MCP
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/mcp.M600282-MCP200 on January 16, 2007.
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Data
Right arrow All Versions of this Article:
M600282-MCP200v1
6/4/717    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Glossary
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Manes, N. P.
Right arrow Articles by Heffron, F.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Manes, N. P.
Right arrow Articles by Heffron, F.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?
Molecular & Cellular Proteomics 6:717-727, 2007.
© 2007 by The American Society for Biochemistry and Molecular Biology, Inc.

Research

Targeted Protein Degradation by Salmonella under Phagosome-mimicking Culture Conditions Investigated Using Comparative Peptidomics*,S

Nathan P. Manes{ddagger}, Jean K. Gustin§, Joanne Rue§, Heather M. Mottaz, Samuel O. Purvine, Angela D. Norbeck{ddagger}, Matthew E. Monroe{ddagger}, Jennifer S. D. Zimmer{ddagger}, Thomas O. Metz{ddagger}, Joshua N. Adkins{ddagger}, Richard D. Smith{ddagger} and Fred Heffron§,||

From the {ddagger} Fundamental Science Division, Environmental Molecular Science Laboratory, Pacific Northwest National Laboratory, Richland, Washington 99352 and § Department of Molecular Microbiology and Immunology, Oregon Health and Sciences University, Portland, Oregon 97239


   ABSTRACT
TOP
 ABSTRACT
EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
The pathogen Salmonella enterica is known to cause both food poisoning and typhoid fever. Because of the emergence of antibiotic-resistant isolates and the threat of bioterrorism (e.g. contamination of the food supply), there is a growing need to study this bacterium. In this investigation, comparative peptidomics was used to study S. enterica serovar Typhimurium cultured in either a rich medium or in an acidic, low magnesium, and minimal nutrient medium designed to roughly mimic the macrophage phagosomal environment (within which Salmonella are known to survive). Native peptides from cleared cell lysates were enriched by using isopropanol extraction and analyzed by using both LC-MS/MS and LC-FTICR-MS. We identified and quantified 5,163 peptides originating from 682 proteins, and the data clearly indicated that compared with Salmonella cultured in the rich medium, cells cultured in the phagosome-mimicking medium had dramatically higher abundances of a wide variety of protein degradation products, especially from ribosomal proteins. Salmonella from the same cultures were also analyzed using traditional, bottom-up proteomic methods, and when the peptidomics and proteomics data were analyzed together, two clusters of proteins targeted for proteolysis were tentatively identified. Possible roles of targeted proteolysis by phagocytosed Salmonella are discussed.

A wide variety of physiological processes require and/or are regulated by bioactive peptides (1–3). However, comprehensive analyses of the peptide complement (i.e. the peptidome, which is generally restricted to peptides <100 amino acids or <10 kDa) of microorganisms and of cells from higher organisms have proven difficult, as evidenced by the limited application of peptidomics reported in the literature. Bottom-up proteomics has greatly benefited from recent advances in high throughput liquid chromatography coupled with mass spectrometry (LC-MS and LC-MS/MS) (4, 5), yet peptidomics remains an emerging field of study. In part, this is because peptidomic masses are generally a small fraction of their corresponding proteomic masses, and substantial sample losses can be incurred during isolation of the peptide fraction. However, peptidomics studies have the potential to identify short amino acid sequences (e.g. tiny proteins, protein fragments, and bioactive peptides) that bottom-up studies might miss. In addition, choosing to ignore the entire peptidome introduces the possibility that native peptides are affecting bottom-up proteomics analyses. For example, because tryptic peptides derived from intact and fragmented forms of the same protein are indistinguishable, targeted proteolysis in vivo could pass undetected. In contrast, a peptidomics study could potentially identify these protein fragments. Therefore, besides including potentially bioactive components, this set of biopolymers can provide important insights into cellular processes such as targeted proteolysis.

Because many peptides are known to have important roles during host-pathogen interactions (e.g. antimicrobial peptides, pro-inflammatory peptides, and bacterial peptide toxins), we expanded our proteomics investigation of Salmonella pathogenesis (6) to include a peptidomic component. The Gram-negative bacterium Salmonella enterica serovar Typhimurium (S. typhimurium) is the major causative agent of salmonellosis (food poisoning) in ~7,000 cases per year in the United States (7). The closely related strain S. enterica serovar Typhi causes typhoid fever, which results in ~600,000 deaths per year worldwide (8). In response to the emergence of antibiotic-resistant isolates (9) and the threat of bioterrorism (e.g. the contamination of the food supply by bioterrorists in Oregon in 1984 (10)), we are applying high throughput techniques to identify the proteins required for Salmonella pathogenesis. Key to this effort is the premise that expression levels of bacterial genes required for virulence are disproportionately increased during infection. As the survival of phagocytosed Salmonella is an important element of its pathophysiology, we focused our investigation on this stage of infection. Validation of putative virulence factors will require orthogonal experimentation (e.g. pathogenicity assays of Salmonella knockouts).

Despite the fact that peptides are generally incapable of catalysis, binding, transport, or other protein functions, they have enormous functional diversity (1). Unfortunately, their physiological roles are often difficult to identify, and most techniques designed to do so are low throughput. This is further exacerbated by the fact that the majority of peptides are bioinactive byproducts of protein turnover. However, if the above premise is correct, then the putative identification of bioactive peptides may be possible via parallel comparative proteomics and peptidomics investigations. In this article we describe the use of high throughput LC-MS and LC-MS/MS to identify and quantify thousands of peptides generated by S. typhimurium grown in host-free culture conditions designed to mimic the environment of the phagosome. As such, it constitutes the first global description of a bacterial peptidome. Possible physiological roles for subsets of the Salmonella peptidome are discussed.


   EXPERIMENTAL PROCEDURES
TOP
 ABSTRACT
 EXPERIMENTAL PROCEDURES
RESULTS
 DISCUSSION
 REFERENCES
 
Reagents—
Water was purified using a NANOpure® system (≥18 m{Omega}xcm, Barnstead International, Dubuque, IA). Reagents were obtained from Sigma unless otherwise specified and included acetic acid, acetone, acetonitrile, ammonium bicarbonate, isopropanol, methanol, and trifluoroacetic acid.

Cell Culture—
Wild type S. enterica serovar typhimurium strains 14028 and LT2 were grown to logarithmic (Log)1 phase, stationary (Stat) phase, magnesium minimal medium- (MgM-) shocked stationary phase, and linear in N-Salts medium phase and then harvested following standard batch culture techniques (11). N-Salts medium consisted of 100 mm Bis-Tris pH 5, 0.5 mm KCl, 7.5 mm (NH4)2SO4, 0.5 mm K2SO4, 1 mm KH2PO4, 0.2% (v/v) glycerol, 0.1% (w/v) casamino acids, 8 µm MgCl2, and was similar to MgM pH 5 (12) but with less potassium (the concentration and role of potassium in the phagosome is the area of active investigation (13, 14)). Cultures were initiated by inoculating 5 ml of either LB (Luria-Bertani) broth (Log, Stat, and MgM samples) or N-Salts medium (N-Salts samples) with cells from a single colony and then grown for 16 h at 37 °C. A 2-liter flask containing 300 ml of either LB broth (Log, Stat, and MgM samples) or N-Salts medium (N-Salts samples) was inoculated with 3 ml of the starter culture and shaken at 200 rpm at 37 °C. The Log, Stat, and N-Salts cultures were harvested at room temperature at an OD600 nm of 0.6, 2.0, and 0.2, respectively. The MgM samples were grown to stationary phase as described above, rinsed twice with MgM pH 5, and then incubated for 4 h in MgM pH 5 at 37 °C while shaking at 200 rpm. Harvested cell cultures were centrifuged at 5000 x g, cell pellets were washed twice with Dulbecco's phosphate buffered saline (Mediatech, Inc., Herndon, VA), and cell pellets with an approximate wet weight of 0.1 g/tube were stored at –80 °C.

