Francisella tularensis Proteome: Low Levels of ASB-14 Facilitate the Visualization of Membrane Proteins in Total Protein Extracts Susan M. Twine,* Nadia C. S. Mykytczuk, Mireille Petit, Tammy-Lynn Tremblay, J. Wayne Conlan, and John F. Kelly* Institute for Biological Sciences, National Research Council of Canada, 100 Sussex Drive, Ottawa, Ontario, Canada Received April 13, 2005
Abstract: Proteomic analysis of bacterial pathogens isolated from in vivo sources, such as infected tissues, provides many challenges not the least of which is the limited quantity of sample available for analysis. It is, therefore, highly desirable to develop a one-step cellular lysis and protein solubilization method that minimizes protein losses and allows the maximum possible coverage of the proteome. Here, we have used standard sample buffer constituents including urea, thiourea and DTT, but varied the detergent composition of the buffers in order to achieve the best quality of gels and the greatest spot resolution. We found that the most efficient solubilizing solution in this case consisted of 7 M urea, 2 M thiourea, 1% DTT, 0.5% amidosulfobetaine-14 (ASB-14) and 4% 3-[(3-Cholamidopropyl)dimethylammonio]-1-propanesulfonate (CHAPS). Inclusion of low levels of ASB-14 in solutions allowed visualization of a subset of 24 new protein spots in the Live Vaccine Strain (LVS) of Francisella tularensis and 21 spots in a virulent A-strain of the pathogen. Further investigation showed that 15 of the 24 enriched LVS spots were membrane or membraneassociated proteins suggesting that the optimized lysis and solubilization solution aids in the detection of more hydrophobic proteins. This methodology is now being applied to the analysis of Francisella obtained from in vivo sources. Keywords: 2DE • detergents • bacteria • proteomics
Introduction Francisella tularensis (F. tularensis) is a facultative intracellular pathogen and is the causative agent of tularemia in small rodents and humans. Two of the four Francisella subspecies routinely cause human infectionssType A (subspecies tularensis) and Type B (subspecies holarctica). In humans, tularemia caused by the virulent type A subspecies of the pathogen has a high mortality rate if left untreated.1-5 A combination of the ease of dissemination of the pathogen, high infectivity, and associated mortality has resulted in F. tularensis being catego* To whom correspondence should be addressed. Tel: (613) 998-5263. E-mail:
[email protected],
[email protected].
1848
Journal of Proteome Research 2005, 4, 1848-1854
Published on Web 08/10/2005
rized as a Type A biowarfare agent.4,6,7 Since F. tularensis has not proved readily amenable to traditional genetic manipulations and generation of gene knock-outs, a detailed study of the proteome of this pathogen is likely the best method for revealing virulence factors and key protective antigens. The current lack of knowledge regarding the latter is inhibiting the development of a defined tularemia vaccine.8 A number of studies have utilized in vitro grown strains F. tularensis in proteomics studies. One group is working toward preparing a proteome database of F. tularensis and has conducted a comparative analysis of the proteomes of strains of the pathogen.9-11 Other studies have used 2DE to separate radiolabeled F. tularensis proteins, after exposure of cell to stress conditions in vitro, such as hydrogen peroxide.11-15 Such in vitro studies of bacterial pathogens subjected to defined stress conditions, designed to simulate the conditions seen by the pathogen in vivo are widely used and can provide valuable insight into the physiological response of pathogens to specific environmental stimuli.16-19 For example, a proteomics study of hydrogen peroxide exposed F. tularensis Live Vaccine showed the increased synthesis of heat shock proteins, such as DnaK and GroEL, and a 23 kDa protein.12 However, while in vitro experiments can provide useful information, it is more likely that proteomic analysis of bacteria isolated from their host environment (i.e., the tissues of infected animals) will reveal the proteins required for in-vivo survival of the pathogen. These experiments will never yield large quantities of bacteria for analysis. Therefore, it is important to develop proteome analysis techniques that can maximize the information acquired on these small samples. Advances in various mass spectrometry based techniques, such as isotopic labeling of cysteine residues, are becoming increasingly popular for differential proteomic analyses, but two-dimensional gel electrophoresis (2DE) is still the most widely used method for protein separation and analysis. Although the technique is known to be biased toward proteins that are in higher abundance or that are hydrophilic,20 it provides a highly visual representation of the proteome and allows the detection of protein isoforms and post-translational modifications.21 Previously reported proteomics studies of F. tularensis have used a protocol which involves cellular lysis by cycles of freezethawing into a high salt buffer, followed by protein precipitation and resuspension into a urea/3-[(3-Cholamidopropyl)dimethylammonio]-1-propanesulfonate (CHAPS)/DTT containing buffer for isoelectric focusing.9-11 When working with samples where 10.1021/pr050102u CCC: $30.25
2005 American Chemical Society
technical notes
Twine et al.
