Global Dynamics of the Escherichia coli ... - American Chemical Society

Article pubs.acs.org/jpr

Global Dynamics of the Escherichia coli Proteome and Phosphoproteome During Growth in Minimal Medium Nelson C. Soares,† Philipp Spaẗ ,† Karsten Krug, and Boris Macek* Proteome Center Tuebingen, University of Tuebingen, Germany S Supporting Information *

ABSTRACT: Recent phosphoproteomics studies have generated relatively large data sets of bacterial proteins phosphorylated on serine, threonine, and tyrosine, implicating this type of phosphorylation in the regulation of vital processes of a bacterial cell; however, most phosphoproteomics studies in bacteria were so far qualitative. Here we applied stable isotope labeling by amino acids in cell culture (SILAC) to perform a quantitative analysis of proteome and phosphoproteome dynamics of Escherichia coli during five distinct phases of growth in the minimal medium. Combining two triple-SILAC experiments, we detected a total of 2118 proteins and quantified relative dynamics of 1984 proteins in all measured phases of growth, including 570 proteins associated with cell wall and membrane. In the phosphoproteomic experiment, we detected 150 Ser/Thr/Tyr phosphorylation events, of which 108 were localized to a specific amino acid residue and 76 were quantified in all phases of growth. Clustering analysis of SILAC ratios revealed distinct sets of coregulated proteins for each analyzed phase of growth and overrepresentation of membrane proteins in transition between exponential and stationary phases. The proteomics data indicated that proteins related to stress response typically associated with the stationary phase, including RpoS-dependent proteins, had increasing levels already during earlier phases of growth. Application of SILAC enabled us to measure median occupancies of phosphorylation sites, which were generally low (75% for a phosphorylation on an S/ T/Y were reported as localized. Furthermore, the fragmentation spectra were manually inspected for good b- and y-ion series coverage. The SILAC ratios of the phosphorylation sites were further normalized to the protein ratios to eliminate quantitation bias due to changing protein expression. The two triple SILAC experiments were merged by using the Lys4 labeled early stationary sample (T4) as common reference point. All SILAC ratios reported for the different growth phases are relative to this common reference point.

Mass Spectrometry

The nano-LC−MS/MS proteome analyses of GeLC and OffGel samples were performed on an EasyLC nano-HPLC (Proxeon Biosystems) coupled to the LTQ Orbitrap Elite mass spectrometer as described previously.37 The peptide mixtures were directly applied onto a 15 cm nano-HPLC column, inhouse packed with reverse-phase 3 μm C18 spheres (Dr. Maisch, Ammerbuch, Germany) at a flow rate of 500 nL/min in 0.5% acetic acid. The peptide elution was achieved in a linear, segmented 90 min gradient of 5−33% of Solvent B (80% acetonitrile in 0.5% acetic acid) at a constant flow rate of 200 nL/min. Separated peptides were ionized in the electrospray ion source (ESI) (Proxeon Biosystems, Odense, Denmark). The mass spectrometer was operated in the positive ion mode with the following acquisition cycle: one initial full scan in the

Clustering Analysis

Similar protein expression profiles were grouped by a hierarchical clustering approach. First we calculated the standard deviations of log2-transformed protein and phosphorylation site profiles across the five growth phases as a measure of their expression change along the growth curve. For cluster analysis we only considered proteins among the top 25% of C

dx.doi.org/10.1021/pr3011843 | J. Proteome Res. XXXX, XXX, XXX−XXX

Journal of Proteome Research

Article

Figure 1. Histograms of detected protein and phosphorylation site ratios in separate time points. Depicted are relative frequencies (y-axis) in bins of log2 transformed ratios (x-axis). Note that all distributions are relative to the time point T4, which was used as a common point for integration of data from two triple-SILAC experiments.

The proteins were blotted (XCell II Blot Module; Invitrogen) onto a PVDF membrane (Roth, Karlsruhe, Germany) and blocked overnight with PBS and 1% Tween at 4 °C. After washing the membrane with PBST, primary antibodies were incubated 1:2000 for 2 h at RT. Affinity purified antibody against GAPDH was from Pierce, Thermo Fisher Scientific and sera against MBP and OmpX were produced in rabbit and kindly provided by Dr. Dirk Linke. After washing, HRP linked secondary antibody (Cell Signaling Technology Inc., Danvers, MA) was incubated 1:5000 for 1 h and proteins were visualized using LumiGLO reagent (Cell Signaling Technology Inc.).

calculated standard deviations to ensure a substantial expression change across the growth phases. Cluster analysis was done using the “hclust” R-function on Z-score transformed profiles using Euclidian distance as metric in combination with the “ward” clustering method. Functional Enrichment Analysis

Gene Ontology (GO) annotation for the E. coli K12 proteome was derived from the UniProt-GOA Database (www. http:// www.ebi.ac.uk/GOA/; downloaded on 29.02.2012). Fisher’s exact test was used to test for enrichment or depletion (twosided test) of specific annotation terms among members of resulting protein clusters. Derived p-values were further adjusted to address multiple hypotheses testing by the method proposed by Benjamini and Hochberg.40 GO terms having adjusted p-values below 0.05 in any of the clusters were treated as significant and corresponding p-values across all clusters were visualized by a heatmap using the “heatmap.2” function from the “gplots” R-package. GO term analysis of detected phosphoproteome was performed in the same way.

