Introduction

Brucella is an intracellular bacterial pathogen that infects ruminants and other mammals as primary host producing abortions and reduction of fertility1. Brucella can also be transmitted to humans as a secondary host producing a worldwide disease named brucellosis, which is characterized by undulant fever and general malaise. Among all Brucella species, B. abortus, B. melitensis and B. suis have the most zoonotic prevalence2. The ability to survive within macrophages, a phagocytic host cell type which is the first line of defense against microbes, is critical for the persistence of Brucella within the host3. Due to its longstanding association with mammalian cells, Brucella has evolved molecular mechanisms to hijack host cell machinery and subvert programmed cell death, thereby allowing Brucella to survive within host cells4, 5. After macrophage internalization Brucella becomes enclosed in membrane bound compartment named as Brucella-containing vacuole (BCV). Initially, BCV interacts with early endosome vacuoles and later avoids the fusion to the highly degradative lysosome in order to survive. After 10–12 hours post-infection, those BCVs that have successfully escaped from entering the endocytic pathway fuse with endoplasmic reticulum-derived membranes becoming a permissive compartment for Brucella replication. At 24–48 hours post-infection, Brucella replicates actively within the cistern of an endoplasmic reticulum-like network that completely occupies the host cell cytosol6.

It has been reported that the activity of the Brucella type four protein secretion system (T4SS)7 and the production of periplasmic cyclic β1–2 glucans (CβG)8, 9 are important traits that allow the bacterium to successfully acquire its intracellular replicative niche within macrophages. T4SS is a “nanomachine” induced within the host cell10 that allows Brucella to translocate a battery of effector proteins to the host cell cytosol in order to modulate its normal physiology11,12,13,14,15,16. CβGs are critical for allowing BCVs to escape from the endocytic pathway, likely due to its ability to occlude cholesterol within its inner ring extracting it from membranes disrupting lipid rafts17. CβGs also have been involved in modulation of host inflammatory and immune responses18, 19.

To gain insights into the molecular mechanisms involved in intracellular adaptation and virulence of Brucella, a comparative proteomic investigation of Brucella grown in culture media or recovered from macrophages at 48 hours post-infection was performed using iTRAQ isobaric tags. This approach allowed a relative and absolute quantitation of modulated proteins from in vitro- and intracellular Brucella cells.

Our results indicate that Brucella undergoes an extensive rearrangement of its proteome in order to adapt to the intracellular microenvironment within macrophages.

Results and Discussion

The highly sensitive iTRAQ LC–MS/MS method allows the identification of Brucella proteins differentially expressed within macrophages

In this study, a global proteome analysis was conducted to directly investigate the differential proteomic profiles of B. abortus isolated within the host macrophage cell in comparison with in vitro-cultured Brucella. Here, the Brucella proteome was systematically analyzed using iTRAQ combined with liquid chromatography and MALDI TOF-TOF tandem mass spectrometry. This approach allows for the direct determination of relative protein abundance levels between comparative conditions and offers several advantages over traditional proteomic methodologies20, 21.

We selected a post-infection time of 48 hours when Brucella has already reached its replication niche in the ER-like compartment that occupied the whole host cell cytosol. This late post-infection time is particularly interesting because it allows us to understand the molecular mechanism that Brucella deploys in order to acquire its intracellular replicative niche within mammalian macrophage, a cell type which is responsible for engulfing and destroying bacterial pathogens.