Peptide Extraction and LC-MS/MS—
Bacterial cells were lysed using 0.1-mm zirconia/silica beads (BioSpec Products, Inc., Bartlesville, OK) and cleared by centrifugation as described in Ref. 6. The S. typhimurium strain 14028 cells were lysed in the presence of a protease inhibitor mixture formulated for use with bacterial cell extracts (Sigma). Also, half of each S. typhimurium strain 14028 cell lysate was incubated at 22 °C for 20 min to mimic poor sample handling, whereas the other half was in a cooling block at ~7 °C.

Isopropanol was then added to the cleared lysates to 1:1, 3:2, 2:1, or 5:2 (v/v) proportions (respectively) as indicated. The samples were then microcentrifuged at 16,000 x g at 4 °C for 20 min, and the resulting supernatants were concentrated in a SpeedVac (Thermo Electron Corp., Waltham, MA). Peptide concentrations were determined by BCA assays (Pierce), and SDS-PAGE analyses were performed using 10–20% Tris-Tricine Ready Gels® (Bio-Rad). Gels were fixed for 30 min in 40% methanol, 10% acetic acid and then stained for 60 min using GelCode® Blue reagent (Pierce).

Prior to mass spectrometric analysis, the concentrated isopropanol extracts were either solid phase extracted using C-18 pipette tips or were fractionated by strong cation exchange (SCX) HPLC. Each C-18 solid phase extraction was performed by applying 30 µg (peptide mass) of concentrated isopropanol extract to an OMIX® C-18 100-µl pipette tip (Varian, Palo Alto, CA) and by following the manufacturer's instructions. The C-18 SPE extracts were concentrated in a SpeedVac (Thermo Electron Corp.) and then analyzed by mass spectrometry. Each SCX fractionation was performed as described previously (6), each was performed using ~150 µg (peptide mass) of concentrated isopropanol extract, and each resulted in 25 fractions. SCX fractionation was performed on four of the concentrated isopropanol extracts (the 14028 strain Log, Stat, MgM, and N-Salts samples that had been kept in the cooling block at ~7 °C), and the resulting 100 fractions were concentrated in a SpeedVac. Note that the acidic conditions under which the C-18 SPE and SCX HPLC took place probably resulted in the precipitation and removal of some isopropanol-soluble proteins. The concentrated C-18 SPE extracts and SCX fractions were then analyzed by reversed-phase microcapillary HPLC-nanoelectrospray ionization-LTQ ion trap tandem mass spectrometry as described previously (6). Six of the samples were analyzed by LC-MS/MS in duplicate, and a description of all 126 LC-MS/MS analyses is included as Supplemental Table I. Each LC-MS/MS gradient was 100 min long, and adding to that the time that was consumed performing column regeneration, system maintenance, and LC-MS/MS analyses of quality control standards the total time that the LTQ mass spectrometer was dedicated to these samples was ~300 h.

LC-MS/MS Data Analyses—
Peptides were identified by using SEQUEST® (TurboSEQUEST® (cluster) v.27 (revision 12), Thermo Electron Corp.) to search the resulting MS/MS spectra against the annotated S. typhimurium FASTA data file of proteins translated from genetic code provided by The Institute for Genomic Research (4,550 protein sequences, September 19, 2004, Stanford University, www.tigr.org/) (15). These analyses used a standard parameter file with peptide_mass_tolerance = 3, fragment_ion_tolerance = 0, and no amino acid modifications. Also, these analyses searched for all possible peptide termini (i.e. not limited to only tryptic termini). Separate SEQUEST searches that used the above FASTA data file but with scrambled amino acid sequences were performed in parallel.

SEQUEST generally returns multiple peptide identifications for each MS/MS spectrum and for each parent ion charge state. Therefore, for each MS/MS spectrum and for each parent ion charge state, only the peptide identification with the highest XCorr value (i.e. the "top ranked hit") was retained. Note that if a MS/MS spectrum returned both normal and scrambled peptide identifications, both top ranked hits were retained.

The estimated percentage of false positive peptide identifications was defined as %FPest. = 100% x (number of scrambled peptide identifications)/(number of normal peptide identifications) (16). %FPest. was calculated for each charge state, XCorr_Cutoff value (i.e. the minimum XCorr value requirement, which ranged from 1.5 to 5 in units of 0.02), and {Delta}Cn_Cutoff value (i.e. the minimum {Delta}Cn value requirement, which ranged from 0 to 0.4 in units of 0.005). For each parent ion charge state, the optimal pair of cutoff values (XCorr_Cutoff and {Delta}Cn_Cutoff) was identified by using a two step process. First, the pair of cutoff values was required to corresponded to %FPest. ≤5%. Second, of the remaining pairs of cutoff values, the pair that was satisfied by the most normal (i.e. not scrambled) peptide identifications was identified as optimal.

The optimal XCorr_Cutoff and {Delta}Cn_Cutoff values at each parent ion charge state (+1 through +5) was determined to be 1.84 and 0.21 (+1), 2.1 and 0.21 (+2), 2.8 and 0.23 (+3), 3.56 and 0.265 (+4), and 4.16 and 0.22 (+5), respectively. Peptide identifications that failed to pass these cutoff values were discarded, as were those that corresponded to multiple S. typhimurium protein sequences. The XCorr_Factor was calculated for each peptide identification (XCorr_Factor = XCorr/XCorr_Cutoff) by using the above optimal XCorr_Cutoff values, and in the rare instance that a single MS/MS spectrum generated multiple filter-passing peptide identifications (each corresponding to a different parent ion charge state), only the peptide identification with the highest XCorr_Factor was retained. The estimated percentage of false positive protein identifications was defined as %FPrest. = 100% x (number of scrambled protein identifications)/(number of normal protein identifications). If ≥10 filter-passing peptide identifications were required for each protein identification, then %FPrest. = 1.99%. For each protein, spectrum counting (i.e. tallying of filter-passing peptide identifications) was used as a rough measure of the abundance of its fragments within the peptidome (17).

To better estimate the percentage of false positive peptide identifications, an alternative method was also used (18). All of the MS/MS spectra were reanalyzed with SEQUEST using a concatenated normal + scrambled S. typhimurium FASTA data file of proteins (i.e. the scrambled protein sequences were appended to the end of the normal FASTA data file, therefore resulting in 9,100 protein sequences total). This estimated percentage of false positive peptide identifications was defined as %FPest.2 = 100% x 2 x (number of scrambled peptide identifications)/(total number of peptide identifications). Using the above optimal XCorr_Cutoff and {Delta}Cn_Cutoff value pairs, %FPest.2 was found to be equal to 3.48%, and 88.1% of the filter-passing peptide identifications that resulted from the search of the normal FASTA file were also filter-passing identifications that resulted from the search of the concatenated FASTA file.

