2-DE analysis indicates that Acinetobacter baumannii displays a robust and versatile metabolism

Background Acinetobacter baumannii is a nosocomial pathogen that has been associated with outbreak infections in hospitals. Despite increasing awareness about this bacterium, its proteome remains poorly characterised, however recently the complete genome of A. baumannii reference strain ATCC 17978 has been sequenced. Here, we have used 2-DE and MALDI-TOF/TOF approach to characterise the proteome of this strain. Results The membrane and cytoplasmatic protein extracts were analysed separately, these analyses revealed the reproducible presence of 239 and 511 membrane and cytoplamatic protein spots, respectively. MALDI-TOF/TOF characterisation identified a total of 192 protein spots (37 membrane and 155 cytoplasmatic) and revealed that the identified membrane proteins were mainly transport-related proteins, whereas the cytoplasmatic proteins were of diverse nature, although mainly related to metabolic processes. Conclusion This work indicates that A. baumannii has a versatile and robust metabolism and also reveal a number of proteins that may play a key role in the mechanism of drug resistance and virulence. The data obtained complements earlier reports of A. baumannii proteome and provides new tools to increase our knowledge on the protein expression profile of this pathogen.


Background
Acinetobacter baumannii is a Gram-negative, nonmotile, aerobic coccobacillus that is often found in health care settings. There is increasing concern as regarding the prevalence of this bacterium in hospital environments, especially in intensive care units, where it has been associated with different types of infections such as pneumonia, meningitis, bacteraemia, urinary tract infections and oth-ers [1][2][3]. In addition, A. baumannii has a remarkable capacity to acquire resistance to most antibiotics used in clinical practice and to cause outbreaks throughout cities, countries and even continents [4][5][6]. Despite the emerging awareness of the potential hazards of this bacterium, little is known about the mechanisms that operate during antibiotic resistance, virulence, or persistence strategies of A. baumannii.
Genomic approaches have recently been successfully applied for A. baumannii. For instance, genomic comparison between SDF and AYE strains with the soil-living A. baylyi strain ADP1 revealed important exclusive features of each strain, which may partly explain their existence in different ecological niches [7,8]. The recently described sequence of A. baumannii ATCC 17978 isolate revealed that its genome is composed by 3,976,746 base pairs (bp) and that it has 3,830 open reading frames (ORFs), of which nearly 17% are situated in 28 putative alien islands [9], which suggests that the genome has acquired a significant amount of foreign DNA.
Nevertheless, it is widely accepted that most ORFs are expressed under different conditions and/or states of growth and environmental stress [10]. Moreover, the identification of proteins via MS/MS analyses validate predicted genes, correct erroneous gene annotation and reveal some completely missed genes [10,11]. Therefore as pointed out by Brötz-Oesterhelt et al. (2005) [12], the application of further techniques such as proteomics is required in order to establish the protein composition of a given cell under certain conditions, independently of its linear gene sequence.
Proteomics analysis has been applied to fractions enriched with A. baumannii cell membranes, and a group of 22 proteins, composed by ribosomal proteins, chaperones elongation factors and outer membrane proteins identified [13]. A total of 135 proteins (inner and outer membrane) and 23 periplasmic proteins were identified in A. baumannii in another study [14]. More recently, it was characterise the proteome of outer membrane vesicles from a clinical A. baumannii isolate [15]. These studies provide important data regarding the proteome of Acinectobacter, nevertheless as far as we are aware, the cytoplasmatic proteome of A. baumannii remains to be characterised. In the current study, we used 2-DE and MALDI-TOF/TOF to analyse the membrane and cytoplasmatic proteome of A. baumannii, and the data obtained confirms the results of previous genomic studies showing that A. baumannii displays a robust and versatile metabolism capable of exploiting a variety of carbon sources and energy.

Bacterial strain and growth conditions
A. baumannii ATCC 17978 was grown overnight in Mueller Hinton broth (Fluka, Madrid, Spain) at 37°C under constant shaking. Fresh nutrient media (500 mL) was inoculated with a 1:100 dilution of the overnight culture and grown to OD 600 = 0.4-0.6, at 37°C with vigorous shaking.