A total of five independent iTRAQ experiments, two with intracellular Brucella (iTRAQ labels 117&119) and three with in vitro-cultured Brucella (iTRAQ labels 115, 116 & 118) were performed (Fig. 1), and a total number of 998 Brucella proteins were identified. Initially, from the five independent experiments, six ratio intracellular/in vitro-cultured control (I/C) were generated. To be considered as a differential expressed protein we selected the following criteria, (i) proteins with (I/C) ratios above 1.1 and below 0.9, (ii) protein ratios (I/C) with P values < 0.05, (iii) differentially expressed in at least two independent ratios (I/C), iv) proteins that displayed discrepancy in ratio direction (up or down regulated) were not considered. From this initial analysis, 207 proteins were identified as intracellular modulated proteins. To add a biological dimension to the initial cutoff value, proteins with I/C ratios values closer to the initial cutoff were selected and a Western blot analysis was performed to confirm their status as intracellular modulated protein. Using antibodies against ribosome recycling protein (Frr) with I/C ratios of 0.83 and 0.75 (Fig. 2A and Supplementary Table S1) or against choloylglycine hydrolase (Cgh) with I/C ratios of 1.18, 1.27 and 1.33 (Fig. 2A and Supplementary Table S1), a new cutoff value was selected in order to include as modulated proteins all the I/C ratios present in Frr and Cgh (I/C ratio below 0.83 or above 1.18). Considering this new cutoff, in addition of the rest of criteria mentioned before, 197 proteins were characterized as intracellular modulated protein, with 84 proteins being up-regulated and 113 down-regulated (Fig. 3A and Supplementary Table S1). Representative MS/MS spectra of peptides from differentially expressed proteins along with the intensity of their reporter ions (iTRAQ labels) are shown in Supplementary Fig. S1. Modulated proteins were assigned to functional categories using the cluster of orthologous groups (COGs) analysis. As shown in Fig. 3B, most of the proteins were grouped into those involved in nutrient transport (amino acids, carbohydrates & inorganic ions), metabolism (carbohydrates, amino acids), protein homeostasis (proteases, chaperones, ribosomal proteins, etc.), and ATP production (Tables 1, 2, 3, 4 and 5 and S2, S3 and S4).

Figure 1
figure 1

Schematic representation of the workflow of the iTRAQ experiment.

Full size image
Figure 2
figure 2

Western blot validation of up and down-regulated protein identified by iTRAQ proteomic. (A) Quantification of the intensity of bands (below) of the Western blots analysis (top) of the whole cell lysates of intracellular Brucella (I) or Brucella growth in culture media (control) (C). Blots were probed with mouse sera specific against; (a) universal stress protein (UspA), (b) ferric uptake regulator (Irr), (c) choloylglycine hydrolase (Cgh) (d) proline racemase (PrpA) and (e) ribosome recycling factor (Frr). Monoclonal antibody against anti-Omp10 was used as a normalizing control. Blots are representative of two independent experiments. Displayed Western blots bands were cropped from the original blot for clarity. Corresponding full-length blots are shown in Supplementary Figure S3. Unpaired two-tailed Student’s t-test (P < 0.05) was performed. (B) The table shows the comparison of the results of the mean ratio (I/C) obtained for iTRAQ or Western blot analysis.

Full size image
Figure 3
figure 3

Global analysis of differentially expressed proteins. (A) Mean of log2 I/C ratio distribution of intracellular modulated proteins. The red and blue dots represent up-regulated and down-regulated proteins, respectively. Shade gray area represent proteins involved in iron transport and metabolism. Dash lines indicate the upper and lower log2 ratio cutoff (0.24, −0.26) used to consider proteins as being modulated. Numbers located at the right, indicate the total number of up, down and non-modulated proteins identified in this study. (B) Distribution of the identified modulated proteins accordingly to their assigned functional category (COGs).

Full size image
Table 1 Inorganic ion transport and metabolism (P)a.
Full size table
Table 2 Energy production and conversion (C)a.
Full size table
Table 3 Carbohydrate transport and metabolism (G)a.
Full size table
Table 4 Translation, ribosomal structure and biogenesis (J)a.
Full size table
Table 5 Posttranslational modification, protein turnover, chaperones (O)a.
Full size table

The expression profile of proteins modulated within macrophages suggests that BCV is an iron-limiting compartment