■

RESULTS We performed two triple-SILAC experiments to analyze five different time points of the in vitro E. coli growth curve: T1) exponential growth; T2) entry into retardation phase; T3) exit from the retardation phase (transition to stationary phase); T4) early stationary phase; T5) late stationary phase. Exact growth conditions and points of harvesting are described in the experimental section. SILAC Experiment 1 included three differentially labeled cultures that were harvested at three distinct points of the growth curve: T1 (Lys0), T4 (Lys4) and T3 (Lys8); SILAC Experiment 2 included three differentially

Western Blot Analysis

For protein separation, 25 μg crude protein extract from each growth phase were applied to SDS-PAGE as described above. D



Article

Figure 2. Clustering of coregulated protein groups and functional enrichment analysis. Differentially regulated proteins could be clustered into six different profiles. Representative protein functions and pathways are indicated in the boxes below the respective clusters; red arrows signify overrepresentation of corresponding GO terms in respective cluster, blue arrows indicate underrepresentation. Detailed results of GO enrichment analysis are presented in the Supplementary Table 2, Supporting Information.

replicates was high, as evidenced by Pearson correlation coefficients of 0.79−0.95 (Supplementary Figure 5, Supporting Information). Overall, the SILAC quantitation achieved very good biological reproducibility in all phases with pronounced proteome and phosphoproteome dynamics.

labeled cultures that were harvested at the following points: T2 (Lys0), T4 (Lys4) and T5 (Lys8) (Supplementary Figure 1, Supporting Information). The T4 (Lys4) sample was identical in both SILAC experiments and was used for integration of the two data sets (see below). In all cases the incorporation levels of the SILAC amino acid were greater than 95% (Supplementary Figure 2, Supporting Information). In each SILAC experiment, protein extracts were mixed in a ratio 1:1:1 (based on Bradford assay and a pilot MS measurement) and taken for further proteomics and phosphoproteomics analysis. The actual mixing ratios measured by MS are shown in Supplementary Table 1, Supporting Information. For analysis of the proteome we employed two analytical strategies: 1D SDS-PAGE protein separation with subsequent in-gel protein digestion (GeLC− MS) and, separately, isoelectric focusing of in-solution protein digestions (OffGel). For analysis of the phosphoproteome we performed two stages of phosphopeptide enrichment, SCX and TiO2 chromatographies. The complete experimental workflow is depicted in the Supplementary Figure 3, Supporting Information.

Proteome Dynamics

Combined, proteome and phosphoproteome analysis of the two triple-SILAC experiments resulted in identification of 502 088 MS/MS spectra corresponding to 15380 non redundant peptide sequences and 2118 protein groups. The estimated false discovery rate (FDR) was 0.56% at peptide level and 1.1% at protein group level. All detected protein groups and phosphorylation events are presented in the Supplementary Tables 2 and 3, Supporting Information, respectively. Of the detected proteins, 2053 fulfilled the prerequisite of having at least two quantification events (ratio counts) in at least one of the triple-SILAC experiments and were thereby considered as quantified. Of these, 1984 were quantified in all five measured phases of growth (Supplementary Figure 6, Supporting Information). Our experimental strategy included the T4 phase (Lys4) as a shared common point between two SILAC experiments (Figure 1), which enabled us to integrate the two data sets and obtain a complete five time-point expression profile. In order to determine the extent of relative expression changes in proteome and phosphoproteome, we calculated the standard deviation of SILAC ratios of each protein and phosphorylation event in all five measured points of the growth curve. We applied a threshold corresponding to the 25% of proteins and phosphorylation sites showing the highest fluctuation throughout the growth curve (Supplementary Figure 7, Supporting Information). This analysis showed that the fluctuations were more pronounced at the phosphoproteome than at the proteome level, which was also evident from the individual distribution plots of the ratios of all the quantified proteins and phosphorylation events (Figure 1A−D). To assess which groups of proteins are coregulated (show similar expression profiles) during growth we applied hierarchical clustering of protein/phosphoprotein profiles that

Reproducibility of SILAC Quantification

In order to assess reproducibility of SILAC quantification we analyzed the correlation between respective GeLC−MS and OffGel proteome measurements. The calculated Pearson correlation coefficients between these paired measurements were 0.7−0.9. In biological replicates (BR), the Pearson correlation coefficients between distant points in the growth curve were 0.7−0.8; however, the correlation was considerably lower between closer time points (e.g., T4/T3 (BR1) vs T4/ T3(BR2); Pearson correlation = 0.3−0.25) or between points at later stages of growth (e.g., T4/T5 (BR1) vs T4/T5 (BR2); Pearson correlation = 0.2−0.16). This is due to the fact that in some cases the protein regulation between two phases was not pronounced, as discussed later in the text. Correlation plots for paired proteome measurements and biological replicates are presented in the Supplementary Figure 4A−G, Supporting Information. In the phosphoproteome analysis, the correlation between the phosphorylation site ratios measured in two biological E



Article

Figure 3. Profiles of three exemplary proteins detected by mass spectrometry and Western blot. The changes in the protein abundance of E. coli (A) Glyceraldehyde-3-phosphate dehydrogenase A, (B) Outer membrane protein X and (C) Maltose-binding periplasmic protein over five different growth stages measured by mass spectrometry in two biological replicates as well as the corresponding Western blot in three replicates are shown. Each replicate of the blots was done from the same protein master mix, therefore GAPDH serves as a common loading control.