Table 1. Protein Solubilization Buffer Detergent Components and Average Numbers of Spots Observed on 2DEa solution
1 2 3 4 5 6 7
CHAPS (%)
SB3-10 (%)
4 4 4 4
2
ASB-14 (%)
2 0.5 1.0 0.5 1.0
LVS no. spots
FSC033 no. spots
569 ( 32 401 ( 18 395 ( 20 480 ( 22 439 ( 16 434 ( 29 410 ( 13
n/a n/a 400 ( 10 425 ( 19 428 ( 13 420 ( 12 419 ( 22
a Solutions 1-7 were used for cellular lysis and total protein extraction. Solutions 1 and 3-7, in addition to the detergents listed, contained 7 M urea, 5 M thiourea, 1% DTT. Solution 2 contained 5 M urea, 2 M thiourea, 1% DTT in addition to detergents. Mean numbers of spots were determined from the numbers of spots detected using PDQuest software (Biorad) where n ) 4.
the initial concentration of bacteria is not limiting, this protocol clearly works well. However, to achieve our aims, we favored the use of a single-step cell lysis and solubilization protocol that would minimize protein losses, such as those frequently associated with protein precipitation techniques. The data we present here began as an effort to optimize the cell lysis and solubilization solution in order to achieve the maximum coverage of F. tularensis proteins by 2DE. Two strains of in vitro grown F. tularensis, the Live Vaccine Strain (LVS, an empirically attenuated Type B strain) and FSC033 (a virulent Type A strain), were used for this study, since they are both relevant to our research. To date, only one other study has presented data regarding the proteome of virulent A strain bacteria.10 During the course of this work, we determined that is possible to obtain excellent quality 2DE separations using the lysis and solubilization solution for IEF and that the addition of 0.5% (w/v) ASB-14 to this solution improved the number of hydrophobic proteins observed in the 2DE gels of the total proteome extracts of both LVS and FSC033.
Experimental Procedures Sample Preparation for 2D Gel Electrophoresis. Francisella tularensis LVS (ATCC29684) and type A strain, FSC033 (CDC standard) were grown in modified Mueller-Hinton broth for 24-36 h at 37 °C with shaking until bacterial density reached 108 -1010 CFU/mL. Bacterial concentrations were determined by plating 10-fold serial dilutions on cysteine heart agar supplemented with 1% (w/v) hemoglobin (CHAH). Colonies were counted after 48-72 h of growth at 37 °C. A-strain bacteria were grown, harvested and lysed within a BioSafety (BS) Level 3 containment facility. Bacterial cultures were harvested in 1 mL aliquots (∼109-1010 bacteria) by centrifugation and the pellets were washed three times with sterile, distilled water. Any remaining liquid was aspirated and the cell pellets were then resuspended in 200 µL of one of the seven protein solubilization described in Table 1. Resuspension of cell pellets in lysis solutions, except solution 2, resulted in immediate cell lysis. Cells resuspended in solution 2 required sonication, to affect a more complete lysis of cells. This was carried out on ice using a Sonic Dismembrator (Fisher Scientific, Ontario, Canada) using 0.5 s bursts for three cycles of 30 s. Cells were shaken for 30 min at room temperature and remaining unlysed cells removed by centrifugation at 14 000 × g for 10 min. Protein concentrations of the extracts were determined using the RC-DC protein assay (Bio-Rad, Hercules, CA) or using a modified Bradford Assay.22 Cells from a single liquid culture
were processed in parallel using solutions 1-7. Biological repeats were carried out weeks apart using different batches of broth culture inoculated with the same bacterial freezer stock. In each case growth conditions, such as culture volume, growth time and temperature were maintained as closely as possible between repeat experiments. Sodium Carbonate Extraction. Performed essentially as described by Fujiki et al.23 Bacterial cells were harvested from broth culture to give a final pellet containing ∼1010 bacteria. The pellet was washed twice with distilled water before finally resuspending in 4 mL of 50 mM Tris/HCL, pH 7.3 with 0.7 mg DNase I (Sigma). The cells were disrupted by sonication as described above and unbroken cells removed by centrifugation at 2500 × g for 10 min. The supernatant was diluted to a final volume of 50 mL with ice-cold 0.1 M sodium carbonate, pH 11. The resulting solution was stirred gently at 4 °C for 1 