As shown in Table 1, the comparison of the protein expression profile of Brucella residing within the host cell versus proteins expressed by the bacteria grown in culture medium showed a differential expression of several proteins involved in iron transport and metabolism. Remarkably, the master regulator Irr (Iron-responsive regulator protein) (BAB1_2175), which is responsible for sensing variations in iron concentration, was highly up-regulated in the BCV. Brucella Irr has been reported to be stable in iron-limiting growth conditions and degraded when iron is available22. As a consequence of its stabilization, Irr promotes the induction of iron acquisition systems. Accordingly, isochorismate synthase (BAB2_0015), one of the enzymes required for Brucella siderophore biosynthesis, was detected up-regulated (Table 1). Siderophores are small iron-chelators molecules that the bacteria excrete to the extracellular medium to capture iron from the environment. The induction of the operon responsible for siderophore synthesis was also observed in an intracellular Brucella transcriptome analysis23. Iron-loaded siderophores are transported back to the bacteria mediated by specific ABC transporters. The uptake of ferri-siderophores by the bacteria is challenging since it requires passage through the bacterial outer and inner membranes, a process that proceeds against an electrochemical gradient in the absence of any ATP production24. Thus, bacteria have evolved proteins like the periplasmic TonB that energize the process of transporting across the membrane by transducing a proton motive force generated in the inner membrane. Similar to the previously reported transcriptomic study23, TonB (BAB1_1681) was the protein most over-expressed in intracellular Brucella (Table 1 and Fig. 3A, see shaded area). No changes were observed in the expression of its inner membrane partners ExbB (BABI_1679) and ExbD (BABI_1680). Ferric uptake through the outer membrane can be mediated by siderophores or by hijacking host ferric-binding proteins such as transferrin or lactoferrin mediated by receptors. Once the iron cargo is incorporated either by siderophores or by siderophore-independent pathways, a periplasmic transferrin-like protein named FbpA, captures free iron for further delivery to the cytosol mediated by ferric-specific ABC transporters25. We observed that Brucella FbpA (BAB2_0539) was up-regulated within the host cell. In addition, we also detected the intracellular up-regulation of an enterochelin-like ABC transporter (BAB2_0564), a periplasmic component of a Fe3+ transport system (BAB2_0113), and a Fe2+ specific ABC transporter (BAB2_0840). Moreover, Brucella iron storage proteins, bacterioferritin (BAB2_0675) and Dps (BAB1_2150) were down-regulated. Interestingly, in Bradyrhizobium, the gene transcription of bacterioferritin is repressed when the bacterium grows in iron restricted condition in an Irr-dependent manner26. A recent report showed that Salmonella downregulated Dps in response to oxidative stress27. The iron related proteins described above had the highest level of up-regulation in this proteomic analysis (Fig. 3A, see shaded area). All these results showed that Brucella is subjected to iron deprivation within the BCV microenvironment.

At this point it was interesting to evaluate the modulation of Brucella iron-regulated proteins also at transcriptional level. A set of these proteins were selected to study their modulation by using RT-qPCR. As shown in Supplementary Fig. S2 B. abortus recovered from macrophages induced the transcription of irr, entC genes, and reduced the expression of dps gene. Therefore, modulation in the expression level of Irr, EntC and Dps proteins can be explain as a consequence of the induction or repression of their respective genes.In a previous report Martinez et al.22 also observed the up-modulation of Irr when B. abortus was grown in an iron-depleted culture medium. They further concluded that up-regulation of Irr was not due to the induction of irr transcription (using a transcriptional fusion approach) but rather a consequence of a reduction of Irr degradation. Here, in addition of detecting the up-modulation of Irr protein by iTRAQ, we also observed the induction of irr gene transcription in intracellular Brucella by RT-qPCR as shown above. Apparent discrepancies between both observations at transcriptional level it might be partially explained by differences in sensitivity of the used techniques in both cases. However, it would be interesting to speculate that transcriptional and translational up-regulation that was detected here it might be the result of a long-term iron deprivation found by Brucella when located within the host cells for 48 hours, compared with the in-vitro iron deprivation condition. Interestingly, in an intracellular Brucella melitensis and Brucella canis transcriptomic report23, it was found that the transcription of irr gene was reduced at 24 hours post-infection. If differences observed by Eskra et al.23 compared with our results are due to a differential regulation of irr transcription in different species of Brucella or it is simply a consequence of an alternative experimental setting (24 hour post-infection and RAW cells) it remains to be explored further.