To test for the remote possibility of contamination of the samples, MS/MS spectra that resulted from the analyses of the eight 14028 S. typhimurium C-18 SPE samples were searched using SEQUEST against a non-redundant, multiorganism database from the Protein Information Resource (283,416 protein sequences, PIR-PSD release 78.00, Georgetown University, pir.georgetown.edu/) (19). The resulting data were filtered using the above optimal XCorr_Cutoff and {Delta}Cn_Cutoff value pairs, and although 207 new peptides were identified, they seemingly randomly corresponded to 127 apparently unrelated proteins.

LC-MS—
The eight S. typhimurium (14028 strain) concentrated C-18 SPE extracts were analyzed by reversed-phase microcapillary HPLC-nanoelectrospray ionization-FTICR-MS (11.5 tesla) (20). These analyses all used the same chromatographic column and were run in the order given in Table II. The resulting MS spectra were analyzed using the accurate mass and elution time (AMT) tag approach (5). Briefly, the calculated mass and the observed normalized elution time (NET) of each filter-passing peptide identification from the LC-MS/MS analyses were used to construct a reference database of AMT tags (note that none of the peptide identifications that had a +4 or +5 parent ion charge state were included in the AMT tag database). Features from LC-MS analyses (i.e. m/z peaks deconvolved of isotopic and charge state effects and then correlated by mass and NET) were matched to AMT tags to identify peptides in a manner that resulted in roughly 5% estimated false positive identifications. For each protein, the sum of its peptide peak areas (NET versus peak height) was used as a measure of the abundance of its fragments within the peptidome.


View this table:
[in this window]
[in a new window]

  		TABLE II LC-MS sample masses and peptide matches

 
Proteomics Analyses—
Peptidomics data were compared with proteomics data from a prior bottom-up proteomics study in our laboratory (6) that used the same preparations of Log (LT2), Stat (LT2), MgM (LT2), and N-Salts (14028) S. typhimurium cells. Data from N-Salts samples were not included in (6), but these samples were prepared and analyzed using the same protocols. Briefly, cells were lysed by boiling in RapiGestTM surfactants (Waters Corp., Milford, MA), proteins were digested with trypsin, and the resultant peptides were fractionated by SCX HPLC. A single unfractionated sample and a full set of 25 SCX fractions were then analyzed by reversed-phase LC-MS/MS. MS/MS spectra were searched using SEQUEST and filtered to reduce false positive peptide identifications. The total number of filter-passing peptide identifications from each phase was 75,201 (Log), 68,927 (Stat), 93,211 (MgM), 39,529 (N-Salts). Normalized spectrum counts (i.e. spectrum counts divided by the total number of filter-passing peptide identifications from each phase) were used as a rough measure of protein abundance within the proteome (Supplemental Table VI) (17).

Data Visualization and Cluster Analysis—
Maps of the estimated percentage of false positive peptide identifications and of the number of peptide identifications versus XCorr_Cutoff and {Delta}Cn_Cutoff were visualized using OmniViz® 3.8 (OmniViz Inc., Maynard, MA) (21) (Supplemental Fig. 1). Protein abundance data (derived by spectrum counting or by summing FTICR peak areas) were subjected to hierarchical cluster analyses using OmniViz with correlations determined by magnitude and shape (Euclidean distance).


   RESULTS
TOP
 ABSTRACT
 EXPERIMENTAL PROCEDURES
 RESULTS
DISCUSSION
 REFERENCES
 
Isolation of a Low Molecular Mass Peptide Fraction from S. typhimurium—
To simplify sample preparation and to avoid host peptides, S. typhimurium were grown in four host-free culture conditions. Although studies of free-living cultures produce only a limited model of host-pathogen pathophysiology, they are generally an important foundation to much more complex dual-organism investigations (74). The environments within logarithmic and stationary phase cultures grown in rich medium represented early stage and late-stage, host-free growth, respectively (e.g. during food contamination). The MgM and N-Salts media roughly mimicked the acidic, low magnesium, and nutrient poor environment of the phagosome (6, 12, 22–24). The N-Salts cultures produced the highest srfH (a type III (SPI-2) effector) expression compared with a number of cultures grown in a variety of phagosome-mimicking media and grown under a variety of growth conditions (data not shown). The expression of srfH was found to be regulated in a fashion consistent with the expression of Salmonella proteins during pathogenesis (data not shown) and was therefore used as an indicator of general type III secretion system expression.

Ultrafiltration of cleared cell lysates was initially used in an attempt to isolate S. typhimurium peptides. However, the filtration membranes clogged and the number of species that managed to traverse the membrane was negligible when analyzed by LC-MS/MS (data not shown). Consequently, this approach was abandoned, and the use of organic solvents to effect protein precipitation was investigated.

The addition of water-miscible organic solvents to an aqueous solution causes a reduction in its dielectric constant, an increased attraction between charged molecules, and protein aggregation (for a detailed review, see section 4.4 of Ref. 25). In general, high molecular mass proteins (~100 kDa) precipitate at low concentrations of organic solvents (~30% v/v), whereas low molecular mass proteins (~10 kDa) require higher concentrations of organic solvents (~65% v/v). Peptides are generally soluble even at high concentrations of organic solvents. Protein depletion by use of organic solvents is additionally attractive because the addition of organic solvents causes protein denaturation and therefore negligible protein-peptide co-precipitation. However, some peptide precipitation is likely at very high organic solvent concentrations. Therefore, four different organic solvents were tested, and isopropanol was found to perform the best. Afterward, to identify an ideal isopropanol concentration, the results of extractions at four different concentrations were compared. These experiments are described below.

One of four organic solvents (acetone, acetonitrile, isopropanol, or methanol) was added to cleared bacterial cell lysates to a 2:1 (v/v) proportion (respectively). In this experiment, instead of S. typhimurium cells, Shewanella oneidensis cells were used, and this experiment was performed in triplicate (i.e. there were 12 samples total). The extracts were then analyzed by LC-MS/MS, and it was found that isopropanol extraction resulted in the most filter-passing peptide identifications (data not shown). Acetonitrile performed almost as well, but acetone and methanol performed poorly. It was also observed that the protein pellets resulting from isopropanol precipitation were much more compact than the acetonitrile pellets. Consequently, isopropanol was used to extract S. typhimurium peptides from cleared cell lysates. Log, Stat, and MgM phase, LT2 strain S. typhimurium cells were lysed by bead beating and cleared by centrifugation, and four concentrations of isopropanol were added. Cleared cell lysates and isopropanol extracts were then analyzed by SDS-PAGE (Fig. 1). Although peptides <6.5 kDa were too small to be observed by SDS-PAGE, Fig. 1 shows that increasing concentrations of isopropanol progressively precipitated the S. typhimurium proteins, as expected.


Figure 1
View larger version (18K):
[in this window]
[in a new window]

  		FIG. 1. SDS-PAGE gel of cleared cell lysates and concentrated isopropanol extracts. Each lane was loaded with an amount of sample that corresponded to an equivalent amount of cleared cell lysate.

 
Identification of S. typhimurium Peptides Generated during Phagosome-like Conditions—
Concentrated isopropanol extracts were solid phase extracted using C-18 pipette tips and then analyzed by LC-MS/MS. SEQUEST was then used to search the MS/MS spectra against a dataset of S. typhimurium protein sequences, and a parallel search was performed using scrambled sequences. By calculating the ratio of scrambled to normal identifications, the estimated percentage of false positive peptide identifications (%FPest.) was determined for each parent ion charge state, XCorr_Cutoff value, and {Delta}Cn_Cutoff value (Supplemental Fig. 1). In this way, %FPest. was constrained to ≤5%.