Protein extraction
The cell were harvested by centrifugation at 3,500 g for 15 min at 4°C and washed twice with 10 mL 0.9% (w/v) NaCl. The resultant pellet was resuspended in 3 or 5 mL of disintegration buffer [13] (7.8 g/L NaH 2 PO 4 , 7.1 g/L Na 2 HPO 4 , 0.247 g/L MgSO4 7.H 2 O + protease inhibitor mix (GE Healthcare, USA) + nuclease mix (GE Healthcare, USA)) and sonicated on ice for 3 periods of 5 min. The unbroken cells were separated by centrifugation at 1,500 g. The supernatant was centrifuged for 30 min at 4°C at 4,500 rpm and was then clarified through a 0.45 μM filter (Milipore, USA) to remove the cell debris. Finally, the extract was processed with a 2-DE Cleanup Kit (GE Healthcare, USA).
The same extraction method was used to extract cell surface membrane. However, after the separation of unbroken cells, the lysate was treated as described by Molly et al. (2000) [16]. Briefly, an equal volume of ice-cold 0.1 M sodium carbonate (pH 11) was added to the resulting supernatant and the mixture was stirred slowly overnight, on ice. The carbonate treated membranes were then collected by ultracentrifugation at 100,000 g for 45 min at 4°C, and the membranes were then re-suspended in 500 μl H 2 O. Finally, as with the soluble fraction, the extract was processed with a 2-DE Cleanup Kit (GE Healthcare, USA).

Two-dimensional gel electrophoresis (2-DE)
Protein concentration in the extracts was determined with a Biorad protein assay kit (Biorad, Germany), by a modified Bradford assay [17] as suggested by Ramagli et al. (1999) [18].
For isoelectric focusing (IEF), the IPGphor III system was used (GE Healthcare, USA) with 3-10 non-linear (NL), 4-7 (L) or 6-11(L) pH gradient strips (IPG strips, GE Healthcare, Sweden). Proteins were solubilised in 8 M urea, 2% (w/v) CHAPS, 40 mM DTT and 0.5% (v/v) corresponding IPG buffer (GE Healthcare, Sweden). IEF was carried out at 30 V for 12 h, followed by 250 V for 1 h, 500 V for 1.5 h, 1,000 V for 1.5 h, a gradient to 8,000 V over 1.5 h and maintenance at 8,000 V for a further 4 h, all at 20°C. Note, for 6-11 (L) pH gradient strips, IEF was carried out at 30 V for 12 h, followed by 500 V for 1 h, 1,000 V for 1 h, a gradient to 8,000 V for 2.5 h and maintenance at 8,000 V for 0.5 h. Prior to the second dimension (SDS-PAGE), the focused IPG strips were equilibrated for 2 × 15 min in buffer containing 50 mM Tris-HCl (pH 8.8), 6 M urea, 30% (v/v) glycerol, 2% (w/v) SDS and a trace of Bromophenol Blue. DTT at 1% (w/v) was added to the first equilibration step and 2.5% (w/v) iodoacetamide to the second. SDS-PAGE was performed on 12% or 15% polyacrylamide gels [19]. For analytical 2-DE gels, silver staining was performed according to Blum et al. (1987) [20] and gels were loaded with 25 to 40 μg of total protein. For preparative gels, MS-compatible silver staining [21] was used and the gels were loaded with at least 350 μg of total protein.

Image acquisition and 2-DE analyses
Gels were scanned with an Image Scanner v3.3 densitometer (GE Healthcare) and analysed with Image Master Platinum software, V.6.0, as described by the manufacturer (GE Healthcare). Briefly, a matched set consisting of three images was created. In order to confirm reproducibility, at least three biological replicates were processed. For spot detection, the parameters were adjusted in the following order: smooth 2; MinArea 5; Saliency 1.00000, and only spots observed on all three gels of a replicated group were considered for further analyses. Numerical data for individual spots were generated automatically and expressed as % of spot volume.