Iron scarcity promotes a reduction in the activity of the TCA cycle and oxidative phosphorylation

It has been reported that iron deprivation can lead to a perturbation of energy producing metabolic pathways. This is due to the fact that enzymes of these pathways contain iron-sulfur (Fe-S) proteins or heme-containing enzymes which require iron for their activity28. Accordingly, we found in this study, that Brucella aconitase (BAB1_0090), one of the critical enzymes from the TCA cycle, was severely down-regulated within the host cell (Table 2 and Fig. 4). Similar findings were reported in the bacterium Staphylococcus or in the fungus Paracoccidioides brasiliensis when iron source is limited29, 30. Bacterial aconitases are iron-sulfur proteins that have been described as moonlighting or multitasking proteins that in addition to their metabolic role in TCA and glyoxylate cycles, also have a role as mRNA-binding proteins that modulates post-transcriptionally the expression of proteins involved in iron metabolism31.

Figure 4
figure 4

Protein expression changes of metabolic pathways involved in (A) central carbon metabolism or (B) oxidative phosphorylation in B. abortus during its intracellular life. Locus Tags depicted either in red or blue indicate up or down-regulated proteins respectively.

Full size image

ATP and NADH are produced during glycolysis and in the TCA cycle. In addition to its role in redox reaction enzymes (especially in catabolism), NADH is also the first electron donor of the oxidative phosphorylation chain, at Complex I (NADH:ubiquinone oxidoreductase) or Nuo complex. The Nuo complex plays a central role in the generation of ATP, since it couples electron transference between NADH and ubiquinone to proton translocation32, 33. We found that all the subunits of the dehydrogenase domain (NuoE, NuoF and NuoG) (BAB1_0826, BAB1_0827, and BAB1_0828, respectively) and one subunit of the connecting domain (NuoD, BAB1_0825) of Nuo Complex I were down-regulated (Table 2 and Fig. 4). Interestingly NuoE, NuoF and NuoG possess iron sulfur clusters, while the connecting domain protein NuoD has a Ni-Fe cluster. All these results also suggest that down-regulation of Complex I proteins is related to iron deficiency.

In addition to the down-regulation of the Nuo complex, components of Complex III (PetA, BAB1_1559) and Complex IV (CcoP, BAB1_0389), an iron sulfur protein and a heme-containing protein, respectively, were also down regulated. It has been reported in bacteria and fungus that the oxidative phosphorylation chain is interrupted when microorganisms are grown in an iron-free medium, with a consequent reduction in ATP production29, 30. The last component of the oxidative phosphorylation chain is the F0F1 complex that utilizes the proton gradient associated to the electron transfer to phosphorylate ADP. As protons in the periplasm are transported back to the cytosol, conformational changes in F0F1 ATP synthase promote ATP production. When oxidative phosphorylation is interrupted, F0F1 still has the ability to reverse the reaction of H+ extrusion from cytosol by consuming ATP, a process which is critical for adaptation of bacteria to acidic environment34. As shown in Table 2, we detected that the alpha (BAB1_1809) and gamma (BAB1_1808) subunits of the F0F1 complex were up-regulated in intracellular Brucella. It is conceivable to speculate that up-regulation of F0F1 might reflect a requirement of ATP hydrolysis to generate a proton gradient across the inner membrane and to allow Brucella to maintain a physiological intracellular pH.

It was reported that microorganisms reinforce the use of metabolic pathways such as glycolysis when they face limitations in energy production (e.g., as a consequence of reduction in the oxidative phosphorylation) since they can synthesize ATP by substrate-level phosphorylation29, 30. We have also identified an up-regulation of four enzymes of glycolysis in intracellular Brucella (Fig. 4 and Table 3): fructose-1,6-bisphosphate aldolase (BAB2_0365), phosphoglycerate kinase (BAB1_1742), enolase (BAB1_1155), and pyruvate kinase (BAB1_1761). Both phosphoglycerate kinase and pyruvate kinase contribute to ATP formation by substrate-level phosphorylation, thereby preventing an energy shortage if the synthesis of ATP through oxidative phosphorylation is impaired.