Highly observed proteins were clustered, and a disproportionate number of protein fragments were observed in the MgM samples (Fig. 2). Because it was possible that native proteases generated these peptides after cell lysis, the experiment was repeated but this time with a bacterial protease inhibitor mixture solution added to the lysis buffer. This time the more virulent and pathophysiologically relevant 14028 strain (compared with the LT2 strain) was used. Also, half of each lysate was incubated at 22 °C for 20 min to mimic poor sample handling, whereas the other half was in a cooling block at ~7 °C. All of these samples were extracted with isopropanol (66.7% v/v, final concentration), C-18 solid phase extracted, and analyzed by LC-MS/MS.


Figure 2
View larger version (42K):
[in this window]
[in a new window]

  		FIG. 2. Cluster analysis of the LC-MS/MS data. Each column corresponds to the sample analyzed by LC-MS/MS, each row represents an individual protein, and the grayscale bar (bottom) indicates the abundance of the fragments of the protein within the peptidome (as determined by spectrum counting). Only highly observed proteins (≥10 filter-passing peptide identifications in ≥1 of the 21 samples) were included in the cluster analysis. LT2 strain: Log, Stat, and MgM cleared cell lysates were extracted with 50, 60, 66.7, or 71.4% isopropanol (indicated by the slanting, black bars), cleaned up by using a C-18 SPE pipette tip, and analyzed by LC-MS/MS (the Log 71.4% sample was analyzed twice). 14028 strain: Log, Stat, MgM, and N-Salts cleared cell lysates were divided into two equal portions, and half (L, S, M, and N columns, respectively) were kept in a cooling block (~7 °C), whereas the other half (L, S, M, and N columns, respectively) were incubated at 22 °C for 20 min to serve as a control for native proteolysis during sample handling, and subsequently all were extracted with 66.7% isopropanol, cleaned up by using a C-18 SPE pipette tip, and analyzed by LC-MS/MS.

 
Highly observed proteins from the LT2 and 14028 experiments were clustered (Fig. 2). Similar to before, a large number of protein fragments were disproportionately detected in the MgM and N-Salts samples. No significant differences between the samples incubated at 7 and 22 °C were observed. In particular, the Log and Stat samples incubated at 22 °C did not contain protein degradation fragments similar to those detected in any of the MgM or N-Salts samples. Therefore, proteolysis subsequent to cell lysis was probably insignificant. Four of the concentrated isopropanol extracts were fractionated by SCX HPLC and analyzed by LC-MS/MS (see Table I for a summary of the peptide identifications and Supplemental Table I for a description of the LC-MS/MS datasets). These LC-MS/MS analyses of offline SCX fractions were performed to improve the AMT tag database for subsequent LC-MS analyses. In addition, a cluster analysis of this data was performed as in Fig. 2, and the results were similar (Supplemental Fig. 3). Because all of these samples were not proteolytically digested in preparation for LC-MS/MS, we suspected that large (≥30 amino acids), highly charged (+4 or +5) parent ions might be identifiable. However, because of the small number of such species detected in our early experiments (Table I), the MS/MS spectra of subsequent experiments were only searched for parent ions with ≤+3 charge states. Overall, there were 45,965 filter-passing peptide identifications of 14,388 peptides from 2,003 proteins.


View this table:
[in this window]
[in a new window]

  		TABLE I Filter-passing LC-MS/MS peptide identifications

na, not analyzed by SEQUEST.

 
It was observed that many of the proteins in Fig. 2 are ribosomal proteins. One hypothesis to explain this was that the most abundant peptides simply derived from the most abundant proteins via a nonspecific degradation process. To test this hypothesis, the N-Salts peptidomics data were compared with data from our bottom-up proteomics experiments (specifically, to N-Salts RapiGest experiments performed as in Ref. 6). A correlation analysis of the proteomics and peptidomics data resulted in a very weak correlation (R2 = 0.062) (Fig. 3). Therefore, the lack of a significant correlation between the proteomic and peptidomic protein abundances and the inhomogeneous clustering of the peptides within the MgM and N-Salts samples strongly suggest that most of this protein degradation resulted from targeted proteolysis.


Figure 3
View larger version (15K):
[in this window]
[in a new window]

  		FIG. 3. Correlation of the proteomics and peptidomics protein abundance values. N-Salts phase samples were analyzed proteomically as described in Ref. 6 and peptidomically as described in the text. Spectrum counting was used to determine both the proteomic and peptidomic abundances of each protein. It total, 960 proteins had both ≥1 proteomics and ≥1 peptidomics filter-passing peptide identification.

 
Various physical properties of the identified peptides and proteins were analyzed. For example, the theoretical isoelecric point and predicted extent of intrinsic structural disorder were calculated for each protein, and the theoretical hydrophobic moment was calculated for each peptide. Although the ribosomal proteins were predicted to be alkaline, as expected, no statistically significant differences from the S. typhimurium proteome as a whole were observed (data not shown).

Of note however was the over-representation of carboxy-terminal arginine and lysine residues in the peptide identification data. Peptides with carboxy-terminal arginine and lysine residues accounted for 24.64 and 27.81% of the peptide identifications, respectively. This resulted in a 6.38- and 4.00-fold over-representation (respectively) compared with what would be expected from a random distribution of the observed amino acids (Supplemental Table XI). It is known that peptides that contain arginine and/or lysine residues are more effectively ionized during nanoelectrospray ionization (reviewed in detail in Chapter 3 of Ref. 26). In addition, peptides having carboxy-terminal arginine or lysine residues are detected more efficiently after collisionally induced dissociation (reviewed in detail in Chapter 4 of Ref. 26).

Ten proteins (STM 0359, 0462, 1121, 1244, 1438, 1728, 1810, 1924, 3362, and 3419) observed in the peptidomics data (i.e. had ≥4 unique (i.e. differing sequences), filter-passing peptide identifications) were not detected in the corresponding bottom-up proteomics data (i.e. had ≤1 filter-passing peptide identification (total)) from our previous investigation (6). To test whether isopropanol extraction enriched a set of proteins that was not detected during our bottom-up proteomics investigation (e.g. low molecular mass proteins), the 12 S. typhimurium LT2 concentrated isopropanol extracts were digested with trypsin, analyzed by LC-MS/MS, and searched using SEQUEST. Two additional proteins (STM 3363 and 3450) were found (i.e. had ≥4 unique, filter-passing peptide identifications and were not in the corresponding bottom-up proteomics data). These 12 proteins do not seem to be related to each other in any way, and it was concluded that isopropanol extraction did not enrich a substantial set of proteins that had been missed by our traditional proteomics investigation.

