteins did not alter during exposure to 4 °C for 2 weeks. ATP synthase subunit β and elongation factor G are important proteins in the production of energy and protein synthesis respectively, and were up-regulated in the cold-treated samples.

JO U RN A L OF P ROTE O M ICS 1 21 ( 20 1 5 ) 4 4 –55

53

Fig. 5 – A section of the GC–MS chromatograms from S. aureus which measured the TMS derivatives to generate the metabolite profiles of organic acids, amino acids, purines, pyrimidines and sugars. This section shows the appearance of substantial quantities of citric acid in the cytoplasm following exposure to the prolonged cold stress at 4 °C for 2 weeks compared with the reference control grown under ideal conditions at 37 °C.

A similar finding was observed in Escherichia coli when it was exposed to prolonged cold stress [37]. The up-regulation of these proteins could suggest that these two also represent general cold stress proteins in bacteria. The metabolic profile analyses demonstrated that prolonged exposure of S. aureus to 4 °C resulted in substantial changes in cytoplasmic composition of amino acids as well as a substantial increase in citric acid concentration. As indicated in Table 2, 15 amino acids were reduced in concentration and 4 amino acids were increased compared with the control cells. The major cytoplasmic amino acid in the control conditions was glutamic acid at 150 ± 12.9 nmol mg−1 dry cell mass representing 45% of this pool of amino acids; this was reduced to 5 nmol mg−1 dry cell mass (6%) following prolonged exposure to cold. The major cytoplasmic amino acid component after prolonged cold stress was ß-aminoisobutyric acid which was 4.05 nmol mg−1 dry cell mass (1.2%) in the control increasing to 12.7 nmol mg−1 dry cell mass (15%) following prolonged exposure to 4 °C. The reductions in cytoplasmic amino acids may reflect utilisation of the free pool of amino acids to produce new essential stress proteins or to repair damaged or misfolded proteins to facilitate acclimation of the bacterial cell to the changing environment. It has been noted in prior investigations that prolonged cold exposures resulted in increased levels of peptidoglycan biosynthesis in the cell wall of S. aureus as well as in cell wall associated proteins [8,38,39]. Reduced metabolic rates would then only support lower standing levels of cytoplasmic amino

acids and some of the amino acids such as glutamic acid may be used as an efficient source of energy to support the transition to surviving new environmental conditions. This would represent an essential strategy in adjusting the cellular homeostasis under cold stress [40,41]. The alterations in amino acid composition were accompanied by a substantial increase in the relative abundance of citric acid following prolonged exposure of cells to 4 °C as shown in Fig. 5. Citric acid is a powerful chelation agent and it may play a role in managing concentrations of cations such as Ca2+ for survival of S. aureus during prolonged exposure to cold stress. Evidence suggests that calcium is involved in the regulation of a range of processes in prokaryotes including cell division and gene expression in response to external stimuli [42]. Non-proteogenic amino acids may play important roles as intermediates in, and regulation of, metabolism. For instance, beta-aminoisobutyric acid showed an extensive increase in concentration in cold treated samples suggesting that it may have a function in adapting to cold stress. Beta-aminoisobutyric acid was found to improve the resistance of the plant Arabidopsis to microbial pathogens and environmental stressors [43]. In addition, isobutyrate was observed to be substantially produced in the cells of Mesorhizobium sp. strain N33 when grown at 4 °C, which was proposed to be due to the use of this compound as a precursor for the cold-regulated fatty acid adjustment to cold stress acclimation [40].

54

JO U R N A L OF P ROTE O M ICS 1 21 ( 20 1 5 ) 4 4–55

In conclusion, proteomic and metabolomic approaches provided evidence that S. aureus adapted to prolonged cold stress by altering its proteins and metabolites. The key findings of this study were that ribosomal proteins represented the primary response group of cytoplasmic proteins and the amino acids, citric acid and beta-aminoisobutyric acid underwent substantial alterations in cytoplasmic composition following prolonged cold stress. It is proposed that these substantial and significant alterations in proteins and metabolites represent the activation of a phenotypic shift as an adaptation mechanism for conferring survival under cold stress conditions.

Conflict of interest

[7]

[8]

[9]

[10]

The authors have declared that no conflict of interests exists. [11]

Acknowledgement Mousa Alreshidi was supported by the Saudi government. This work was also supported by the University of Newcastle (grant no. G0189306) Australia funds for post-graduate research as well as the Harold Stannet Williams and Judith Mason Research Foundation. The funders had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. http://www.anz.com.au/resources/2/f/ 2fce2b804a47773887c8cfac93b0266b/MasonNatMedGuidelines. pdf?CACHEID=2fce2b804a47773887c8cfac93b0266b.