As mentioned above, ATP and NADH are the main products of glycolysis. Since aerobic glucose oxidation is not proceeding at the level of the respiratory chain, ATP demand lead to an increased metabolism due to incremented glycolysis and consequently an accumulation of NADH. Along with NADH accumulation, NAD+ is not regenerated which is a critical cofactor required for an effective continuation of glycolysis. Interestingly, BAB1_0128, a zinc-containing alcohol dehydrogenase, an enzyme that consumes NADH to produce ethanol and regenerate NAD+ was detected up-regulated in intracellular Brucella (Table 3 and Fig. 4). The increased activity of this alcohol dehydrogenase could contribute to maintain the redox balance by recycling NADH. Methylglyoxal is another product derived from glycolysis which is produced from dihydroxyacetone phosphate by the action of methylglyoxal synthase. Accumulation of metylglyoxal is very toxic for the cell and there are glyoxalase enzymes (glyoxalases 1 and 2) that metabolize this compound to lactate35. In intracellular Brucella, methylglyoxal synthase (BAB2_1009) (Table 3 and Fig. 4) is down-regulated while glyoxalase (BAB1_1286) (Supplementary Table S2 and Fig. 4) is up-regulated, suggesting that intracellular Brucella is controlling the increase in concentration of this compound to avoid toxicity.

Pyruvate dehydrogenase complex (PDH) plays a key role in carbon metabolism. It transform pyruvate derived from glycolysis into acetyl-coenzyme A that can either enter the TCA cycle or be available to enter a number of alternative pathways such as fatty acids or amino acids biosynthesis. As shown in Table 2 and Fig. 4, all the subunits of PDH (BAB1_1149, BAB_1150, BAB1_1151 and BAB1_1152) were up-regulated by Brucella within the host cell. This has also been reported in the closely-related α2-proteobacteria Sinorhyzobium meliloti in which the PDH complex is also up-regulated when the bacterium is located intracellularly within the plant cell36.

Living within the macrophage is a stressful condition for Brucella abortus

At 48 hours post-infection, a global down-regulation of the expression of Brucella ribosomal protein (r-proteins) was observed suggesting a reduction of protein synthesis (Table 4). Reduction of the r-proteins content has been linked to the stringent response (SR), the adaptive response to different types of starvation stresses37. A recent transcriptomic study of intracellular Brucella also found a critical down-regulation in r-protein transcription38. A set of Brucella proteins characterized as stress adaptation proteins were also modulated within the macrophage. This included the universal stress protein UspA (BAB1_1068) (Supplementary Table S4) which was reported to respond to different types of stressors (temperature, oxidative and osmotic) in the human pathogen Escherichia coli O157:H739.

Another critical challenge that intracellular bacteria have to face is the oxidative stress response within professional macrophages. As shown in Table 5, both components of the Brucella alkyl hydroperoxide reductase complex AhpC (BAB2_0531) and AhpD (BAB2_0532), which are involved in detoxification of H2O2 produced by oxidative burst in macrophages were up-regulated. In addition, an enzyme that quenches superoxide anion radicals (O2 −) such as superoxide dismutase C (BAB2_0535) was also up-regulated (Table 1). In contrast, the Fe-S containing enzyme catalase (BAB2_0848) (Table 5) was down-regulated, probably also as a consequence of iron deprivation as discussed above. Oxidative stress promotes bacterial protein denaturation and the formation of protein aggregates which are toxic for bacterium. To restore protein homeostasis, a set of bacterial proteins that assist the process of protein refolding (chaperones) or degradation of protein aggregates (proteases) were over-expressed. As shown in Table 5, three major proteases HtrA (BAB1_1413), CtpA (BAB1_1845) and FtsH (BAB1_1703) and a putative protease M20/M25/M40 (BAB1_0118) were all up-regulated in intracellular Brucella. Depending on their activity, molecular chaperones have been classified in i) “folding chaperones” (e.g., DnaK and GroEL), ii) “holding chaperones” (e.g., Hsp33, Hsp31 and IbpB) that maintain unfolded proteins in solution awaiting availability of “folding” chaperones, and iii) chaperones that solubilize protein aggregates produced by stress (e.g., ClpB). Although DnaK, a critical bacterial chaperone, was not modulated, its cochaperones GrpE (BAB1_0170) and DnaJ (BAB1_0446) were up-regulated and down-regulated, respectively (Table 5). Interaction of DnaK chaperone with its cochaperones GrpE and DnaJ determines the avidity of DnaK for misfolded proteins. When the nucleotide exchange factor GrpE replace ADP by ATP on DnaK, this chaperone is locked in an “open” conformation, with low affinity for peptides. This conformation favors the release of peptides allowing DnaK to start a new cycle of folding. It has been described that over-expression of GrpE tends to inhibit DnaK function40, 41. Whether the combination of GrpE up-regulation and DnaJ down-regulation reflects a reduced DnaK activity remains to be determined.