and was enriched in proteins involved in D-galacturonate catabolic process (Figure 2F). To assess the precision of our relative quantitative strategy, we assessed expression profiles of members of several known protein complexes in our data set: ClpAP/XP and HslVU ATPdependent protease complexes; UvrABC nucleotide excision repair complex; and CusCFBA Cation efflux system complex. As expected, most of the members of these complexes showed similar expression profiles throughout the measured growth phases, demonstrating a good precision of our relative quantification strategy (Supplementary Figure 8, Supporting Information). In addition, we validated expression profiles of three exemplary proteins by Western blot: the Glyceraldehyde3-phosphate dehydrogenase A (GAPDH); the outer membrane protein X (OmpX); and the maltose-binding periplasmic protein (MBP). These proteins were selected for Western blot validation according to the following criteria: (a) the antibodies/sera were readily available; (b) the proteins showed different dynamics: unchanging (GAPDH), increasing (OmpX) and decreasing (MBP); (c) dynamics of two proteins was not reported before (OmpX and MPB), whereas expression of GAPDH was known to be stable during growth and was therefore used as a positive control. In all three cases, Western blot analysis was in agreement with the MS quantification (Figure 3).

fulfilled the following requirements: they were quantified in at least three of the stages of growth and they were among 25% of the proteins that showed the highest fluctuation along the growth curve. Hierarchical clustering analysis revealed groups of proteins having similar expression patterns across the growth curve (Figure 2). The enrichment analysis of gene ontology (GO) terms further indicated that all six of the detected clusters could be associated with a specific cellular process and/or localization. Proteins associated with Cluster 1 had generally decreased levels along the growth curve and included cytosolic proteins mainly associated with peptide transport, iron binding, oxidation−reduction and aerobic respiration (Figure 2A). Cluster 2 included cytosolic and periplasmic proteins with increased levels at time points T1 and T2, that were functionally related with metal ion binding, amino acid biosynthesis, general metabolism, ribosome and proteins related to translation and iron−sulfur binding (Figure 2B). Cluster 3 grouped proteins with increased levels at time points T1 and T3; interestingly, this cluster was almost exclusively composed by proteins associated with membrane (Figure 2C). Cluster 4 included proteins that showed drastically increased levels from T1 to T2 phase and slightly decreased levels during T5; the proteins related with iron ion binding and siderophore transport were overrepresented in this cluster (Figure 2D). Cluster 5 included proteins that were continuously increasing along the growth curve and were involved in tryptophan biosynthetic process, lyase activity and biosynthesis of aromatic amino acids (Figure 2E). Finally, proteins grouped to Cluster 6 included proteins with increased levels at later stages of growth

Phosphoproteome Dynamics

In SILAC-based phosphoproteomics experiments we detected a total of 180 phosphorylation events, of which 128 were F



Article

kinases or to an expression-independent mode of kinase regulation. The general increase of phosphorylation levels in the late stationary phase prompted us to test whether we would increase the total number of identified phosphorylation sites by specifically analyzing the E. coli phosphoproteome in late stationary phase. To that end five milligram of total protein was extracted from late stationary unlabeled cells and was subjected to SCX TiO2-based phospho enrichment and phosphoproteomics analysis. A total of 156 localized phosphosites were identified, out of those 95 were newly identified phosphosites (Supplementary Table 4, Supporting Information).

localized to a specific amino acid residue. After manual validation of peptide spectra, 150 phosphorylation events were retained as high confidence, of which 108 were localized on a specific residue: 75.9% on serine (pS), 16.7% on threonine (pT) and 7.4% on tyrosine (pY). To exclude quantitation bias due to possible fluctuations in protein levels, we normalized the SILAC ratios of phosphorylation sites by ratios of corresponding phosphoproteins. After normalization, we grouped the phosphorylation sites according to the number of growth phases in which they were quantified; 76 p-sites were quantified in all five time points (Supplementary Figure 9, Supporting Information). Importantly, the majority of phosphorylation sites had significantly increased levels during later stages of growth, especially in the stationary phase. Application of SILAC enabled us to calculate occupancy of those phosphorylation sites for which phosphopeptide ratio, unmodified counterpart peptide ratio and corresponding unmodified protein ratio were obtained.41 Despite of the general increase in later growth phases, the measured occupancy of phosphorylation sites remained low in all measured time points. The median occupancy values in separate time points were: T1 = 4.4%; T2 = 6.7%; T3 = 8.9%; T4 = 7.6%; and T5 = 11.5%. The low measured occupancy of S/T/Y phosphorylation sites is in agreement with observation from previous qualitative mass spectrometry studies in E. coli (and other bacteria), where even spectra of phosphopeptides from abundant proteins had in general a low intensity.15 The GO enrichment analysis revealed that the measured E. coli phosphoproteome included both cytoplasmic as well as membrane proteins associated with protein binding, translation and translation elongation factor activity and response a variety of stresses (Figure 4; Supplementary Table 3, Supporting Information). Interestingly, the levels of five known S/T/Y protein kinases quantified in our data set did not change significantly (Supplementary Figure 10, Supporting Information), pointing to the action of other

■

DISCUSSION

SILAC Is Routinely Applicable in Quantitative Proteomics Analyses in E. coli

Although SILAC has been successfully used to label E. coli proteome,42−46 this study presents its first systematic application to quantitative analysis proteome and phosphoproteome dynamics in different phases of E. coli growth. We measured high (>95%) incorporation rates of heavy-labeled lysine in the wild type E. coli strain and chose to use this strain to avoid any potential influence of lysine deletion mutant on physiological processes during growth. We showed high reproducibility of SILAC quantification in biological replicates, especially in distant time points that were characterized with pronounced changes in proteome dynamics. As expected, closer time points showed less pronounced changes and resulted in poorer correlation of SILAC ratios. In addition, late time points (T4 and T5) also showed lower correlation, pointing to increased heterogeneity of the cell population and deregulation of protein levels at these stages which may be attributed to increased protein degradation in the late stationary phase.47 Overall, this study resulted in one of the largest quantitative (phospho)proteomic data sets reported in bacteria thus far. Growth in Glucose Minimal Medium Requires Differential Regulation of Stress-related Proteins Throughout the Growth Curve