[12]

[13]

[14]

[15]

[16]

Appendix A. Supplementary data

[17]

Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.jprot.2015.03.010. [18]

REFERENCES

[1] Lowy FD. Staphylococcus aureus infections — reply. N Engl J Med 1998;339:2026–7. [2] Gill SR, Fouts DE, Archer GL, Mongodin EF, DeBoy RT, Ravel J, et al. Insights on evolution of virulence and resistance from the complete genome analysis of an early methicillinresistant Staphylococcus aureus strain and a biofilm-producing methicillin-resistant Staphylococcus epidermidis strain. J Bacteriol 2005;187:2426–38. [3] Schneider LA, Korber A, Grabbe S, Dissemond J. Influence of pH on wound-healing: a new perspective for wound-therapy? Arch Dermatol Res 2007;298:413–20. [4] Baumert N, von Eiff C, Schaaff F, Peters G, Proctor RA, Sahl HG. Physiology and antibiotic susceptibility of Staphylococcus aureus small colony variants. Microb Drug Resist 2002;8:253–60. [5] von Eiff C. Staphylococcus aureus small colony variants: a challenge to microbiologists and clinicians. Int J Antimicrob Agents 2008;31:507–10. [6] Abu-Qatouseh LF, Chinni SV, Seggewiss J, Proctor RA, Brosius J, Rozhdestvensky TS, et al. Identification of differentially expressed small non-protein-coding RNAs in Staphylococcus

[19]

[20]

[21]

[22] [23]

[24] [25]

[26]

aureus displaying both the normal and the small-colony variant phenotype. J Mol Med 2010;88:565–75. Morikawa K, Ohniwa RL, Ohta T, Tanaka Y, Takeyasu K, Msadek T. Adaptation beyond the stress response: cell structure dynamics and population heterogeneity in Staphylococcus aureus. Microbes Environ 2010;25:75–82. Onyango LA, Dunstan RH, Gottfries J, von Eiff C, Roberts TK. Effect of low temperature on growth and ultra-structure of Staphylococcus spp. PLoS One 2012;7:e29031. Alreshidi MM, Dunstan HR, Roberts TK, Onyango LA. Staphylococcal phenomics: metabolomic and proteomic responses to environmental stressors. In: Méndez-Vilas A, editor. Microbial Pathogens and Strategies for Combating Them: Science, Technology and Education. Badajoz: Spain Formatex Research Center; 2013. p. 690–701. Onyango LA, Dunstan RH, Roberts TK, Macdonald MM, Gottfries J. Phenotypic variants of staphylococci and their underlying population distributions following exposure to stress. PLoS One 2013;8. von Eiff C, Peters G, Becker K. The small colony variant (SCV) concept — the role of staphylococcal SCVs in persistent infections. Injury 2006;37(Suppl. 2):S26–33. Kriegeskorte A, Konig S, Sander G, Pirkl A, Mahabir E, Proctor RA, et al. Small colony variants of Staphylococcus aureus reveal distinct protein profiles. Proteomics 2011;11:2476–90. Hunter AJ, Ottoson N, Boerth N, Koretzky GA, Shimizu Y. Cutting edge: a novel function for the SLAP-130/FYB adapter protein in beta 1 integrin signaling and T lymphocyte migration. J Immunol 2000;164:1143–7. Hecker M, Becher D, Fuchs S, Engelmann S. A proteomic view of cell physiology and virulence of Staphylococcus aureus. Int J Med Microbiol 2010;300:76–87. Butt HL, Dunstan RH, McGregor NR, Roberts TK, Zerbes M, Klineberg IJ. An association of membrane-damaging toxins from coagulase-negative staphylococci and chronic orofacial muscle pain. J Med Microbiol 1998;47:577–84. Brown GK, Martin AR, Roberts TK, Aitken RJ. Detection of Ehrlichia platys in dogs in Australia. Aust Vet J 2001;79:554–8. Alreshidi MM, Dunstan RH, Macdonald MM, Smith ND, Gottfries J, Roberts1 TK. Data Set for Metabolomic and Proteomic Responses of Staphylococcus aureus to Prolonged Cold Stress Data in Brief. J Proteomics 2015;121:44–55. Phadtare S. Recent developments in bacterial cold-shock response. Curr Issues Mol Biol 2004;6:125–36. Durack J, Ross T, Bowman JP. Characterisation of the transcriptomes of genetically diverse Listeria monocytogenes exposed to hyperosmotic and low temperature conditions reveal global stress-adaptation mechanisms. PLoS One 2013; 8:e73603. Korobeinikova AV, Garber MB, Gongadze GM. Ribosomal proteins: structure, function, and evolution. Biochem Biokhim 2012;77:562–74. Stelzl U, Sean Connell M, Nierhaus KH, Wittmann-Liebold B. Ribosomal proteins: role in ribosomal functions. Encyclopedia of Life Sci 2001:1–12. Aseev LV, Boni IV. Extraribosomal functions of bacterial ribosomal proteins. Mol Biol 2011;45:805–16. Squires CL, Zaporojets D. Proteins shared by the transcription and translation machines. Annu Rev Microbiol 2000;54: 775–98. Jeffery CJ. Moonlighting proteins: old proteins learning new tricks. Trends Genet 2003;19:415–7. Smith A, McCann M, Kavanagh K. Proteomic analysis of the proteins released from Staphylococcus aureus following exposure to Ag(I). Toxicol in Vitro 2013;27:1644–8. Floden AM, Watt JA, Brissette CA. Borrelia burgdorferi enolase is a surface-exposed plasminogen binding protein. PLoS One 2011;6:e27502.