Another chaperone, Hsp33 (BAB1_0333), was found to be up-regulated. It has been described that Hsp33 is inactive when its four cysteins located in the active site are in a reduced state. As a consequence of oxidative stress, the cysteins become oxidized leading to the activation of Hsp33 that form dimers or oligomers that bind to protein targets42. As shown in Table 5, the chaperone ClpA (BAB1_1138) was also up-regulated. This protein is critical in Brucella suis to survive against oxidative stress43. We also observed the up-regulation of a peptidyl-prolyl-cis-trans isomerase, PpiC (BAB1_1944), which is involved in protein folding. In a recent report we showed that cyclophilins CypA and CypB (one type of peptidyl-prolyl-cis-trans isomerase enzymes) were also up-regulated intracellularly playing a critical role in adaptation of Brucella within host cells44. The up-regulation of proteins dedicated to reduce the concentration of reactive oxygen species, protein chaperones that prevent misfolding of proteins, and proteases that degrade toxic protein aggregates, suggest that Brucella is adapting to a stressful condition within macrophages, protecting itself from damage likely due to oxidative burst.

Validation of proteomic results

To confirm iTRAQ results by an alternative method such as Western blot analysis we selected five proteins that were modulated intracellularly (three up and two down-regulated). As a reference internal control (RIC), the protein Omp10 was used because (i) it was not modulated in our proteomic study, (ii) it was not reported to be affected in different Brucella proteomic studies and (iii) it is an immunogenic protein that allow us to use it as a reference with high sensitivity. The amount of modulated protein in comparison with RIC in both intracellular and in vitro conditions is shown in Fig. 2A. The results were in accordance with those observed in the iTRAQ analysis (Fig. 2B) showing that intracellular Brucella (relative to the in vitro control) contains increased amounts of universal stress protein (UspA, BAB1_1068), ferric-uptake regulator (Irr, BAB1_2175) or choloylglycine hydrolase (Cgh, BAB1_1488) and decreased amounts of prolil racemase (PrpA, BAB1_1800) or ribosome recycling factor (Frr, BAB1_1181).

In addition to the required validation of proteomic results from a statistical point of view or by Western blot analysis, it is interesting to analyze our results regarding metabolic or functional criteria. For instance, here we observed that all the subunits of different protein complexes (or most of them) show a coordinated modulation. Thus, the PDH complex (Table 2), the F0F1 ATP synthase complex (Table 2), the 2-oxoglutarate dehydrogenase complex (Table 2) and the electron transfer flavoprotein complex (Table 2) displayed a coordinated up-regulation. In addition, several ribosomal proteins (Table 4) and the Nuo complex (Table 2) were down-regulated. An interesting observation was that proteins of two different systems dedicated to the same function, the Fe-S cluster assembly, SufB (Table 5) and Nif U (BAB1_ Table 5), were both up-regulated.

Iron concentration is an important signal that Brucella sense for intracellular adaptation

As shown in Supplementary Table S5, a more detailed analysis determines that about the 28% of modulated proteins were somehow related to iron metabolism or regulated by iron concentration. In addition, our results highlighted the critical role of the master regulator Irr also in intracellular adaptation of Brucella. Accordingly, Irr was up-regulated at transcriptional and translational level in intracellular Brucella (Table 1 and Figs 2 and S2). It has been described that Irr is able to recognize and bind a DNA sequence motif named ICE-Box (also known as Irr binding motif)26, 45, 46. Irr can act either as an activator or repressor factor depending on the proximity of ICE consensus motif to the promoter of the Irr-regulated genes26. A search for ICE consensus motif was performed along the intergenic regions of intracellularly modulated proteins using the Find Individual Motif Ocurrences (FIMO) algorithm software (Supplementary Table S5)47. Of interest, from the 197 modulated proteins, 55 of them (28%) were proteins related to iron, from which 35 have also a predicted ICE motif, a strong suggestion that Irr is playing a central role in intracellular adaptation (Supplementary Table S5).