The induction of stress responsive proteins is commonly associated with later stages of growth. For example, in E. coli the induction of stress responsive proteins in the late stationary is coordinated by transcription factor RpoS.29,48−50 Here we have identified several RpoS dependent proteins that showed increasing expression already at early stages of growth, namely at time points T1 and/or T2. This is in agreement with previous reports, which indicate that the regulation of RpoSdependent proteins depends on cell culture conditions, specifically on the carbon source avaibility.51,52 Cells in minimal medium have to synthesize their building blocks from a single carbon source, which leads to intracellular accumulation and excretion of metabolic compounds such as acetate. The extracellular accumulation of this compound lowers the external pH and consequently induces acid stress.52 RpoS-dependent genes comprise a system for acid tolerance, which are induced in the exponential stage in cells growing in the minimal medium.52 In concordance, here we detected two acidic resistance chaperones HdeA and HdeB apparently coded by the same RpoS-dependent operon,51 that are induced in the early stages of growth curve. Specifically, HdeA peaked at T2 at which the protein abundance was 5-fold higher than that observed in T3 or T5, similarly HdeB peaked at T2 at which the protein abundance was 3.6-fold higher than that detected in

Figure 4. Global dynamics of Ser/Thr/Tyr phosphoproteome. All 76 phosphorylation sites quantified in the five stages of the growth showed a general increase in the late stationary phase. Main functional classes overrepresented in the GO enrichment analysis are indicated in the box and presented in Supplementary Table 3, Supporting Information. G



Article

T5 (Supplementary Table 2, Supporting Information). However, it is interesting to note that an acid resistance membrane protein HdeD, also described as RpoS-dependent protein,51 that gradually increased along growth and peaking at T5 where the protein abundance was 3.8-fold higher than that detected at T1, most likely as a response to further decrease in pH, which in our experiment dropped from 7.2 during inoculation to 5.8 during the late stationary phase (Supplementary Figure 1, Supporting Information). Our results also revealed early induction of RpoS-dependent proteins related with carbon starvation stress, such as Carbon starvation A53 (peaking as early as T1 and decreasing by 3-fold at T5) and protein CsiD54 (peaking at T2 and decreasing by 4-fold at T3/ T4). Similar to what was described with acid-resistance proteins, transition to stationary phase required the induction of new set of carbon stress related proteins; for instance, in our data set the starvation-sensing proteins RspA and RspB peaked at T5 and T4, respectively, and phosphate starvation-inducible protein psiF was 3.6-fold more abundant in T5 than T1. Overall, our results indicate a coordinated interplay between stress proteins in the exponential and late stationary phase of the E. coli growth.

This RpoS-regulated protein has been implicated in the development of increased cell envelope resilience and enhanced pressure resistance in stationary phase49 and it has been hypothesized that it may cross-link the outer membrane and peptidoglycan and thereby restrain the growth rate.57 Therefore, the measured dynamics of these proteins in our data set is in full agreement with previous data on their expression and physiology. General Increase of Phosphorylation Levels in the Late Stationary Phase

As mentioned earlier, as the cells enter into the stationary phase, individual cells become gradually exposed to a hostile environment, including lack of nutrients, oxygen depletion, accumulation of toxic products and lower pH. Endurance of bacteria at this stage heavily relies on the promptness and suitability of the response to external stimuli. Many of these responses are mediated by signal transduction networks that include protein phosphorylation.4 Our results clearly indicate a general increase of phosphorylation levels in the later stages of growth, with a maximum in the late stationary phase (T5). This trend is not only related to detection of a higher number of phosphorylation sites but also to increased occupancy of quantified sites. However, we note that the measured median occupancy of S/T/Y phosphorylation sites is considerably lower than in eukaryotes,41 pointing to potentially different regulatory capacity and mode of action of this modification in bacteria. Phosphorylation of ribosomal proteins was previously detected in E. coli,15,59 Bacillus subtilis,14 Pseudomonas putida,17 Mycobecterium tuberculosis,21 Lactococcus lactis19 and Sreptomyces coelicolor,6 suggesting that S/T/Y phosphorylation may be involved in regulation of translation across bacterial species. Interestingly, our results revealed that proteins related to translation (e.g., small subunits (30S) S1, S2, S7; large subunit (50) L7/L12, L9, L19), together with several elongation factors (Tu, Ts, G), had in general decreased levels in later phases of growth, whereas at the same time their phosphorylation sites had increased levels. This suggests that S/T/Y phosphorylation may provide a mechanism for coordinating translation during bacterial adaptation to changing environmental conditions. For instance, it has been demonstrated that Lys-9 and/or Arg-12 in the N-terminal of EF-Ts are crucial in the interaction with EFTu.60 Here we report that N-terminal Ser-6 is increasingly phosphorylated upon entering the stationary phase and thus can likely interfere with EF-Tu-Ts complex formation, essential for peptide elongation cycle. It was recently demonstrated in Mycobacterium that phosphorylation of EF-Tu by PKnB reduced its interaction with GTP, resulting in a consequent reduction in the level of protein synthesis.61 The same study showed that Mycobacterium EF-Tu was found to be phosphorylated by PKnB on multiple sites, including Thr118, which is required for optimal activity of the protein, suggesting that protein phosphorylation may serve different functions on different sites of the EF-Tu. In our study, we detected two phosphorylation sites on EF-Tu with opposite dynamics: Thr-39 was increased only during T2, whereas Ser313 was steadily increasing throughout growth curve and peaked in the late stationary phase (T5). Considering the differences between T2 and T5 in terms of cell division and protein translation rate, it would of interest to further investigate the impact of these phosphorylation events on overall protein synthesis.