To identify non-annotated S. typhimurium proteins, MS/MS spectra were searched using SEQUEST against a protein FASTA file constructed from translations of S. typhimurium LT2 DNA sequences flanked by stop codons (i.e. stop-to-stop translations). These searches included MS/MS spectra from both the peptidomics experiments described above and the bottom-up proteomics experiments performed previously in our laboratory (6). Care was taken to discard any non-annotated peptide identification that had resulted from the analysis of a MS/MS spectrum that had already been associated with an annotated peptide identification from the prior SEQUEST searches. In addition, care was taken to discard any identification of a non-annotated peptide which could have derived from a "S. enterica" or "S. typhimurium" protein annotated in the UniProt database (7,714 protein sequences, August 9, 2006, us.expasy.org/) (27). Ultimately, nine non-annotated proteins were identified (requiring ≥3 unique (i.e. differing sequences) and ≥15 total filter-passing peptide identifications). However, the percentage of these protein identifications that are false positives has yet to be estimated, and none of these identifications have yet been validated using orthogonal experimentation (e.g. Western blotting). Therefore, these results are preliminary. It was noted that seven of the nine non-annotated proteins had a high degree of homology with proteins identified from Basic Local Alignment Search Tool (BLAST) searches against the Entrez protein sequence database provided by the National Center for Biotechnology Information (www.ncbi.nlm.nih.gov/). In addition, seven of the nine non-annotated genes partially overlapped other, annotated genes, and it is possible that these non-annotated genes had been identified as pseudogenes for this reason. Studies of non- and misannotated S. typhimurium proteins are ongoing in our laboratory.

Identification and Quantification of Peptides Using LC-MS—
The eight S. typhimurium (14028 strain) concentrated C-18 extracts were analyzed by LC-FTICR-MS. Equalization of the sample masses took place following isopropanol extraction and prior to C-18 SPE (Table II). Equalization was designed to take into account any sample preparation biases, and no subsequent sample or data normalization was performed. Note that the peptidome masses recovered after the C-18 cleanups were similar.

LC-MS features (i.e. m/z peaks deconvolved of isotopic and charge state effects and then correlated by mass and NET) were matched to AMT tags (i.e. filter-passing LC-MS/MS peptide identifications) in a process that resulted in an average mass error of 1.09 ppm and an average NET error of 0.74%. Using the AMT tag approach, the eight LC-MS analyses resulted in 15,025 peptide identifications of 5,163 peptides from 682 proteins. As described above, 14,388 peptides from 2,003 proteins were identified by LC-MS/MS. Most likely, compared with the high resolution LC-MS analyses of the unfractionated samples, a greater depth of coverage resulted from the LC-MS/MS analyses of the SCX fractions. The total number of LC-MS features and the percentage of matching features increased in rough proportion to the stress that the cells underwent (Table II), which correlated well with the spectrum counting results (Fig. 2).

Highly abundant proteins were clustered (Fig. 4, left), and the results were very similar to the spectrum counting results (Fig. 2). The peptidomics data were compared with proteomics data (Fig. 4, right) from a prior study performed in our laboratory (6). Although individual Z-scores of ~1.5 are generally not statistically significant by themselves, the overall trend of the data suggests that ribosomal proteins were most abundant during logarithmic phase growth in rich medium, as expected (28). In contrast, a number of known stress response factors were found to be most abundant during suboptimal culture conditions. The two-component regulator phoP (29, 30), the histone-like protein hupB (31), and the universal stress protein ybdQ (32) were most abundant in the Stat samples, and the chaperone dnaK (30, 33, 34) and the histone-like protein hns (31) were most abundant in the N-Salts samples.


Figure 4
View larger version (64K):
[in this window]
[in a new window]

  		FIG. 4. Cluster analysis of the LC-MS data. Left, each column corresponds to the 14028 strain sample analyzed by LC-MS (see Fig. 2 legend for sample descriptions), each row represents an individual protein, and the color bar (bottom) indicates the abundance values (calculated from the sum of the peptide peak areas). Only the highly abundant proteins (≥3 in ≥1 of the 8 samples) were included in the cluster analysis. Right, for the corresponding proteins and cell cultures, proteomics data from a prior study (6) are displayed. Cells were lysed by boiling in RapiGest surfactant, proteins were digested with trypsin, and the resultant peptides were fractionated by SCX HPLC. Unfractionated and SCX fractionated samples were then analyzed by LC-MS/MS, and spectrum counting was used as a rough measure of protein abundance. Z-scores of the protein abundance values were calculated across the 4 samples and are indicated by the color bar (bottom).

 
To investigate deeper into the S. typhimurium peptidome and proteome, data from this work and from our prior study (6) were co-clustered (Fig. 5), and two prominent clusters were identified. The first, designated putative growth proteins (Gr.P.), consisted of 48 proteins highly abundant in the Log samples and highly fragmented in the N-Salts samples. Almost all of these (37 of 48) were ribosomal proteins. The second, designated putative stress response factors (S.R.F.), consisted of 71 proteins both highly abundant and highly fragmented in the N-Salts samples. Many of these proteins are known stress response factors. In addition to the proteins listed in Fig. 4, these proteins included six involved in copper (copA), iron (feoA, feoB, sitB) and magnesium (mgtA, mgtB) transport, six involved in amino acid synthesis (ilvH, serA, thrC, usg, yifE, ynhG), the response regulators phoB, phoN, and phoR, and the rRNA RNase (rna (35, 36)) (note, underlined genes denote putative functions). Possible physiological roles of these two protein clusters are discussed below.


Figure 5
View larger version (82K):
[in this window]
[in a new window]

  		FIG. 5. Peptidomics and proteomics co-cluster analysis. Each column corresponds to the sample analyzed by LC-MS (see Fig. 4 legend for sample descriptions), and each row represents an individual protein (278 total). Peptidome columns, Z-scores of the protein fragment abundance values were calculated across the 8 samples. Only abundant proteins (≥0.1 in ≥1 of the 8 samples) were included. Proteome columns, for the corresponding proteins and cell cultures, proteomics data from a prior study (6) are displayed. Z-scores of the protein abundance values were calculated across the 4 samples as in Fig. 4. All columns, all of the data were simultaneously clustered, and the Z-score values are indicated by the color bar (bottom). Two protein clusters, putative growth proteins (Gr.P.) and putative stress response factors (S.R.F.), are discussed in the text.

 

   DISCUSSION
TOP
 ABSTRACT
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
REFERENCES
 
In this investigation, high throughput LC-MS and LC-MS/MS were used to identify and quantify thousands of native S. typhimurium peptides generated during conditions designed to mimic the environment of the phagosome. The first challenge was to remove as much of the protein from the samples as possible to prevent it from overwhelming the SCX and reversed-phase media that were critical to the downstream analyses. Fortunately, a number of descriptions of peptide enrichment strategies, mostly applied to serum and other biofluids, have recently been published. Because it is probable that extraction with isopropanol results in the loss of some peptides, peptidomics coverage could be expanded by using orthogonal enrichment protocols such as acid precipitation coupled to reversed-phase or SCX SPE, restricted access media chromatography, and preparative electrophoresis. Ultrafiltration was unsuccessful probably because under native conditions some level of protein precipitation is unavoidable, and under denaturing conditions filtration efficiency is low because the ultrafiltration of linearized polypeptides performs poorly in general.

A number of S. typhimurium proteins were found to be degraded upon exposure of cells to phagosome-like culture conditions. Because the MgM and N-Salts media contain a low concentration of nutrients (including 10% of the concentration of amino acids of LB broth), it is possible that the cells were primarily responding to starvation. During the "stringent response," bacteria down-regulate rRNA and ribosomal protein expression (28), dimerize and inactivate ribosomes (37), and degrade rRNA (35, 36) (which liberates complexed ribosomal proteins (38)). In addition, bacteria produce polyphosphate (PolyP), which results in the degradation of uncomplexed ribosomal proteins by the PolyP-activated Lon protease (39–41). In this study, the stationary phase samples had relatively little ribosomal protein fragmentation even though these cells were likely experiencing significant carbon starvation. This suggests that low magnesium and/or low pH are also required for ribosomal protein proteolysis to be up-regulated. Magnesium is known to be required for ribosomal stability (42), and it is also a known inhibitor of the rRNA RNase rna (36, 38).