Proteomic studies showed common features in microbial pathogen-host interactions

There are common features in the intracellular protein expression profile of different intracellular pathogens such as Shigella, Salmonella or Brucella. Intracellular Salmonella showed an up-regulation of proteins involved in iron transport and metabolism48 while intracellular Shigella modifies its central carbon metabolism with reduction of oxidative phosphorylation49, as described here for Brucella.

Results depicted in Fig. 5, showed the comparison of our results with a previous intracellular proteomic report in Brucella abortus 2308 at 44 hours post-infection50. Intersection in the Venn diagram representation shows common proteins that were differentially expressed intracellularly (23 proteins) in both studies (Fig. 5). Only one protein showed discrepancy in the direction of the modulation (up or down regulation) between both reports.

Figure 5
figure 5

Comparison of the results of previous intracellular proteomic report in B. abortus 2308 with our study. Intersection of the Venn diagram shows common differentially expressed proteins identified in both studies.

Full size image

In this report we have identified a set of 174 new differentially modulated proteins in intracellular Brucella. We believe these results are providing new insights into the molecular mechanisms that the bacterium utilizes in order to adapt to its intracellular replication niche within macrophages.

Concluding Remarks

Of the total number of proteins that B. abortus modulates within macrophages, 28% of them were either related with iron transport, iron storage, iron as cofactor, the iron master regulator Irr and proteins which expression is controlled by Irr. As reported in an earlier publication51 Brucella grown in iron-restricted conditions had a greater invasiveness and virulence. This observation is particularly interesting because allow us to speculate if iron scarcity in addition of being a nutritional stressor it might also function as a “positioning” signal that tells Brucella about its intracellular localization within host cells. Results presented here are consistent with the idea that iron deprivation is an important cue that conduit B. abortus to shape its proteome in order to increase its intracellular fitness. Here, by using the highly sensitive iTRAQ/MS/MS technique, we provided new insights into the Brucella virulence mechanisms that will redounds in the design of new strategies to control this worldwide zoonosis.

Materials and Methods

Growth of B. abortus

All experiments involving live Brucella abortus 2308 (Brucella melitensis biovar Abortus 2308) were conducted under Biosafety level 3 (BSL3) conditions. B. abortus strains were grown in Brucella broth (BB) (Difco Laboratories, Detroit, Mich.) at 37 °C on a rotary shaker (250 rpm). Culture media were supplemented with nalidixic acid (Nal) at 5 μg/ml.

Cell culture and infection assay

Cell culture and infection assay of J774 macrophages for proteomic studies was performed as described previously44. Aliquots of the resulting lysates were serially diluted in PBS and plated on BB agar to determine CFUs.

Isolation of Brucella from infected J774 macrophages

Isolation of Brucella from infected macrophages was performed as described previously44. Infected cell lysates were centrifuged at 210 × g for 10 min at 4 °C to remove host cell debris. The culture supernatant was collected and centrifuged at 20,000 × g for 30 min at 4 °C, and the resulting pellet was resuspended in 3 ml of Tris-sucrose (TS) buffer (33 mM Tris-hydrochloride containing 0.25 M sucrose [pH 7.4]). Three milliliters of the bacterial suspension were loaded onto 27 ml of Percoll (GE Healthcare Life Sciences) prepared at 30% (vol/vol) in polycarbonate centrifuge tubes. Tubes were centrifuged at 25,000 × g for 60 min at 4 °C to allow the development of a self-forming-gradient by isopycnic centrifugation leading to the development of two gradient bands. The lower band of the gradient, containing more than 85% of Brucella cells, was collected and diluted 10-fold in ice-cold PBS (pH 7.4) and then centrifuged at 20,000 × g for 30 min at 4 °C. The pellet was resuspended in PBS and recentrifuged to eliminate residual Percoll. The final pellet from each gradient was resuspended in PBS, and protein content and Brucella viability were determined. In addition, bacteria in vitro cultured were subjected to the same purification steps.