Exponential to Stationary Phase Transition Is Characterized by Changes at Membrane Level

Changes occurring during transition from exponential to stationary phase (in our data set time point T3) are of great interest, since this time point is characterized by complex processes of adaptation including morphological and molecular changes, which ultimately determine cell population growth rate. However, quantitative information concerning protein expression in these growth points was so far very scarce. Despite of overall underrepresentation of membrane proteins in our data set, which was expected due to their biophysical properties, bioinformatic analysis of the proteome data set revealed that one of the clusters of coregulated proteins (Cluster 3, Figure 2C) was composed of a significant number of membrane proteins39 out of 71 proteins in this cluster were associated with the membrane. Interestingly, this cluster contained proteins with expression maxima in T1 and T3 points, indicating that important changes at the membrane level take place during these distinct phases of growth. Cluster 3 included several permeases and membrane transporters, which will inevitably influence membrane permeability. For instance, we detected increased levels of several proteins involved in the uptake of amino acids (e.g., Serine transporter, Arginine ABC transporter permease, Threonine/serine transporter TdcC). Considering that bacterial cells were inoculated in a minimal media containing a single amino acid (lysine), it is understandable that the lack of amino acids in the medium would trigger the uptake of alternative amino acids. We also observed a differential induction of membrane proteins involved in regulation of osmolarity and growth. For example, metalloprotease YggG had increased levels in early time points T1 and T2 (Cluster 2) and decreased by nearly 7-fold at T5 (Cluster 2). It was reported that the yggG gene is induced when E. coli is subjected to media of low osmolarity55 and it has been suggested that its expression highly improves growth of stressed E. coli cells.56 Conversely, the osmotically inducible lipoprotein B (OsmB) had increased levels by 3-fold in later time points T3-T5 (Cluster 6). It was previously reported that osmB expression is induced by elevated osmolarity57,58 and environmental factors present during stationary phase of growth.57 H



Article

discussions and help in manuscript preparation. We thank the Juniorprofessoren-Programm of the Landesstiftung BW and the SFB766 of the German Research Council (DFG) for financial support. Nelson C. Soares is financed by the postdoctoral program Angeles Alvariño Dez 2009- Xunta de Galicia, Spain. We also thank PD Dr. Dirk Linke for the provision of the OmpX and MBP antibodies.

We detected phosphorylation on several proteins related to stress response, including oxidative stress, protein folding and starvation, whose phosphorylation increased in the late stationary phase (T5). For instance, our results indicate that stringent starvation protein A (SspA) is increasingly phosphorylated as the culture goes into stationary phase, although its protein level stays constant. SspA has been described as an RNA polymerase-associated protein that is induced during stationary phase as well as upon carbon starvation. It has been demonstrated that SspA is required for acid resistance by inhibiting stationary-phase accumulation of H-NS, a global regulator gene related with a variety of environmental stresses.62 However, the exact mechanism of that regulation is not known. Of proteins related to oxidative stress, we detected phosphorylation on MnSOD (pSer-2 and pSer-23); Thiol peroxidase (pSer17); Alkyl hydroperoxide reductase protein C22 (pSer-180); Peroxiredoxin osmC (pSer-112); Uncharacterized oxidoreductase ydgJ (pSer-38), that were all phosphorylated on a Ser residue and had phosphorylation levels increased by at least 5-fold at T5 in comparison to T4. It has been shown that reactive oxygen species (ROS) accumulate during late stages of growth in batch culture and that late stationary cells of several bacterial species are more protected against oxidative stress.34 Interestingly, it was shown in Listeria monocytogenes that the cytoplasmic MnSOD can be post-translationally controlled by phosphorylation.63 In this context, it will be important to further investigate whether the phosphorylation events detected in our study can influence the regulation of stress response in bacteria.