Trypsin digestion of proteins and characterisation by MALDI-TOF/TOF
Selected spots were manual excised from the gels and transferred to microcentrifuge tubes. Samples were then destained with a solution containing 20% (w/v) sodium thiosulphate and 1% (w/v) potassium ferricyanide for 5 min. Destained spots were then in-gel reduced, alkylated and digested with trypsin as suggested be Sechi and Chait 1998 [22]. Briefly, spots were washed twice with 25 mM ammonium bicarbonate in 50% (v/v) acetonitrile (ACN) for 20 min. The gel spots were then shrunk with 100% (v/ v) in acetonitrile and dried in a speed-vac (Savant, USA). The samples were reduced with DTT and subsequently alkylated with iodoacetamide. Samples were digested with 20 ng/μl sequencing grade trypsin (Roche Applied Science, USA), overnight at 37°C. After digestion the supernatant was collected and 1 μl was spotted onto a MALDI target plate (384-spot Teflon ® -coated plates) and allowed to air dry at room temperature. Where necessary, tryptic peptides were passed through C18 Zip-Tips (Millipore, USA) and mixed with 3 mg/ml of an α-cyano-4hydroxycinnamic acid in 0.1% (v/v) TFA and 50% (v/v) ACN was added to the dried peptide digest spots and allowed to air dry. The samples were analyzed using a MALDI-TOF/TOF mass spectrometer 4800 Proteomics Analyzer (Applied Biosystems). MALDI-TOF spectra were acquired in reflector positive ion mode using 1000 laser shots per spectrum. Data Explorer version 4.2 (Applied Biosystems) was used for spectra analyses and generating peak picking list. All mass spectra were internally calibrated using autoproteolytic trypsin fragments and externally calibrate using standard peptide mixture (Sigma-Aldrich). TOF/TOF fragmentation spectra were acquired by selecting the 10 most abundant ions of each MALDI-TOF peptide mass map (excluding trypsin autolytic pep-tides and other background ions) and averaging 2,000 laser shots per fragmentation spectrum. The parameters used to analyze the data were a signal to noise threshold of 20, minimum area of 100 and a resolution higher than 10,000 with a mass accuracy of 20 ppm.

Database queries and protein identification
The monoisotopic peptide mass fingerprinting data obtained from MS and the amino acid sequence tag obtained from each peptide fragmentation in MS/MS analyses were used to search for protein candidates using Mascot version 1.9 from Matrix Science http:// www.matrixscience.com. Peak intensity was used to select up to 50 peaks per spot for peptide mass fingerprinting and 50 peaks per precursor for MS/MS identification. Tryptic autolytic fragment-, keratin-, and matrix-derived peaks were removed from the data set utilised for databased search. The search for peptides mass fingerprints and tandem MS spectra were performed in NCBInr database without any taxonomy restriction. Fixed and variable modifications were considered (Cys as S-carbadiomethyl and Met as oxidised methionine, respectively), allowing one trypsin missed cleavage. MS/MS ions search were conducted with a mass tolerance of ± 1.2 Da on the parent and 0.3-0.8 Da on fragments, in all cases the peptide charge was +1. Decoy search was done automatically by Mascot on randomized database of equal composition and size. Mascot scores for all protein identifications were higher than the accepted threshold for significance (at the p < 0.050 level, positive rate measured to be 0.047). For protein subcellular localisation prediction it was utilised PSORTb v.2.0.4, available free online at http:// www.psort.org/psortb/index.html.