h. Carbonate treated membranes were collected by ultracentrifugation in a Beckman 55.2 Ti rotor at 100 000 × g for 1 h at 4 °C. The supernatant was discarded and membrane pellet resuspended in 5 mL of ice cold 50 mM Tris/HCl to remove contaminants, and then collected by centrifugation at 100 000 × g for 30 min. This wash procedure was repeated a second time, again discarding the supernatant. The final membrane protein containing pellet was solubilized for 2D electrophoresis with 1.0 mL of IEF solution (7 M urea, 2 M thiourea, 1% (w/v) ASB-14, 4% (w/v) CHAPS, 1% (w/v) DTT, and 0.5% (v/v) Biolytes 3-10 (Bio-Rad, Hercules, CA). Two-Dimensional Gel Electrophoresis and Image Analysis. The extracted proteins were separated using immobilized pH gradient strips (IPG), either linear pH 4-7, 17 cm (Biorad, Hercules, CA) or linear pH 6-11, 18 cm (Amersham Biosciences, Uppsula, Sweden). 100-300 µg of each protein solution was diluted with the appropriate solubilization buffer (solutions 1-7), with 0.5% v/v pH 3-10 Biolytes (Biorad, Hercules, CA) or pH 6-11 Pharmalytes (Amersham Biosciences, Uppsula, Sweden) and 0.003% Orange G. Proteins were loaded onto the IPG strips by in-gel rehydration overnight. Isoelectric focusing was conducted as described previously,24 using a Protein IEF Cell (Biorad, Hercules, CA). Briefly, the running conditions were as follows: 200 V for 1 h, 500 V for 1 h, 6 h ramp to 5000 V, hold at 5000 V for a total of 80 000100 000 Vh. These conditions were those routinely used within the laboratory, and optimized for this IEF cell, for focusing of 100-400 µg protein. Following IEF, the IPG strips were equilibrated for 25 min in equilibration buffer containing 2% SDS, 50 mM Tris/HCl pH 8.8, 6 M urea, 30% glycerol and 1% DTT. This was immediately followed by a second 25 min equilibration strip in the same solution containing 4% iodoacetamide in the place of DTT. The IPG strips were then rinsed in SDS gel running buffer (Biorad, Hercules, CA) and embedded onto a 12% homogeneous polyacrylamide gels (190 × 190 × 1.5 mm) with 1% agarose overlay with bromophenol blue (Biorad, Hercules, CA). Gels were run using PowerPac 1000 (Biorad, Hercules, CA) at 24 mA for 5 h with the temperature during electrophoresis maintained at 20 °C by water cooling. Sypro Ruby (Molecular Probes, Eugene, Oregon) staining was carried out as described in the manufacturer’s instructions. Images of gels were collected using FluorS (Biorad, Hercules, CA). Gels were subsequently stained with silver nitrate, to facilitate spot excision. Protein spots were matched using PDQuest software (Biorad, Hercules, CA), manually assigning landmark spots, to facilitate software matching. The normalized intensity for each spot was calculated as the ratio of the spot intensity versus the Journal of Proteome Research • Vol. 4, No. 5, 2005 1849
Francisella tularensis Proteomics
technical notes
Figure 1. 2DE of Total Proteome Extracts of Francisella tularensis Live Vaccine Strain, Extracted in Solutions Differing in Detergent Composition. Upper panel: 2DE show LVS total protein extract in (a) solution 3 and (b) solution 4. Arrows with numeric annotation indicate protein spots enriched in solutions of 0.5% ASB-14 (Solution 4). The spot numbers refer to Francisella tularensis locus tags and the corresponding protein identities listed in Table 1. Arrowed spots without FTT numbers were not able to be identified. Lower panel: Boxed regions A-C in Figure 1(b) are shown below in Panels A-C, with equivalent regions from 2DE of cells lysed in solutions 1-5, as indicated. The zoom panels highlight several points, including the increase in resolution of broad spots into multiple spots, in the higher molecular weight region of LVS in solution 1 (panel A) and the increasingly poor vertical spot resolution observed with decreasing molecular mass with the same solubilization solution (Solution 1, panels B and C). Arrows with numeric annotation indicate protein spots enriched in solutions of 0.5% ASB-14 (Solutions 4-5). Arrows without numeric annotation indicate equivalent positions in other gels where no corresponding spot intensity was observed. Separations were in the pH range 4-7, using 12% homogeneous polyacrylamide gels, with 200 µg of LVS total protein extract. Gels are stained with Sypro Ruby (Molecular Probes, Eugene, Oregon).