Most ribosomal proteins are small, alkaline, and function solely by stabilizing rRNA folding by binding to it (and likely to PolyP) with non-globular, positively charged extensions (43, 44). These extensions are probably structurally disordered when unbound to polynucleotides because they are disproportionately detected by proteome-wide predictions of regions of intrinsic structural disorder (45, 46). Lon is known to target defective proteins (47–51), probably because its unfoldase can process proteins that contain intrinsically disordered regions faster than proteins without such regions. This suggests a mechanism by which Lon targets ribosomal proteins. First, during starvation Salmonella rRNA is quickly degraded (52), releasing ribosomal proteins. Simultaneously, polyphosphatase is down-regulated, and the non-globular extensions of uncomplexed ribosomal proteins bind to PolyP (note that only uncomplexed ribosomal proteins are vulnerable to Lon proteolysis (refer to Supplemental Fig. 2, Ref. 53). Next, Lon binds to PolyP and possibly traverses up and down the phosphate polymer. Lastly, PolyP-bound, non-globular extensions of ribosomal proteins are rapidly processed by the Lon unfoldase. Bacterial histone-like proteins have similar DNA-binding extensions (54, 55), and three (hupA (56), stpA (57), and ymoA (58)) are known Lon substrates. Paradoxically, although Lon down-regulates the expression of invasion genes and the efficiency of host-cell invasion (59, 60), the study of Lon-null mutants has demonstrated that Lon is essential for systemic infection (61). Because the availability of nutrients within the phagosome is limited, ribosomal protein degradation may foster pathogenesis by generating much needed amino acids, thereby increasing the survivability of phagocytosed Salmonella. In addition, Lon degradation of ribosomal proteins might produce bioactive peptides such as host-cell toxins.

Although many of the detected peptides were derived from ribosomal proteins, a number of other proteins were also observed. Two of these proteins, trigger factor (tig) (62) and protein chain initiation factor I (infA) (63), are known to bind to ribosomes, and it is possible that the uncomplexed forms of these proteins are somehow more vulnerable to degradation. Many of the remaining proteins were found to be most abundant in stressed Salmonella (i.e. in the Stat, MgM, or N-Salts samples) and thus are likely to have stress response functions. Fragments of many of these were most abundant in the N-Salts samples. That Salmonella would up-regulate the expression of stress response factors only to target them for degradation is seemingly paradoxical. However, translational fidelity is known to dramatically decrease during starvation, which results in mistranslated proteins that are significantly less likely to properly fold, and thus their exposure to proteolysis is increased (64). In addition, misfolded proteins are more vulnerable to modification by reactive chemicals (e.g. carbonylation by reactive oxygen species), and these modifications further undermine proper folding, which makes these proteins even more susceptible to degradation (64). Assuming that ribosomes are indeed being catabolized, then this introduces a novel explanation of why starvation results in the production of defective proteins as partially digested ribosomes almost certainly have reduced translational fidelity. Two stress response factors depicted in Fig. 4, dnaK (33, 50, 65) and hns (33, 66), are known to be carbonylated in response to starvation, and it is possible that they are subsequently targeted for proteolysis. Although Lon is the major protease of aberrant proteins in bacteria (47–51), ClpABPX (49–51, 67), HslUV (50), and HtrA (49, 68) also have important roles. Therefore, two modes of protein degradation are likely up-regulated by phagocytosed Salmonella, PolyP/Lon-dependent directed proteolysis of ribosomal proteins and ClpABPX/HslUV/HtrA/Lon proteolysis of defective stress response factors.

It is likely that some of the identified peptides are bioactive. For example, Salmonella peptides are believed to interfere with host-cell MHC-I peptide presentation via an unknown PhoPQ- and YejABEF-dependent mechanism (69). Some of the amphipathic peptides may act by lysing the host-cell phagosomal and/or plasma membranes or may act as biofilm surfactants (70). Some may be bacteriocins (e.g. the Helicobacter pylori ribosomal L1 derived bacteriocin (71)), quorum sensors (e.g. Salmonella acetyltransferase repressing factor (72)) (perhaps via PhoPQ (29)), quorum sensor-mimics (e.g. against Gram-positive bacteria in the gastrointestinal tract), host hormone-mimics, host-cell toxins, pro-inflammatory peptides, or the cause of autoimmune disorders (i.e. via molecular mimicry on major histocompatibility complex molecules). The extent of peptidomic host-Salmonella interactions is largely unknown because only recently have researchers had the ultrasensitive, high throughput tools necessary to conduct peptidomics investigations. In the future, linguistic models of peptide sequences may be capable of distinguishing bioactive peptides from bioinactive ones (73). For now, however, although comparative peptidomics alone was insufficient to tentatively identify bioactive peptides, it did prove to be a powerful tool for identifying proteins targeted for proteolysis.


   ACKNOWLEDGMENTS
 
We gratefully acknowledge the contributions of Marina Gritsenko, Ronald Moore, Karin Rodland, Dwayne Elias, Liang Shi, Joshua Turse, Bhairavi Shanghavi, Penny Colton, and Harold Udseth.


  FOOTNOTES
 
Received, August 1, 2006, and in revised form, January 8, 2007.

Published, MCP Papers in Press, January 16, 2007, DOI 10.1074/mcp.M600282-MCP200

1 The abbreviations used are: Log, logarithmic; Stat, stationary; MgM, acidic, magnesium-depleted minimal medium; LB, Luria-Bertani; SCX, strong cation exchange; SPE, solid phase extraction; %FPest., estimated percentage of false positive peptide identifications; AMT, accurate mass and elution time; NET, normalized elution time; PolyP, polyphosphate. Back

* Portions of this work were supported by the National Institute of Allergy and Infectious Diseases, (National Institutes of Health (NIH)/Department of Health and Human Services through interagency agreement Y1-AI-4894-01) and the NIH National Center for Research Resources (RR18522). Additional supporting results and protocols are available at the National Institute of Allergy and Infectious Diseases funded Administrative Resource for Biodefense Proteomics Research Centers under the PNNL data section (www.proteomicsresearch.org). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Back

S The on-line version of this article (available at http://www.mcponline.org) contains supplemental material. Back

|| To whom correspondence and reprint requests should be addressed: Dept. of Molecular Microbiology and Immunology, OHSU Mail Drop L220, 3181 S. W. Sam Jackson Park Rd., Portland, OR 97239. E-mail: heffronf{at}ohsu.edu


   REFERENCES
TOP
 ABSTRACT
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Adermann, K., John, H., Standker, L., and Forssmann, W. G. (2004) Exploiting natural peptide diversity: novel research tools and drug leads. Curr. Opin. Biotechnol. 15, 599 –606[CrossRef][Medline]

  2. Ivanov, V. T., and Yatskin, O. N. (2005) Peptidomics: a logical sequel to proteomics. Expert Rev. Proteomics 2, 463 –473[CrossRef][Medline]

  3. Schulz-Knappe, P., Schrader, M., and Zucht, H. D. (2005) The peptidomics concept. Comb. Chem. High Throughput Screen. 8, 697 –704[CrossRef][Medline]

  4. Bogdanov, B., and Smith, R. D. (2005) Proteomics by FTICR mass spectrometry: top down and bottom up. Mass Spectrom. Rev. 24, 168 –200[CrossRef][Medline]