Bimodal distribution of Acinetobacter baumannii proteome
In the current study, a 2-DE based proteomic analyses of A. baumannii ATCC 17978, grown under rich medium conditions and harvested during exponential phase were performed. The study was based on the analysis of two distinct protein fractions, a fraction enriched in membrane proteins and a fraction enriched in cytoplasmatic proteins; for convenience, these are referred in the text as membrane and cytoplasmatic fractions respectively. Analysis of three biological replicates revealed that at least 239 spots were reproducibly detected in the membrane fraction, whereas the cytoplasmatic fraction contained a total of 511 reproducible spots.
As a first approach, both the membrane and the cytoplasmatic protein fractions were analysed by use of 2-DE gels with immobilized pH gradient (IPG) strips (pH 3-10) (see Fig. 1A and 1B, respectively). However, for the cytoplasmatic fraction, these analyses revealed a clear predomi-nance of acidic proteins, although a number of highly basic proteins remained unresolved close to the cathode (pI > 9) (see Fig. 1B). Therefore, in an attempt to overcome the intrinsic under-representation of the alkaline proteins [12,23], the acidic and basic cytoplasmatic subproteomes were analysed using immobilized pH gradient strips of various pH range (see Fig. 2 and 3). Use of 2-DE 12% polyacrylamide (pI ranges of 4-7) silver stained gels revealed 390 spots that could be reproducibly detected on the acidic range ( Fig. 2A). Additionally, complementary analyses using 15% polyacrylamide silver stain gels in the range pH 3-10 revealed the presence of further 30 protein spots of low molecular weight (between 9 and 12 kDa) (Fig. 2B). Use of 2-DE gels 12% polyacrylamide (pI ranges of 6-11) added a further 91 reproducible alkaline protein spots (nearly 21% of the total soluble protein spots) (Fig. 3).
In concordance with previous reports referring other bacteria species [23], the 2-DE gels analysis revealed that the spot pattern of this pathogen is characterised by clear predominance of high molecular protein spots and in terms of pI there is visible a bimodal distribution of protein spots, where the most crowded regions is found between pI 4-7 and 9-11 (Fig. 4).  The cytoplasmatic fraction was analysed with IPG strips of various pH range and silver stained gels with different con-centrations of acrylamide    S1, Additional file 1 and table  S2, Additional file 2). These observations are a common feature of 2-DE gels [14,24], and possibly arise from the heterogeneity in post-translation modifications. Therefore within this context, membrane and cytoplasmatic fractions together contained 153 unique proteins. Other spots shared homology with the same protein families (e.g. spot # 1, spot # 2 and spot # 35), but with different corresponding gene sequences, indicating the presence of multi-gene families in the A. baumannii proteome.

2-DE gels showing A. baumannii proteins
The results demonstrate that the A. baumannii proteome is diverse in nature, and can be conveniently divided into at least 16 distinct functional groups, namely: amino acid transport and metabolism; carbohydrate transport and metabolism; cell division and chromosome partitioning; coenzyme metabolism; cell envelope and outer membrane biogenesis (transport); defence; energy production and conversion; general function predicted only; multifunctional; nucleotide transport and metabolism; postranslation modification, protein turnover, chaperones; redox reactions; RNA and protein synthesis; signalling; translation, ribosomal structure and biogenesis and transcription (see Fig. 5), in addition there was further 4 hypothetical proteins with unknown function. Whenever possible the proteins were classified according to the clusters of orthologous groups (COG) [25].   Curiously, when comparing the data here presented with previous reports, it become visible that several proteins here identified are also present in the membrane proteome of other A. baumannii strains, including clinical isolates [13][14][15] (see table S2, Additional file 2). It is therefore tempting to propose that these proteins are constitutively present in the membrane of A. baumannii and if that so, then this valuable information can be utilised for development of new drugs, vaccines and/or diagnostic.

2-DE (12% SDS) gel containing basic proteins within a pH range of 6-11
A number of Omps associated with the transport across the membrane of several compounds were identified in the present study. Including proteins such as putative copper receptor (OprC, spot # 459), which allows the penetration of small cations, a putative Omp (OprD, spot # 465) that in Pseudomonas aeruginosa has been demonstrated to be involved in the uptake of basic amino acids, small peptides and of imipenem [27]; putative glucosesensitive porin (OprB-like spot # 466), this porin is often referred to as carbohydrate selective porin, since it acts as a central component of glucose, mannitol, glycerol, and fructose transport across the outer membrane [28]; Omp 38 precursor (spot # 467) belonging to the OmpA family, which has been suggested to be involved with the transport of β-lactams and saccharides up to approximately 800 Da [29,30]; an Omp (with homology with Omp 33-36 kDa, spot # 468) involved in the transport of water and carbapenems [31]; Omp CarO (spot # 469 & spot # 472), which is essential for the uptake of L-ornithine and participates in the selective uptake of carbapenems and other basic amino acids in A. baumannii [32].
Hence, the results indicate that the most abundant membrane proteins were those involved in the transport of dif- ferent components (amino acid, sugars, and fatty acids), which suggests that A. baumannii have an adaptable metabolism that utilises a wide range of carbon sources.