sum of the intensities of all spots present in the gel. Variations in spot intensity were calculated as the ratio of normalized intensities for each spot in gels of lysis solutions. Proteins deemed to have significant changes in intensity were selected based upon a consistent variation between biological repeats. An intensity variation greater than 1.5-fold or less than 0.6fold was considered significant for this analysis. In cases where spots could not be matched between gels, a list of unmatched spots was tabulated. In-Gel Digestion with Trypsin and Protein Identification. Selected protein spots were excised manually and processed using the Progest automated digestion unit (Genomic Solutions, Ann Arbor, MI) as described previously.24 The in-gel digests were analyzed by nano-liquid chromatography-MS/MS using a ‘CapLC’ capillary chromatography system (Waters, Milford, MA) coupled to a ‘QTOF Ultima’ hybrid quadrupole time-of1850
Journal of Proteome Research • Vol. 4, No. 5, 2005
flight mass spectrometer (Waters, Milford, MA). Peptide extracts were injected on a 75 µm internal diameter × 150 mm PepMap C18 nanocolumn (Dionex/LC packings) and resolved by gradient elution (5-75% acetonitrile, 0.l2% formic acid in 30 min, ∼350 nL/min). MS/MS spectra were acquired on doubly, triply and quadruply charged ions. Mascot Daemon (Matrix Science, London, UK) was used to search the Francisella strain Schu 4 genome sequence25 and proteins were subsequently identified by matching peptide sequences derived from MS/MS spectra. The high degree of similarity in gene content between Francisella subspecies26 meant that no problems were encountered in protein identification for either LVS or FSC033 data using Francisella strain Schu 4 genome sequence. Mass spectrometry results were evaluated based upon the Mascot score (>30), the number of peptides identified and the score for each identified peptide. The MS/MS spectra of positive results were manually
technical notes
Twine et al.
Figure 2. 2DE of Total Proteome Extracts of Francisella tularensis FSC033, extracted in Solutions Differing in Detergent Composition. 2DE in the pH range 4-7, 200 µg of FSC033 total protein extract in (a) Solution 3 (b) Solution 4. Gels are stained with Sypro Ruby (Molecular Probes, Eugene, Oregon). Spots annotated in (b) indicate spots that were observed to be present only in the presence of 0.5-1.0% ASB-14. The spot numbers refer to Francisella tularensis locus tags and the corresponding protein identities are listed in Table 1. Some unique spots could not be identified and therefore have not been numerically annotated.
checked to confirm matching. Molecular mass standards and position of protein spots of known pI were used to calibrate gels and allow estimation of the observed molecular mass and pI of protein spots.
Results and Discussion The principal aim of this work was to determine the most favorable detergent combination to affect a single step cellular lysis and protein solubilization for strains of the pathogen Francisella tularensis. These studies were initially conducted with the Live Vaccine Strain (LVS) and the more promising lysis and solubilization solutions were then used to examine the proteome of the more virulent A-strain, FSC033. In total 7 solutions were tested, whose composition are listed in Table 1. With the exception of solution 2, they differ only in the type and proportion of detergent added (CHAPS and/or ASB-14). The concentration of total chaotrope in solution 2 had to be lowered in order to facilitate solubilization of SB3-10. Lysis solutions 2 and 3 are routinely used in our laboratory,24,27 while solution 1 was based on protein solubilization solutions commonly used for the solubilization of enriched hydrophobic protein preparations.28 With the exception of solutions 2, all solutions were able to affect almost complete cellular lysis of LVS, while cellular lysis with solution 2 required sonication in order to achieve full solubilization. This phenomenon was attributed to the lower total chaotrope concentration of solution 2. All three solutions were compatible with isoelectric focusing (IEF) after the addition of carrier ampholytes and tracking dye. 2DE (pH 4-7) analysis of the LVS proteome using solutions 2 and 3 yielded gels of reasonable quality, with similar numbers of protein spots resolved (Table 1, 401 and 395 spots, respectively). Since the detergent SB3-10 appeared to offer no advantage over CHAPS and necessitated the use of lower chaotrope concentration (which can be detrimental to the maintenance of protein solubility29), its use was not pursued further. Solution 1 yielded the greatest number of spots on 2DE (595 spots). This was attributed to the resolution of broader spots into multiple isoforms and a reduction in horizontal streaking