  5. Zimmer, J. S., Monroe, M. E., Qian, W. J., and Smith, R. D. (2006) Advances in proteomics data analysis and display using an accurate mass and time tag approach. Mass Spectrom. Rev. 25, 450 –482[CrossRef][Medline]

  6. Adkins, J. N., Mottaz, H. M., Norbeck, A. D., Gustin, J. K., Rue, J., Clauss, T. R., Purvine, S. O., Rodland, K., Heffron, F., and Smith, R. D. (2006) Analysis of the Salmonella typhimurium proteome through environmental response toward infectious conditions. Mol. Cell. Proteomics 5, 1450 –1461[Abstract/Free Full Text]

  7. Chalker, R. B., and Blaser, M. J. (1988) A review of human salmonellosis: III. Magnitude of Salmonella infection in the United States. Rev. Infect Dis. 10, 111 –124[Medline]

  8. Pang, T., Bhutta, Z. A., Finlay, B. B., and Altwegg, M. (1995) Typhoid fever and other salmonellosis: a continuing challenge. Trends Microbiol. 3, 253 –255[CrossRef][Medline]

  9. Furuya, E. Y., and Lowy, F. D. (2006) Antimicrobial-resistant bacteria in the community setting. Nat. Rev. Microbiol. 4, 36 –45[CrossRef][Medline]

  10. Tucker, J. B. (1999) Historical trends related to bioterrorism: An empirical analysis. Emerg. Infect. Dis. 5, 498 –504[Medline]

  11. Beuzon, C. R., Meresse, S., Unsworth, K. E., Ruiz-Albert, J., Garvis, S., Waterman, S. R., Ryder, T. A., Boucrot, E., and Holden, D. W. (2000) Salmonella maintains the integrity of its intracellular vacuole through the action of SifA. EMBO J. 19, 3235 –3249[CrossRef][Medline]

  12. Beuzon, C. R., Banks, G., Deiwick, J., Hensel, M., and Holden, D. W. (1999) pH-dependent secretion of SseB, a product of the SPI-2 type III secretion system of Salmonella typhimurium. Mol. Microbiol. 33, 806 –816[CrossRef][Medline]

  13. Reeves, E. P., Lu, H., Jacobs, H. L., Messina, C. G., Bolsover, S., Gabella, G., Potma, E. O., Warley, A., Roes, J., and Segal, A. W. (2002) Killing activity of neutrophils is mediated through activation of proteases by K+ flux. Nature 416, 291 –297[CrossRef][Medline]

  14. Wagner, D., Maser, J., Lai, B., Cai, Z., Barry, C. E., III, Honer Zu Bentrup, K., Russell, D. G., and Bermudez, L. E. (2005) Elemental analysis of Mycobacterium avium-, Mycobacterium tuberculosis-, and Mycobacterium smegmatis-containing phagosomes indicates pathogen-induced microenvironments within the host cell's endosomal system. J. Immunol. 174, 1491 –1500[Abstract/Free Full Text]

  15. Peterson, J. D., Umayam, L. A., Dickinson, T., Hickey, E. K., and White, O. (2001) The comprehensive microbial resource. Nucleic Acids Res. 29, 123 –125[Abstract/Free Full Text]

  16. Qian, W. J., Liu, T., Monroe, M. E., Strittmatter, E. F., Jacobs, J. M., Kangas, L. J., Petritis, K., Camp, D. G., II, and Smith, R. D. (2005) Probability-based evaluation of peptide and protein identifications from tandem mass spectrometry and SEQUEST analysis: the human proteome. J. Proteome Res. 4, 53 –62[CrossRef][Medline]

  17. Qian, W. J., Jacobs, J. M., Camp, D. G., II, Monroe, M. E., Moore, R. J., Gritsenko, M. A., Calvano, S. E., Lowry, S. F., Xiao, W., Moldawer, L. L., Davis, R. W., Tompkins, R. G., and Smith, R. D. (2005) Comparative proteome analyses of human plasma following in vivo lipopolysaccharide administration using multidimensional separations coupled with tandem mass spectrometry. Proteomics 5, 572 –584[CrossRef][Medline]

  18. Peng, J., Elias, J. E., Thoreen, C. C., Licklider, L. J., and Gygi, S. P. (2003) Evaluation of multidimensional chromatography coupled with tandem mass spectrometry (LC/LC-MS/MS) for large-scale protein analysis: the yeast proteome. J. Proteome Res. 2, 43 –50[CrossRef][Medline]

  19. Barker, W. C., Garavelli, J. S., McGarvey, P. B., Marzec, C. R., Orcutt, B. C., Srinivasarao, G. Y., Yeh, L. S., Ledley, R. S., Mewes, H. W., Pfeiffer, F., Tsugita, A., and Wu, C. (1999) The PIR-International Protein Sequence Database. Nucleic Acids Res. 27, 39 –43[Abstract/Free Full Text]

  20. Gorshkov, M. V., Pasa Tolic, L., Udseth, H. R., Anderson, G. A., Huang, B. M., Bruce, J. E., Prior, D. C., Hofstadler, S. A., Tang, L., Chen, L. Z., Willett, J. A., Rockwood, A. L., Sherman, M. S., and Smith, R. D. (1998) Electrospray ionization-Fourier transform ion cyclotron resonance mass spectrometry at 11.5 tesla: instrument design and initial results. J. Am. Soc. Mass Spectrom. 9, 692 –700[CrossRef][Medline]

  21. Saffer, J. D., Burnett, V. L., Chen, G., and van der Spek, P. (2004) Visual analytics in the pharmaceutical industry. IEEE Comput. Graph Appl. 24, 10 –15[CrossRef][Medline]

  22. Garcia Vescovi, E., Soncini, F. C., and Groisman, E. A. (1996) Mg2+ as an extracellular signal: environmental regulation of Salmonella virulence. Cell 84, 165 –174[CrossRef][Medline]

  23. Heithoff, D. M., Conner, C. P., Hentschel, U., Govantes, F., Hanna, P. C., and Mahan, M. J. (1999) Coordinate intracellular expression of Salmonella genes induced during infection. J. Bacteriol. 181, 799 –807[Abstract/Free Full Text]

  24. Bringer, M. A., Barnich, N., Glasser, A. L., Bardot, O., and Darfeuille-Michaud, A. (2005) HtrA stress protein is involved in intramacrophagic replication of adherent and invasive Escherichia coli strain LF82 isolated from a patient with Crohn's disease. Infect. Immun. 73, 712 –721[Abstract/Free Full Text]

  25. Scopes, R. K. (1994) Protein Purification: Principles and Practice, 3rd Ed., pp. 85–92, Springer-Verlag, New York

  26. Kinter, M., and Sherman, N. E. (2000) Protein Sequencing and Identification Using Tandem Mass Spectrometry, pp. 29–116, John Wiley and Sons, Inc., New York

  27. Bairoch, A., Apweiler, R., Wu, C. H., Barker, W. C., Boeckmann, B., Ferro, S., Gasteiger, E., Huang, H., Lopez, R., Magrane, M., Martin, M. J., Natale, D. A., O'Donovan, C., Redaschi, N., and Yeh, L. S. (2005) The Universal Protein Resource (UniProt). Nucleic Acids Res. 33, D154 –D159[Abstract/Free Full Text]

  28. Paul, B. J., Ross, W., Gaal, T., and Gourse, R. L. (2004) rRNA transcription in Escherichia coli. Annu. Rev. Genet. 38, 749 –770[CrossRef][Medline]