■

(1) Mijakovic, I.; Macek, B. Impact of phosphoproteomics on studies of bacterial physiology. FEMS Microbiol. Rev. 2011, 36 (4), 877−92. (2) Soufi, B.; Soares, N. C.; Ravikumar, V.; Macek, B. Proteomics reveals evidence of cross-talk between protein modifications in bacteria: focus on acetylation and phosphorylation. Curr. Opin. Microbiol. 2012, 15 (3), 357−63. (3) Macek, B.; Mijakovic, I. Site-specific analysis of bacterial phosphoproteomes. Proteomics 2011, 11 (15), 3002−11. (4) Kobir, A.; Shi, L.; Boskovic, A.; Grangeasse, C.; Franjevic, D.; Mijakovic, I. Protein phosphorylation in bacterial signal transduction. Biochim. Biophys. Acta: Gen. Subj. 2011, 1810 (10), 989−94. (5) Petranovic, D.; Michelsen, O.; Zahradka, K.; Silva, C.; Petranovic, M.; Jensen, P. R.; Mijakovic, I. Bacillus subtilis strain deficient for the protein-tyrosine kinase PtkA exhibits impaired DNA replication. Mol. Microbiol. 2007, 63 (6), 1797−805. (6) Manteca, A.; Ye, J.; Sánchez, J.; Jensen, O. N. Phosphoproteome analysis of Streptomyces development reveals extensive protein phosphorylation accompanying bacterial differentiation. J. Proteome Res. 2011, 10 (12), 5481−92. (7) Shah, I. M.; Laaberki, M.-H.; Popham, D. L.; Dworkin, J. A eukaryotic-like Ser/Thr kinase signals bacteria to exit dormancy in response to peptidoglycan fragments. Cell 2008, 135 (3), 486−96. (8) Klein, G.; Dartigalongue, C.; Raina, S. Phosphorylation-mediated regulation of heat shock response in Escherichia coli. Mol. Microbiol. 2003, 48 (1), 269−85. (9) Lacour, S.; Bechet, E.; Cozzone, A. J.; Mijakovic, I.; Grangeasse, C. Tyrosine phosphorylation of the UDP-Glucose dehydrogenase of Escherichia coli. Is at the crossroads of colanic acid synthesis and polymyxin resistance. PLoS ONE 2008, 3 (8), e3053. (10) Molle, V.; Kremer, L. Division and cell envelope regulation by Ser/Thr phosphorylation: Mycobacterium shows the way. Mol. Microbiol. 2010, 75 (5), 1064−77. (11) Ge, R.; Shan, W. Bacterial phosphoproteomic analysis reveals the correlation between protein phosphorylation and bacterial pathogenicity. Genomics, Proteomics Bioinform. 2011, 9 (4−5), 119−27. (12) Morona, J. K.; Miller, D. C.; Morona, R.; Paton, J. C. The effect that mutations in the conserved capsular polysaccharide biosynthesis genes cpsA, cpsB, and cpsD have on virulence of Streptococcus pneumoniae. J. Infect. Dis. 2004, 189 (10), 1905−13. (13) Schumacher, M. A.; Piro, K. M.; Xu, W.; Hansen, S.; Lewis, K.; Brennan, R. G. Molecular mechanisms of HipA-mediated multidrug tolerance and Its neutralization by HipB. Science 2009, 323 (5912), 396−401. (14) Macek, B.; Mijakovic, I.; Olsen, J. V.; Gnad, F.; Kumar, C.; Jensen, P. R.; Mann, M. The Serine/Threonine/Tyrosine phosphoproteome of the model bacterium Bacillus subtilis. Mol. Cell. Proteomics 2007, 6 (4), 697−707. (15) Macek, B.; Gnad, F.; Soufi, B.; Kumar, C.; Olsen, J. V.; Mijakovic, I.; Mann, M. Phosphoproteome analysis of E. coli reveals evolutionary conservation of bacterial Ser/Thr/Tyr phosphorylation. Mol. Cell. Proteomics 2008, 7 (2), 299−307. (16) Schmidl, S. R.; Gronau, K.; Pietack, N.; Hecker, M.; Becher, D.; Stü l ke, J. The phosphoproteome of the minimal bacterium Mycoplasma pneumoniae. Mol. Cell. Proteomics 2010, 9 (6), 1228−42. (17) Ravichandran, A.; Sugiyama, N.; Tomita, M.; Swarup, S.; Ishihama, Y. Ser/Thr/Tyr phosphoproteome analysis of pathogenic and non-pathogenic Pseudomonas species. Proteomics 2009, 9 (10), 2764−75.

■

CONCLUSIONS Here we present the most comprehensive quantitative proteome and phosphoproteome analysis of bacterial growth to date. We measure relative dynamics of 1984 E. coli proteins and 76 phosphorylation events in five distinct points of growth. Our study indicates that induction of several stress-related proteins, such as RpoS, occurs already in early stages of growth in E. coli cells grown in glucose minimal medium. Our quantitative phosphoproteome data point to a general increase of phosphorylation levels during late stages of growth, implicating that this post-translational modification may play in processes related to bacterial adaptation to adverse surroundings.

■

ASSOCIATED CONTENT

S Supporting Information *

Supplementary figures and tables. This material is available free of charge via the Internet at http://pubs.acs.org.

■

REFERENCES

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Phone: +49/(0)7071/ 29-70556. Fax: +49/(0)7071/29-5779. Author Contributions †

N.C.S. and P.S. contributed equally to this work.

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We thank Prof. Karl Forchhammer, Dr. Boumediene Soufi and the members of the Proteome Center Tuebingen for fruitful I