in the high molecular weight region of the gel. The drawback of this solution was the comigration of the ASB-14 detergent front with the lower molecular weight proteins, which gave rise to poor vertical focusing of spots in the region of the gel at molecular weights less than 30 kDa (Figure 1, lower panel). Use of 1% ASB-14 in combination with CHAPS and as the sole detergent in the lysis solution (solutions 5 & 7) gave reasonable gels, but still suffered slightly from interference of ASB-14 detergent front with lower MW region of the gel, although comparable numbers of spots were observed, when comparing gels prepared with solution 3, 5, & 7 (Table 1). Solutions 5, with 0.5% ASB-14, 4% CHAPS and solution 6, with 0.5% ASB-14, did not suffer interference of ASB-14 detergent front. Figure 1 presents a summary of the most pertinent results. The upper panel of Figure 1, gels (a) and (b) show representative gels of total protein extracts of LVS lysed in solutions 3 and 4, respectively. At first glance, there appears to be only marginal differences in gel quality this pH range (4-7). Since these solutions differ only in the inclusion of 0.5% ASB-14 in solution 4, this may not be unexpected. Upon closer inspection, however, 24 unique spots were consistently observed in gels where proteins were solubilized in solution 4. The positions of the spots, which vary in molecular mass, pI and spot intensity, are annotated in gel b. The boxed areas, labeled A-C, are expanded in the lower panel of Figure 1 and include equivalent areas from equivalent LVS 2DE gels using solutions 1-5. The expanded regions clearly highlight the increased horizontal spot resolution in the high molecular weight region of the gel and also the poor vertical spot resolution at lower molecular weight when solution 1 is used. The same expanded regions also show the presence of protein spots obtained with solutions 4-5 (0.5%-1% ASB-14) that are not observed with solutions 2 and 3 (no ASB-14). The enriched spots are annotated on the solution 4-5 panels and their predicted equivalent positions denoted by arrows on the other gel images. These proteins were identified by nLC-MS/MS and will be discussed shortly. Lysis solutions 3-7 were tested with cell pellets of the virulent strain of the pathogen, FSC033, and yielded comparable results to those obtained with LVS. Spot numbers Journal of Proteome Research • Vol. 4, No. 5, 2005 1851
technical notes
Francisella tularensis Proteomics
Table 2. Identification of Proteins from 2DE of Strains of Francisella Visible in the Presence of Different Concentrations of the Detergents CHAPS and ASB-14h identified protein
MW, pI theoreticalb
sequence coveragec
FTT0049 NusA FTT0188 FtsZ FTT0209ci
55.1, 4.19
308, 31
39.7, 4.45
432, 46
33.8, 5.17
244,30
FTT0583i FopA FTT0615c
41.4, 5.15
301, 19
32.2, 4.47
405, 26
FTT0825ci FTT0842i
12.2, 4.49 23.2, 4.82
543, 53 457, 37
FTT0863ci LemA FTT0879i SodC FTT0897 PurK
22.0, 5.14
397,44
19.8, 5.36
174,18
40.3, 5.61
159, 12
FTT0900
13.9, 4.85
121, 8
FTT0904 LpnB FTT0923 FTT0955c Gor
17.3, 4.71
672, 50
83.5, 7.60 49.5, 5.79
502, 63 73, 5
FTT0991 FTT1088c FTT1103i
21.0, 7.61 28.2, 4.90 38.7, 4.91
195, 37 195, 21 522,41
FTT1155c AroK FTT1346i FTT1351 FTT1416ci
19.7, 4.99
412, 39
Superoxide dismuate (Cu-Zn) precursor Phosphoribosylaminoimidazole carboxylase, ATPase subunit Conserved hypothetical membrane protein Conserved hypothetical lipoprotein (TUL4) Hypothetical protein Pyruvate/2-oxoglutarate dehydrogenase complex, dihydrolipoamide dehydrogenase component Hypothetical lipoprotein Hypothetical protein Conserved hypothetical lipoprotein Shikimate kinase I
14.5, 8.33 24.6, 4.76 14.9, 9.13
159, 26 89, 5 381, 45
Hypothetical protein Hypothetical protein Hypothetical lipoprotein
YP_170298.1 YP_170303.1 YP_170359.1
a
locus tag
protein name
N utilization substance protein A Cell division protein Periplasmic solute binding family protein Outer membraneassociated protein Metal ion transporter protein Hypothetical protein Peptidoglycanassociated lipoprotein LemA-like protein
accession no.d
straine
TMRf
YP_169124.1
FSC033
0
cytoplasm
YP_169249.1
FSC033
1
inner membrane
YP_169268.1
LVS
1
periplasm
YP_169607.1
LVS FSC033 FSC033
0
outer membrane
1
inner membrane
0 1
periplasm inner membrane
1
inner membrane
1
inner membrane
0
cytoplasm
YP_169636.1 YP_169832.1 YP_169847.1 YP_169865.1 YP_169879.1 YP_169894.1
YP_169897.1
LVS
0
cytoplasm
YP_169901.1
FSC033
0
YP_169919.1 YP_169945.1
FSC033 LVS
1 1
periplasm or attached to OM by lipid anchor inner membrane inner membrane
YP_169979.1 YP_170067.1 YP_170079.1
FSC033 LVS LVS FSC033 FSC033
0 0 0
outer membrane cytoplasm periplasm
0
cytoplasm
LVS LVS LVS FSC033 LVS FSC033 LVS
1 1 0
inner membrane inner membrane inner or outer membrane
1
inner membrane
1
periplasm
2
inner membrane
YP_170531.1
LVS FSC033 LVS
0
cytoplasm
YP_170579.1
FSC033
0
outer membrane
YP_170626.1|
LVS
0