  29. Bader, M. W., Sanowar, S., Daley, M. E., Schneider, A. R., Cho, U., Xu, W., Klevit, R. E., Le Moual, H., and Miller, S. I. (2005) Recognition of antimicrobial peptides by a bacterial sensor kinase. Cell 122, 461 –472[CrossRef][Medline]

  30. Rychlik, I., and Barrow, P. A. (2005) Salmonella stress management and its relevance to behaviour during intestinal colonisation and infection. FEMS Microbiol. Rev. 29, 1021 –1040[CrossRef][Medline]

  31. Travers, A., and Muskhelishvili, G. (2005) DNA supercoiling–a global transcriptional regulator for enterobacterial growth? Nat. Rev. Microbiol. 3, 157 –169[CrossRef][Medline]

  32. Siegele, D. A. (2005) Universal stress proteins in Escherichia coli. J. Bacteriol. 187, 6253 –6254[Free Full Text]

  33. Dukan, S., and Nystrom, T. (1998) Bacterial senescence: stasis results in increased and differential oxidation of cytoplasmic proteins leading to developmental induction of the heat shock regulon. Genes Dev. 12, 3431 –3441[Abstract/Free Full Text]

  34. Buchmeier, N. A., and Heffron, F. (1990) Induction of Salmonella stress proteins upon infection of macrophages. Science 248, 730 –732[Abstract/Free Full Text]

  35. Deutscher, M. P. (2003) Degradation of stable RNA in bacteria. J. Biol. Chem. 278, 45041 –45044[Free Full Text]

  36. Deutscher, M. P. (2006) Degradation of RNA in bacteria: comparison of mRNA and stable RNA. Nucleic Acids Res. 34, 659 –666[Abstract/Free Full Text]

  37. Wada, A. (1998) Growth phase coupled modulation of Escherichia coli ribosomes. Genes Cells 3, 203 –208[Abstract]

  38. Raziuddin, Chatterji, D., Ghosh, S., and Burma, D. P. (1979) Site of action of RNase I on the 50 S ribosome of Escherichia coli and the association of the enzyme with the partially degraded subunit. J. Biol. Chem. 254, 10575 –10578[Abstract/Free Full Text]

  39. Kuroda, A. (2006) A polyphosphate-lon protease complex in the adaptation of Escherichia coli to amino acid starvation. Biosci. Biotechnol. Biochem. 70, 325 –331[CrossRef][Medline]

  40. Nishi, K., and Schnier, J. (1988) The phenotypic suppression of a mutation in the gene rplX for ribosomal protein L24 by mutations affecting the lon gene product for protease LA in Escherichia coli K12. Mol. Gen. Genet. 212, 177 –181[CrossRef][Medline]

  41. Nishii, W., Suzuki, T., Nakada, M., Kim, Y. T., Muramatsu, T., and Takahashi, K. (2005) Cleavage mechanism of ATP-dependent Lon protease toward ribosomal S2 protein. FEBS Lett. 579, 6846 –6850[CrossRef][Medline]

  42. Hoa, G. H., Begard, E., Beaudry, P., Maurel, P., Grunberg-Manago, M., and Douzou, P. (1980) Analysis of cosolvent and divalent cation effects on association equilibrium and activity of ribosomes. Biochemistry 19, 3080 –3087[CrossRef][Medline]

  43. Ban, N., Nissen, P., Hansen, J., Moore, P. B., and Steitz, T. A. (2000) The complete atomic structure of the large ribosomal subunit at 2.4 A resolution. Science 289, 905 –920[Abstract/Free Full Text]

  44. Schuwirth, B. S., Borovinskaya, M. A., Hau, C. W., Zhang, W., Vila-Sanjurjo, A., Holton, J. M., and Cate, J. H. (2005) Structures of the bacterial ribosome at 3.5 A resolution. Science 310, 827 –834[Abstract/Free Full Text]

  45. Dunker, A. K., Obradovic, Z., Romero, P., Garner, E. C., and Brown, C. J. (2000) Intrinsic protein disorder in complete genomes. Genome Inform. Ser. Workshop Genome Inform. 11, 161 –171[Medline]

  46. Uversky, V. N., Oldfield, C. J., and Dunker, A. K. (2005) Showing your ID: intrinsic disorder as an ID for recognition, regulation and cell signaling. J. Mol. Recognit. 18, 343 –384[CrossRef][Medline]

  47. Downs, D., Waxman, L., Goldberg, A. L., and Roth, J. (1986) Isolation and characterization of lon mutants in Salmonella typhimurium. J. Bacteriol. 165, 193 –197[Abstract/Free Full Text]

  48. Gottesman, S. (1996) Proteases and their targets in Escherichia coli. Annu. Rev. Genet. 30, 465 –506[CrossRef][Medline]

  49. Laskowska, E., Kuczynska-Wisnik, D., Skorko-Glonek, J., and Taylor, A. (1996) Degradation by proteases Lon, Clp and HtrA, of Escherichia coli proteins aggregated in vivo by heat shock; HtrA protease action in vivo and in vitro. Mol. Microbiol. 22, 555 –571[CrossRef][Medline]

  50. Fredriksson, A., Ballesteros, M., Dukan, S., and Nystrom, T. (2005) Defense against protein carbonylation by DnaK/DnaJ and proteases of the heat shock regulon. J. Bacteriol. 187, 4207 –4213[Abstract/Free Full Text]

  51. Vera, A., Aris, A., Carrio, M., Gonzalez-Montalban, N., and Villaverde, A. (2005) Lon and ClpP proteases participate in the physiological disintegration of bacterial inclusion bodies. J. Biotechnol. 119, 163 –171[CrossRef][Medline]

  52. Hsu, D., Shih, L. M., and Zee, Y. C. (1994) Degradation of rRNA in Salmonella strains: a novel mechanism to regulate the concentrations of rRNA and ribosomes. J. Bacteriol. 176, 4761 –4765[Abstract/Free Full Text]

  53. Kuroda, A., Nomura, K., Ohtomo, R., Kato, J., Ikeda, T., Takiguchi, N., Ohtake, H., and Kornberg, A. (2001) Role of inorganic polyphosphate in promoting ribosomal protein degradation by the Lon protease in E. coli. Science 293, 705 –708[Abstract/Free Full Text]

  54. Ramstein, J., Hervouet, N., Coste, F., Zelwer, C., Oberto, J., and Castaing, B. (2003) Evidence of a thermal unfolding dimeric intermediate for the Escherichia coli histone-like HU proteins: thermodynamics and structure. J. Mol. Biol. 331, 101 –121[CrossRef][Medline]

  55. Ono, S., Goldberg, M. D., Olsson, T., Esposito, D., Hinton, J. C., and Ladbury, J. E. (2005) H-NS is a part of a thermally controlled mechanism for bacterial gene regulation. Biochem. J. 391, 203 –213[CrossRef][Medline]

  56. Bonnefoy, E., Almeida, A., and Rouviere-Yaniv, J. (1989) Lon-dependent regulation of the DNA binding protein HU in Escherichia coli. Proc. Natl. Acad. Sci. U. S. A. 86, 7691 –7695[Abstract/Free Full Text]

  57. Johansson, J., and Uhlin, B. E. (1999) Differential protease-mediated turnover of H-NS and StpA revealed by a mutation altering protein stability and stationary-phase survival of Escherichia coli. Proc. Natl. Acad. Sci. U. S. A. 96, 10776 –10781