Cytoplasmatic extract
MALDI-TOF/TOF analyses of cytoplasmatic extract revealed that of the 155 identified spots, 12 were identified as hypothetical proteins (see table S1, Additional file 1) and of those 9 could be functionally classified according COG (information obtained by MASCOT analyses). The three most representative groups are those related to amino acid and transport metabolism (29), translation, ribosomal structure and biogenesis (28) and energy production and conversion (14). Also there are other important groups (see table S1, Additional file 1 and Fig. 5).
Analyses of the cytoplasmatic proteins revealed that much of the A. baumannii proteome is dedicated to the metabolism of a wide range of components. One of the most highly represented classes of proteins was that involved in amino acid transport and metabolism. . In addition, the multifunctional proteins included proteins such as branched-chain amino acid transferase, which catalyses the transamination of the branched-chain amino acids (spot # 132) leucine, isoleucine and valine to their respective alpha-keto acids [33]. Taking into consideration the number, level of expression (in a population with an average of % spot volume of 0.21 there was: spot # 67 0.6 ± 0.04; spot # 123 0.55 ± 0.03; spot # 285 1.13 ± 0.36; see also Fig. 2) and the diversity of proteins within this group, it is reasonable to conclude that the amino acid metabolism is in fact a major pillar of the overall A. baumannii metabolism.
Interestingly, a phospho-2-dehydro-3-heoxyheptonate aldolase (spot # 123) was identified; in E. coli this type of synthetase catalyses the first step in aromatic amino acid biosynthesis from chorismate [34]. Earlier studies have demonstrated that species of the genus Acinetobacter are capable of degrading aromatic compounds [24,[35][36][37]. The capacity of Acinetobacter spp. to degrade aromatic compounds is of particular interest for bioremediation studies [38] and also for the development of new antibiotics [24]. Therefore, the protein here identified and its possible involvement in aromatic compound degradation may be subject of further investigation.
As mentioned above, a low molecular weight protein named nitrogen assimilation regulatory protein P-II 2 (spot # 352) was identified. In E. coli this protein regulates the assimilation of nitrogen by modulating the activity of glutamase synthase [39], this regulatory mechanism allows the cell to adjust the glutamase synthase activity quickly -in response to alterations in the abundance of the preferred nitrogen source, ammonia [39]. Although further investigation is required, this suggests that nitrogen is another source of energy utilised by this microorganism.
In concordance with previous genomic studies [8,9,40] that indicated the inability of A. baumannii to catabolise glucose, the current proteomic analysis did not detect any enzymes such as hexokinase, glucosekinase or proteins of the phosphotransferase transport system essential for glycolysis. However, it has to be referred the already mentioned presence of the putative glucose porin sensitive outer membrane (OprB-like) suggests the uptake of glucose from A. baumannii. Hence, this information supports the existence of a route passing through the Entner-Doudoroff pathway [41], in which the oxidation of glucose in the periplasm leads to pyruvate formation. Moreover, the presence of both fructose-1,6-bisphosphatase (spot # 138) and fructose-1,6-bisphosphate aldolase (spot # 156), shows that the bacterium in question is capable of catabolising fructose. Hence, it would be interesting to investigate further the capability of these and other sugars to support growth of this bacterium.
Identification of the selected alkaline spots revealed that most were ribosomal proteins (see table S1, Additional file 1) as previously described in other microbe studies [42]. In addition, to ribosome-associated proteins, several other proteins of the translational machinery were also detected. For instance, translation elongation factors EF-G (spot # 5), EF-Tu (spot # 100), EF-Ts (spot # 178), EF-P (spot # 235) and ribosome releasing factor (spot # 248). Ribosomes, ribosome-associated proteins and proteins associated with translation are targets for many antibiotics, which interfere with translation through different molecular mechanisms of action [12]. As far as we know, this is the first proteomic study that has identified such a large number of ribosomes and ribosome-associated proteins in the A. baumannii proteome, and we believe that the data provided here can serve as preliminary data for further studies on this bacterium, including studies of antibiotic drug action.