periplasm
YP_170667.1
LVS FSC033
1
outer membrane
YP_169607.1
FTT1531 FadA FTT1567ci
41.7, 5.70
201, 21
3-ketoacyl-CoA thiolase
YP_170461.1
22.5, 5.07
73, 7
YP_170489.1
FTT1591i
41.6, 4.32
101, 12
Hypothetical membrane protein Lipoprotein
FTT1616 CysS FTT1673 RibA
53.1, 5.06
227, 23
44.6, 5.88
471, 37
FTT1724c TolC FTT1778ci
57.2, 4.93
159, 23
13.7, 8.82
463, 37
Cysteinyl-tRNA synthetase Riboflavin biosynthesis protein ribA/GTPcyclohydrolase II Outer membrane protein tolC precursor Hypothetical membrane protein
LVS LVS FSC033 LVS FSC033 LVS FSC033 LVS FSC033
localizationg
YP_170509.1
a Locus tag according to genome sequence of Francisella tularensis SchuS425 and abbreviated protein name where applicable. b Theoretical mass (kDa) and pI calculated from translated gene sequence.25 c Mascot score, sequence coverage (%). d Accession number according to NCBI. e Strain in which the protein was enriched. f Predicted Transmembrane Regions (TMR), from predicted amino acid sequences using PSORT algorithm.30 g Predicted subcellular location based upon predicted amino acid sequence. Calculated using PSORT algorithm.30 h Peptide mass spectra of tryptic digests of excised spots were recorded on a QTOF ULTIMA nLCMS/MS mass spectrometer (Waters, Milford, MA). Proteins were identified from MSMS spectra using MASCOT software. i Protein spots also detected in 2DE of sodium carbonate preparations of LVS membranes.
observed for gels in the pH range 4-7 are listed in Table 1 and are comparable for FSC033 lysed in solutions 4-7. Figure 2 shows representative gels from FSC033 solubilized in solutions 3 and 4. Here again, as observed with LVS, the gels produced using solutions 5-7 suffered from a slight decrease in vertical spot resolution (data not shown). In qualitative terms, the gels 1852
Journal of Proteome Research • Vol. 4, No. 5, 2005
do not appear different but, upon closer analysis, additional protein spots, varying in molecular mass and pI, were observed when cells were lysed in solutions containing ASB-14. In total, 24 unique spots were consistently located in LVS gels (Figure 1, gel b) and 21 in FSC033 gels (Figure 2, gel b) when the lysis solution contained ASB-14. The unique spots were excised, in-
technical notes
Figure 3. Representative 2DE of Sodium Carbonate Membrane protein fraction of Francisella LVS. 2DE in the pH range 4-7, 200 µg of LVS membrane enriched fraction. Spots annotated in indicate enriched protein spots observed in 2DE of LVS total protein lysates in the presence of ASB-14, which were also located in 2DE of LVS membrane fraction. Spot numbers refer to Francisella tularensis locus tags and the corresponding protein identities are listed in Table 1.
gel digested and identified by nLC-MS/MS (Table 2). In cases where the spot intensity was too weak to permit protein identification, the spot is indicated with an arrow but without an identification number. In total, 20 unique proteins in the case of LVS and 18 for FSC033 (Table 2) and 10 of these proteins were common to both strains. Of the proteins listed, 6 were annotated in the genome as lipoproteins, 5 as membrane proteins and 5 hypothetical proteins. The PSORT algorithm was used to search for possible commonalities in the proteins, giving predictions of the number of potential transmembrane domain (TMDs) and putative subcellular location proteins based upon their amino acid sequences.30 The results of these analyses for proteins identified as being enriched in the presence of ASB-14, are listed in Table 2. Ten of the enriched LVS proteins were predicted to have TMDs, with 10 predicted to be associated with the bacterial inner membrane, 3 with the bacterial outer membrane, 5 with the periplasm and 4 cytoplasmic proteins. Of the enriched proteins observed in the virulent strain FSC033, 9 were predicted to have TMD’s, with 10 associated with the bacterial inner membrane, 1 with the periplasm, 4 with the outer membrane and two cytoplasmic proteins. No more than two TMDs were predicted for the enriched proteins, indicating that highly hydrophobic transmembrane proteins with multiple TMDs are still not compatible with our extraction and separation conditions. This is not surprising, since highly hydrophobic proteins with multiple transmembrane spanning regions are rarely observed on 2DE. From the PSORT prediction, it became apparent that as many as 60% of the proteins enriched by ASB-14 were predicted to be targeted to the inner or outer bacterial membrane, suggesting that inclusion of relatively small amounts of ASB-14 in the lysis and solubilization solution increases the proportion of the membrane proteome visible in 2DE of total protein extracts. To determine how many of the enriched spots were, in fact, membrane or membrane-associated proteins, a membrane enriched fraction of the LVS proteome was prepared
Twine et al.