JO U RN A L OF P ROTE O M ICS 1 21 ( 20 1 5 ) 4 4 –55

[27] Lahteenmaki K, Edelman S, Korhonen TK. Bacterial metastasis: the host plasminogen system in bacterial invasion. Trends Microbiol 2005;13:79–85. [28] Jeffery CJ. Moonlighting proteins—an update. Mol Biosyst 2009;5:345–50. [29] Brandina I, Graham J, Lemaitre-Guillier C, Entelis N, Krasheninnkov I, Sweetlove L, et al. Enolase takes part in a macromolecular complex associated to mitochondria in yeast. Bba Bioenerg 2006;1757:1217–28. [30] Besier S, Zander J, Siegel E, Saum SH, Hunfeld KP, Ehrhart A, et al. Thymidine-dependent Staphylococcus aureus small-colony variants: human pathogens that are relevant not only in cases of cystic fibrosis lung disease. J Clin Microbiol 2008;46:3829–32. [31] Vaudaux P, Francois P, Bisognano C, Kelley WL, Lew DP, Schrenzel J, et al. Increased expression of clumping factor and fibronectin-binding proteins by hemB mutants of Staphylococcus aureus expressing small colony variant phenotypes. Infect Immun 2002;70:5428–37. [32] Schröder A, Kland R, Peschel A, von Eiff C, Aepfelbacher M. Live cell imaging of phagosome maturation in Staphylococcus aureus infected human endothelial cells: small colony variants are able to survive in lysosomes. Med Microbiol Immunol (Berl) 2006;195:185–94. [33] Sendi P, Proctor RA. Staphylococcus aureus as an intracellular pathogen: the role of small colony variants. Trends Microbiol 2009;17:54–8. [34] Sanchez B, Cabo ML, Margolles A, Herrera JJR. A proteomic approach to cold acclimation of Staphylococcus aureus CECT 976 grown at room and human body temperatures. Int J Food Microbiol 2010;144:160–8.

55

[35] Chanda PK, Mondal R, Sau K, Sau S. Antibiotics, arsenate and H2O2 induce the promoter of Staphylococcus aureus cspC gene more strongly than cold. J Basic Microbiol 2009;49:205–11. [36] Cordwell SJ, Larsen MR, Cole RT, Walsh BJ. Comparative proteomics of Staphylococcus aureus and the response of methicillin-resistant and methicillin-sensitive strains to Triton X-100. Microbiol Sgm 2002;148:2765–81. [37] Vidovic S, Mangalappalli-Illathu AK, Korber DR. Prolonged cold stress response of Escherichia coli O157 and the role of rpoS. Int J Food Microbiol 2011;146:163–9. [38] Sianglum W, Srimanote P, Wonglumsom W, Kittiniyom K, Voravuthikunchai SP. Proteome analyses of cellular proteins in methicillin-resistant Staphylococcus aureus treated with rhodomyrtone, a novel antibiotic candidate. PLoS One 2011;6: e16628. [39] de Jonge BL, Chang YS, Gage D, Tomasz A. Peptidoglycan composition of a highly methicillin-resistant Staphylococcus aureus strain. The role of penicillin binding protein 2A. J Biol Chem 1992;267:11248–54. [40] Ghobakhlou A, Laberge S, Antoun H, Wishart DS, Xia J, Krishnamurthy R, et al. Metabolomic analysis of cold acclimation of arctic Mesorhizobium sp. strain N33. PLoS One 2013;8:e84801. [41] Balaban NQ, Merrin J, Chait R, Kowalik L, Leibler S. Bacterial persistence as a phenotypic switch. Science 2004;305:1622–5. [42] Dominguez DC. Calcium signalling in bacteria. Mol Microbiol 2004;54:291–7. [43] Wu CC, Singh P, Chen MC, Zimmerli L. L-Glutamine inhibits beta-aminobutyric acid-induced stress resistance and priming in Arabidopsis. J Exp Bot 2010;61:995–1002.