Proteins with potential role in drug resistance, stress response and in virulence
A recent description of the A. baumannii ATCC 17978 genome revealed that this strain has acquired several genes from its environment, these foreign genes are clustered in small (>10 kb) regions called putative alien islands (pAs) [9]. In addition, it was reported that ATCC 17978 genome have a total of 28 pAs and some of these are associated with drug resistance, virulence, iron uptake, metabolism and others [9]. Of the identified proteins, only 26 out of 192 were encoded by genes mapped in these pAs (see table S1, Additional file 1 and table S2, Additional file 2). This led us to conclude that most of the expressed proteins correspond to genes located in the native A. baumannii chromosome. Hence, it is tempting to speculate that those pAs are mainly expressed under specific conditions, e.g. stress.
Although for the most cases it was not clear the association between the protein and its respective pAs function, there were a few examples where it was possible to establish a more direct association. For instance, heat-shock proteins (Hsps) namely chaperone Hsp60 (spot # 35), and Co-chaperonin GroES (spot # 347), are encoded by genes situated in drug resistance pA 23 [9]. Many of the Hsps are chaperones and are important for protection against environmental stress and provide tolerance to high temperature [43,44], nutrient deprivation [45], salinity and osmotic stress [46]. Moreover, taking into account that Hsp60 chaperones were by far one of the most abundant protein spot (spot # 35 1.13 ± 0.8); and considering the presence of other Hsps (spot # 1, 2 and 393), it appears reasonable to suggest that these Hsps play a relevant role in the responses of A. baumannii to environmental changes and in the survival and adaptation to stress conditions. At this respect, a recent report have shown downregulation of Hsp60 proteins associated with resistance to colistin concomitantly to a decrease in bacterial fitness in an A. baumannii isolate [47]. In addition to Hsps, the Universal stress proteins (Usps) here identified (spot # 332, 333 and 498) are encoded by genes located in pA 18, which is associated with drug resistance [9]. In E. coli the expression level of these small cytoplasmatic proteins increase when cells are exposed to stress including CdCl 2 , H 2 O 2 , DNP, CCCP and osmotic stress [48], moreover Usp enhances the rate of cell survival during prolonged exposure to stress conditions [48]. In A. baumannii, the function of Usps remain elusive, recently we have observed that the expression of these proteins increase at late phases of in vitro growth (data not published), therefore Usps can be an attractive target for further studies concerning drug resistance.
Within the stress response context it is worthy to refer the presence of proteins known to play a relevant role in response to oxidative stress such as alkyl hydroperoxide reductase (spot # 271 and 273), superoxide dismutase (spot # 247), and glutathione peroxidase (478).
Here we have identified several proteins encoded by genes situated in pAs with a predicted role in virulence, however again it is not straightforward the association of those proteins with virulence. Nevertheless, amongst the proteins identified in this report, a few may be more directly related to virulence.  S1, Additional file 1) associated to the OMVs, we find these proteins exclusively in the cytoplasmatic fraction. In the referred study, the authors do not rule out the possibility of the entrapment of these proteins by unknown mechanism. If this is the case, then the data gathered here, together with that presented in the referred study provide a list of potential candidate cytoplasmatic proteins that can be transported in OMVs.
Furthermore, in the current study a putative type III effector (spot # 353) was identified, and the interactions between Pseudomonas synringae-host interaction is mediated in great part by effector proteins, which are injected into plant cells by type III secretion [50]. In A. baumannii the function of type III effector has yet to be elucidated, although it probably participates in bacterium-host interactions.
The specific virulence factors or pathogenic mechanisms of this bacterium remain elusive [51]. In light of the present results, it would be interesting to use a proteomic approach to investigate the expression of these and other proteins with a potential role in virulence, in the presence of virulence-induced stimuli or in the process of host-bacteria interactions (experiments are currently in progress).

Conclusion
Here we report for the first time a 2-DE based analysis, of both membrane and cytoplasmatic fraction of the opportunistic pathogen A. baumannii. Analysis of the most abundant proteins of this microorganism indicates that A. baumannii has a versatile and robust metabolism capable of utilising a wide range of nutrient sources. In addition, this study revealed the presence of highly expressed proteins such as porins (membrane fraction) and chaperones or heat-shock proteins and others, which are likely to play a crucial role not only in mechanisms of virulence and drug resistance, but also in adaptive environmental responses, which are probably related to the persistence in hospital settings, one of the hallmark of this microorganism. Thus, this proteomic approach provides us with new molecular tools to investigate the complexity of mechanisms operating during virulence and adaptive responses of this pathogen, which has been recently outlined by the Infectious Diseases Society of America as one of six important highly dangerous drug resistant microbes in hospitals worldwide.