using the alkaline sodium carbonate extraction procedure.31 This has previously been used to extract membrane proteins from E. coli and other bacteria strains, with minimal contamination from cytoplasmic proteins.31-34 Good quality 2D-gels were obtained with the carbonate extract and a representative gel in shown is Figure 3. A number of other gel spots were excised and all were identified as membrane or membrane associated proteins, indicating that the method gave acceptably pure membrane preparations for F. tularensis. The carbonate gels were matched to the LVS total protein extract gels prepared with solution 4. 12 of the 24 enriched proteins observed in the latter gels were also observed in 2D-gels of the carbonate fraction, indicating that they are highly likely to be membrane associated. The identifications were confirmed by in-gel digestion and nLC-MSMS analysis of the carbonate protein spots (as indicated in Table 2). 8 of the 12 proteins detected in LVS membrane preparation were commonly enriched by ASB-14 in both LVS and FSC033 strains of Francisella. Due to the difficulties and restrictions posed by working within the Biosafety Level 3 facility, membrane preparations of the virulent FSC033 strain were not prepared. These result lend further proof that low concentrations of ASB-14 in the protein lysis and solubilization solution enriches for a larger proportion of membrane and membrane associated proteins. Of those proteins that were enriched by ASB-14 only one, FTT1103, has been previously reported to be identified in F. tularensis samples prepared by other methods.11
Conclusions In this technical note, we describe our efforts to determine the best detergent combination for rapid, single-step lysis and protein solubilization of Live Vaccine Strain and virulent A-strain of Francisella tularensis. A solution containing 7 M urea, 2 M thiourea, 1% DTT, 0.5% ASB-14, and 4% CHAPS appears to provide the best conditions for total proteome analysis by 2DE using a single step lysis and extraction procedure that is completely compatible with IEF. In addition, inclusion of low concentrations of ASB-14 also improves the extraction and/or the resolution of membrane and membrane associated proteins. We have also determined that this optimized solution is compatible with the 2DE analysis of basic proteins (pH 6-11, data not shown) and narrower pH ranges (for example pH 3-6). While we do not suggest that this method is superior to proteome fractionation techniques where protein quantity is not a limiting factor, we believe that it is valuable in cases when sample quantity is restricted and sufficient protein for only 1-2 gels is available. We have applied this methodology to strains of Francisella isolated from in vivo sources and have resolved upward of 400 spots in the neutral/ acidic pH range. These results will be published in a future paper when the role of the identified proteins in the physiology of in vivo pathogen survival has been determined.
Acknowledgment. The authors thank Wen Ding and Luc Tessier for technical assistance with mass spectrometers, Hua Shen and Rhonda Kuo Lee for assistance with studies conducted within the BS Level 3 facility. This work was funded by the National Research Council’s Genome and Health Initiative NRC manuscript no. 42503. References (1) Dienst, F. T., Jr. J. La State Med. Soc. 1963, 115, 114-127.
Journal of Proteome Research • Vol. 4, No. 5, 2005 1853
technical notes