Article

(18) Parker, J. L.; Jones, A. M. E.; Serazetdinova, L.; Saalbach, G.; Bibb, M. J.; Naldrett, M. J. Analysis of the phosphoproteome of the multicellular bacterium Streptomyces coelicolor A3(2) by protein/ peptide fractionation, phosphopeptide enrichment and high-accuracy mass spectrometry. Proteomics 2010, 10 (13), 2486−97. (19) Soufi, B.; Gnad, F.; Jensen, P. R.; Petranovic, D.; Mann, M.; Mijakovic, I.; Macek, B. The Ser/Thr/Tyr phosphoproteome of Lactococcus lactis IL1403 reveals multiply phosphorylated proteins. Proteomics 2008, 8 (17), 3486−93. (20) Ge, R.; Sun, X.; Xiao, C.; Yin, X.; Shan, W.; Chen, Z.; He, Q.-Y. Phosphoproteome analysis of the pathogenic bacterium Helicobacter pylori reveals over-representation of tyrosine phosphorylation and multiply phosphorylated proteins. Proteomics 2011, 11 (8), 1449−61. (21) Prisic, S.; Dankwa, S.; Schwartz, D.; Chou, M. F.; Locasale, J. W.; Kang, C.-M.; Bemis, G.; Church, G. M.; Steen, H.; Husson, R. N. Extensive phosphorylation with overlapping specificity by Mycobacterium tuberculosis serine/threonine protein kinases. Proc. Natl. Acad. Sci. 2010, 107 (16), 7521−6. (22) Hempel, K.; Pané-Farré, J.; Otto, A.; Sievers, S.; Hecker, M.; Becher, D. r. Quantitative cell surface proteome profiling for SigBdependent protein expression in the human pathogen Staphylococcus aureus via biotinylation approach. J. Proteome Res. 2010, 9 (3), 1579− 90. (23) Ong, S.-E.; Blagoev, B.; Kratchmarova, I.; Kristensen, D. B.; Steen, H.; Pandey, A.; Mann, M. Stable isotope labeling by amino acids in cell culture, SILAC, as a simple and accurate approach to expression proteomics. Mol. Cell. Proteomics 2002, 1 (5), 376−86. (24) Soufi, B.; Kumar, C.; Gnad, F.; Mann, M.; Mijakovic, I.; Macek, B. Stable isotope labeling by amino acids in cell culture (SILAC) applied to quantitative proteomics of Bacillus subtilis. J. Proteome Res. 2010, 9 (7), 3638−46. (25) Macek, B.; Mann, M.; Olsen, J. V. Global and site-specific quantitative phosphoproteomics: principles and applications. Annu. Rev. Pharmacol. Toxicol. 2009, 49, 199−221. (26) Schreiber, T. B.; Mäusbacher, N.; Breitkopf, S. B.; GrundnerCulemann, K.; Daub, H. Quantitative phosphoproteomics − an emerging key technology in signal-transduction research. Proteomics 2008, 8 (21), 4416−32. (27) Olsen, J. V.; Blagoev, B.; Gnad, F.; Macek, B.; Kumar, C.; Mortensen, P.; Mann, M. Global, in vivo, and site-specific phosphorylation dynamics in signaling networks. Cell 2006, 127 (3), 635−48. (28) Pan, C.; Gnad, F.; Olsen, J. V.; Mann, M. Quantitative phosphoproteome analysis of a mouse liver cell line reveals specificity of phosphatase inhibitors. Proteomics 2008, 8 (21), 4534−46. (29) Nystrom, T. Stationary-phase physiology. Annu. Rev. Microbiol. 2004, 58 (1), 161. (30) Hengge-Aronis, R. Signal transduction and regulatory mechanisms involved in control of the σS (RpoS) subunit of RNA polymerase. Microbiology and Molecular Biology Reviews 2002, 66 (3), 373−95. (31) Jers, C.; Kobir, A.; Søndergaard, E. O.; Jensen, P. R.; Mijakovic, I. Bacillus subtilis two-component system sensory kinase DegS is regulated by Serine phosphorylation in its input domain. PLoS ONE 2011, 6 (2), e14653. (32) Bernhardt, J.; Weibezahn, J.; Scharf, C.; Hecker, M. Bacillus subtilis during feast and famine: visualization of the overall regulation of protein synthesis during glucose starvation by proteome analysis. Genome Res. 2003, 13 (2), 224−37. (33) Lee, K.-J.; Bae, S.-M.; Lee, M.-R.; Yeon, S.-M.; Lee, Y.-H.; Kim, K.-S. Proteomic analysis of growth phase-dependent proteins of Streptococcus pneumoniae. Proteomics 2006, 6 (4), 1274−82. (34) Soares, N. C.; Cabral, M. P.; Gayoso, C.; Mallo, S.; RodriguezVelo, P.; Fernández-Moreira, E.; Bou, G. Associating growth-phaserelated changes in the proteome of Acinetobacter baumannii with increased resistance to oxidative stress. J. Proteome Res. 2010, 9 (4), 1951−64. (35) Monod, J. The growth of bacterial cultures. Annu. Rev. Microbiol. 1949, 3 (1), 371−94.