Francisella tularensis Proteomics (2) Ellis, J.; Oyston, P. C.; Green, M.; Titball, R. W. Clin. Microbiol. Rev. 2002, 15, 631-646. (3) Tarnvik, A.; Berglund, L. Eur. Respir. J. 2003, 21, 361-373. (4) Oyston, P. C.; Sjostedt, A.; Titball, R. W. Nat. Rev. Microbiol. 2004, 2, 967-978. (5) Tarnvik, A.; Priebe, H. S.; Grunow, R. Scand. J. Infect. Dis. 2004, 36, 350-355. (6) Cronquist, S. D. Dermatol. Clin. 2004, 22, 313-vii. (7) Dennis, D. T.; Inglesby, T. V.; Henderson, D. A.; Bartlett, J. G.; Ascher, M. S.; Eitzen, E.; Fine, A. D.; Friedlander, A. M.; Hauer, J.; Layton, M.; Lillibridge, S. R.; McDade, J. E.; Osterholm, M. T.; O’Toole, T.; Parker, G.; Perl, T. M.; Russell, P. K.; Tonat, K. JAMA 2001, 285, 2763-2773. (8) Conlan, J. W. Expert. Rev. Vaccines 2004, 3, 307-314. (9) Hernychova, L.; Stulik, J.; Halada, P.; Macela, A.; Kroca, M.; Johansson, T.; Malina, M. Proteomics 2001, 1, 508-515. (10) Hubalek, M.; Hernychova, L.; Havlasova, J.; Kasalova, I.; Neubauerova, V.; Stulik, J.; Macela, A.; Lundqvist, M.; Larsson, P. J. Chromatogr. B Anal. Technol. Biomed. Life Sci. 2003, 787, 149177. (11) Hubalek, M.; Hernychova, L.; Brychta, M.; Lenco, J.; Zechovska, J.; Stulik, J. Proteomics 2004, 4, 3048-3060. (12) Ericsson, M.; Tarnvik, A.; Kuoppa, K.; Sandstrom, G.; Sjostedt, A. Infect. Immun. 1994, 62, 178-183. (13) Golovliov, I.; Ericsson, M.; Sandstrom, G.; Tarnvik, A.; Sjostedt, A. Infect. Immun. 1997, 65, 2183-2189. (14) Havlasova, J.; Hernychova, L.; Halada, P.; Pellantova, V.; Krejsek, J.; Stulik, J.; Macela, A.; Jungblut, P. R.; Larsson, P.; Forsman, M. Proteomics 2002, 2, 857-867. (15) Kovarova, H.; Halada, P.; Man, P.; Golovliov, I.; Krocova, Z.; Spacek, J.; Porkertova, S.; Necasova, R. Proteomics 2002, 2, 8593. (16) Hecker, M.; Schumann, W.; Volker, U. Mol. Microbiol. 1996, 19, 417-428. (17) Hecker, M. Adv. Biochem. Eng Biotechnol. 2003, 83, 57-92. (18) Hecker, M.; Volker, U. Proteomics 2004, 4, 3727-3750. (19) Schweder, T.; Hecker, M. Adv. Biochem. Eng Biotechnol. 2004, 89, 47-71. (20) Wilkins, M. R.; Gasteiger, E.; Sanchez, J. C.; Bairoch, A.; Hochstrasser, D. F. Electrophoresis 1998, 19, 1501-1505. (21) Santoni, V.; Molloy, M.; Rabilloud, T. Electrophoresis 2000, 21, 1054-1070.
1854
Journal of Proteome Research • Vol. 4, No. 5, 2005
(22) Ramagli, L. S. Methods Mol. Biol. 1999, 112, 99-103. (23) Fujiki, Y.; Hubbard, A. L.; Fowler, S.; Lazarow, P. B. J. Cell Biol. 1982, 93, 97-102. (24) Zhang, R.; Tremblay, T. L.; McDermid, A.; Thibault, P.; Stanimirovic, D. Glia 2003, 42, 194-208. (25) Larsson, P.; Oyston, P. C.; Chain, P.; Chu, M. C.; Duffield, M.; Fuxelius, H. H.; Garcia, E.; Halltorp, G.; Johansson, D.; Isherwood, K. E.; Karp, P. D.; Larsson, E.; Liu, Y.; Michell, S.; Prior, J.; Prior, R.; Malfatti, S.; Sjostedt, A.; Svensson, K.; Thompson, N.; Vergez, L.; Wagg, J. K.; Wren, B. W.; Lindler, L. E.; Andersson, S. G.; Forsman, M.; Titball, R. W. Nat. Genet. 2005, 37, 153159. (26) Broekhuijsen, M.; Larsson, P.; Johansson, A.; Bystrom, M.; Eriksson, U.; Larsson, E.; Prior, R. G.; Sjostedt, A.; Titball, R. W.; Forsman, M. J. Clin. Microbiol. 2003, 41, 2924-2931. (27) Carrillo, C. D.; Taboada, E.; Nash, J. H.; Lanthier, P.; Kelly, J.; Lau, P. C.; Verhulp, R.; Mykytczuk, O.; Sy, J.; Findlay, W. A.; Amoako, K.; Gomis, S.; Willson, P.; Austin, J. W.; Potter, A.; Babiuk, L.; Allan, B.; Szymanski, C. M. J. Biol. Chem. 2004, 279, 20327-20338. (28) Henningsen, R.; Gale, B. L.; Straub, K. M.; DeNagel, D. C. Proteomics 2002, 2, 1479-1488. (29) Navarrete, R.; Serrano, R. Biochim. Biophys. Acta 1983, 728, 403408. (30) Gardy, J. L.; Laird, M. R.; Chen, F.; Rey, S.; Walsh, C. J.; Ester, M.; Brinkman, F. S. Bioinformatics 2005, 21, 617-623. (31) Molloy, M. P.; Herbert, B. R.; Slade, M. B.; Rabilloud, T.; Nouwens, A. S.; Williams, K. L.; Gooley, A. A. Eur. J. Biochem. 2000, 267, 2871-2881. (32) Molloy, M. P.; Herbert, B. R.; Walsh, B. J.; Tyler, M. I.; Traini, M.; Sanchez, J. C.; Hochstrasser, D. F.; Williams, K. L.; Gooley, A. A. Electrophoresis 1998, 19, 837-844. (33) Nouwens, A. S.; Cordwell, S. J.; Larsen, M. R.; Molloy, M. P.; Gillings, M.; Willcox, M. D.; Walsh, B. J. Electrophoresis 2000, 21, 3797-3809. (34) Molloy, M. P.; Phadke, N. D.; Chen, H.; Tyldesley, R.; Garfin, D. E.; Maddock, J. R.; Andrews, P. C. Proteomics 2002, 2, 899910.
PR050102U