(36) Ishihama, Y.; Rappsilber, J.; Mann, M. Modular stop and go extraction tips with stacked disks for parallel and multidimensional peptide fractionation in proteomics. J. Proteome Res. 2006, 5 (4), 988− 94. (37) Franz-Wachtel, M.; Eisler, S. A.; Krug, K.; Wahl, S.; Carpy, A.; Nordheim, A.; Pfizenmaier, K.; Hausser, A.; Macek, B. Global detection of protein kinase D-dependent phosphorylation events in nocodazole-treated human cells. Mol. Cell. Proteomics 2012, 11 (5), 160−70. (38) Olsen, J. V.; de Godoy, L. M. F.; Li, G.; Macek, B.; Mortensen, P.; Pesch, R.; Makarov, A.; Lange, O.; Horning, S.; Mann, M. Parts per million mass cccuracy on an orbitrap mass spectrometer via lock mass injection into a C-trap. Mol. Cell. Proteomics 2005, 4 (12), 2010−21. (39) Cox, J.; Matic, I.; Hilger, M.; Nagaraj, N.; Selbach, M.; Olsen, J. V.; Mann, M. A practical guide to the MaxQuant computational platform for SILAC-based quantitative proteomics. Nat. Protocols 2009, 4 (5), 698−705. (40) Benjamini, Y.; Hochberg, Y. Controlling the false discovery rate - a practical and powerful approach to multiple testing. J. R. Stat. Soc. Ser. B: Methodol. 1995, 57 (1), 289−300. (41) Olsen, J. V.; Vermeulen, M.; Santamaria, A.; Kumar, C.; Miller, M. L.; Jensen, L. J.; Gnad, F.; Cox, J.; Jensen, T. S.; Nigg, E. A.; Brunak, S.; Mann, M. Quantitative phosphoproteomics reveals widespread full phosphorylation site occupancy during mitosis. Sci. Signal. 2010, 3 (104), ra3. (42) Larance, M.; Bailly, A. P.; Pourkarimi, E.; Hay, R. T.; Buchanan, G.; Coulthurst, S.; Xirodimas, D. P.; Gartner, A.; Lamond, A. I. Stableisotope labeling with amino acids in nematodes. Nat. Methods 2011, 8 (10), 849−51. (43) Matic, I.; Jaffray, E. G.; Oxenham, S. K.; Groves, M. J.; Barratt, C. L. R.; Tauro, S.; Stanley-Wall, N. R.; Hay, R. T. Absolute SILACcompatible expression strain allows Sumo-2 copy number determination in clinical samples. J. Proteome Res. 2011, 10 (10), 4869−75. (44) Neher, S. B.; Villén, J.; Oakes, E. C.; Bakalarski, C. E.; Sauer, R. T.; Gygi, S. P.; Baker, T. A. Proteomic profiling of ClpXP substrates after DNA damage reveals extensive instability within SOS regulon. Mol. Cell 2006, 22 (2), 193−204. (45) Waanders, L. F.; Hankes, S.; Mann, M. Top-down quantitationand characterization of SILAC-labeled proteins. J. Am. Soc. Mass Spectrom. 2007, 18 (11), 2058−64. (46) Zeiler, M.; Straube, W. L.; Lundberg, E.; Uhlen, M.; Mann, M. A Protein epitope signature tag (PrEST) library allows SILAC-based absolute quantification and multiplexed determination of protein copy numbers in cell lines. Mol. Cell. Proteomics 2012, 11 (3), No. O111.009613. (47) Kuroda, A.; Nomura, K.; Ohtomo, R.; Kato, J.; Ikeda, T.; Takiguchi, N.; Ohtake, H.; Kornberg, A. Role of inorganic polyphosphate in promoting ribosomal protein degradation by the Lon protease in E. coli. Science 2001, 293 (5530), 705−8. (48) Ishihama, A. Functional modulation of Escherichia coli RNA polymerase. Annu. Rev. Microbiol. 2000, 54 (1), 499−518. (49) Charoenwong, D.; Andrews, S.; Mackey, B. Role of rpoS in the development of cell envelope resilience and pressure resistance in stationary-phase Escherichia coli. Appl. Environ. Microbiol. 2011, 77 (15), 5220−9. (50) Battesti, A.; Majdalani, N.; Gottesman, S. The RpoS-mediated general stress response in Escherichia coli*. Annu. Rev. Microbiol. 2011, 65 (1), 189−213. (51) Dong, T.; Schellhorn, H. Control of RpoS in global gene expression of Escherichia coli in minimal media. Molecular Genetics and Genomics 2009, 281 (1), 19−33. (52) Tao, H.; Bausch, C.; Richmond, C.; Blattner, F. R.; Conway, T. Functional Genomics: Expression analysis of Escherichia coli growing on minimal and rich media. J. Bacteriol. 1999, 181 (20), 6425−40. (53) Schultz, J. E.; Matin, A. Molecular and functional characterization of a carbon starvation gene of Escherichia coli. J. Mol. Biol. 1991, 218 (1), 129−40. (54) Marschall, C.; Labrousse, V.; Kreimer, M.; Weichart, D.; Kolb, A.; Hengge-Aronis, R. Molecular analysis of the regulation of csiD, a J



Article

carbon starvation-inducible gene in Escherichia coli that is exclusively dependent on σS and requires activation by cAMP-CRP. J. Mol. Biol. 1998, 276 (2), 339−53. (55) Lutticke, C.; Hauske, P.; Lewandrowski, U.; Sickmann, A.; Kaiser, M.; Ehrmann, M. E. coli LoiP (YggG), a metalloprotease hydrolyzing Phe-Phe bonds. Mol. BioSyst. 2012, 8 (6), 1775−82. (56) Huang, Y.; Zhang, B.; Dong, K.; Zhang, X.; Hou, L.; Wang, T.; Chen, N.; Chen, S. Up-regulation of yggG promotes the survival of Escherichia coli cells containing Era-1 mutant protein. FEMS Microbiol. Lett. 2007, 275 (1), 8−15. (57) Jung, J. U.; Gutierrez, C.; Martin, F.; Ardourel, M.; Villarejo, M. Transcription of osmB, a gene encoding an Escherichia coli lipoprotein, is regulated by dual signals. Osmotic stress and stationary phase. J. Biol. Chem. 1990, 265 (18), 10574−81. (58) Jung, J. U.; Gutierrez, C.; Villarejo, M. R. Sequence of an osmotically inducible lipoprotein gene. J. Bacteriol. 1989, 171 (1), 511−20. (59) Soung, G. Y.; Miller, J. L.; Koc, H.; Koc, E. C. Comprehensive analysis of phosphorylated proteins of Escherichia coli ribosomes. J. Proteome Res. 2009, 8 (7), 3390−402. (60) Zhang, Y.; Yu, N.-J.; Spremulli, L. L. Mutational analysis of the roles of residues in Escherichia coli elongation factor Ts in the interaction with Elongation Factor Tu. J. Biol. Chem. 1998, 273 (8), 4556−62. (61) Sajid, A.; Arora, G.; Gupta, M.; Singhal, A.; Chakraborty, K.; Nandicoori, V. K.; Singh, Y. Interaction of Mycobacterium tuberculosis elongation factor Tu with GTP is regulated by phosphorylation. J. Bacteriol. 2011, 193 (19), 5347−58. (62) Hansen, A.-M.; Qiu, Y.; Yeh, N.; Blattner, F. R.; Durfee, T.; Jin, D. J. SspA is required for acid resistance in stationary phase by downregulation of H-NS in Escherichia coli. Mol. Microbiol. 2005, 56 (3), 719−34. (63) Archambaud, C.; Nahori, M. A.; Pizarro-Cerda, J.; Cossart, P.; Dussurget, O. Control of Listeria superoxide dismutase by phosphorylation. J. Biol. Chem. 2006, 281 (42), 31812−22.

K


Global Dynamics of the Escherichia coli ... - American Chemical Society

Recommend Documents