Protein extraction

Protein extraction was performed as described previously44. Protein extraction from Brucella isolated from macrophages or from in vitro-cultured bacteria was performed in the same manner. Aliquots of 75 ul of Brucella cells suspension were mixed with 37.5 ul of acetonitrile (ACN) (to a final concentration of 7.5% (vol/vol)) and 387.5 ul of 8 M urea in a total volume of 500 μl. The cells were sonicated using 5 pulses of 5-s duration each with a 30-s rest between each pulse. After sonication, the cells were centrifuged (15,000 x g, 10 min, 4 °C) and the supernatant was removed and kept. The supernatant containing 0.8 mg/ml of protein was then applied to a Pall 10 K Nanosep column and concentrated to approximately 75 μl. A series of buffer exchange and protein cleaning steps were performed as follows, with reconcentration to 75 μl after each step: step 1, addition of 4% CHAPS {3-[(3-cholamidopropyl)-dimethylammonio]-1-propanesulfonate} buffer (500 μl); step 2, addition of 7.5% ACN plus 4% CHAPS (500 μl); step 3, addition of 4% CHAPS (500 μl) followed by a repeat of step 3; step 4, addition of 0.05% CHAPS (500 μl) followed by a repeat of step 4. After each extraction, the total protein concentration was determined by Bradford analysis.

Trypsin Digest and iTRAQ Labeling

For trypsin digest 100 µg of protein from either intracellular or in vitro grown Brucella were aliquoted and dried down in a centrivap. The protein pellet was then resuspended in 20 µl of Dissolution Buffer (0.5 M triethylammonium bicarbonate (TEAB) (diluted to 0.5 M with dH2O, pH 5.5). Denaturant (1 µl of 2% (wt/vol) SDS) was then added and the sample was vortexed. Then the reducing reagent tris-(2-carboxyethyl) phosphine (TCEP) (1 µl, final concentration 5 mM) was added, the sample was vortexed and centrifuged briefly, and then incubated at 60 °C for 1 hour. Then, fresh iodoacetamide (1 µl of an 84 mM stock) was added and the sample was wrapped in aluminum foil and incubated at room temperature for 30 minutes. Sequencing grade trypsin (Promega) was then reconstituted with Resuspension buffer (50 mM acetic acid), and 2 µl of the resulting suspension was added to the protein sample and the entire reaction was incubated overnight at 48 °C. Samples were then dried in a vacuum concentrator to a total volume of no more than 33 µl.

For this analysis, the 8-plex iTRAQ kit was utilized (Applied Biosystems). Trypsin digests from two intracellular and three in vitro Brucella samples were labeled with iTRAQ according to the manufacture’s protocol. Briefly, all iTRAQ reagents were brought to room temperature and then 50 µl of isopropanol was added to each. The entire contents of each resuspended iTRAQ label was then added to a sample vial and incubated at room temperature for 2 hours. Then, 100 µl of dH2O was added to quench each reaction and incubated for an additional 30 minutes at room temperature. All iTRAQ labeled samples (5 samples including labels 115, 116, 117, 118 and 119) were then combined into a single tube and dried down in a speed vac. Chasing of the sample was then performed by three rounds of resuspension in 100 µl dH2O and drying in a speed vac. The final dried sample was then submitted to The Pennsylvania State University Hershey Medical Center Mass Spectrometry Core Facility (Hershey, PA) for LC-MS/MS analysis.

2D-LC Separations

Strong cation exchange (SCX) separations were performed on a passivated Waters 600E HPLC system using a 4.6 × 250 mm PolySULFOETHYL Aspartmide column (PolyLC, Columbia, MD) at a flow rate of 1 ml/min. Buffer A contained 10 mM ammonium formate, pH 2.7, in 20% acetonitrile/80% water. Buffer B contained 666 mM ammonium formate, pH 2.7, in 20% acetonitrile/80% water. The gradient was Buffer A at 100% (0–22 minute following sample injection), 0–40% Buffer B (16–48 min), 40–100% Buffer B (48–49 min), then isocratic 100% Buffer B (49–59 min), then at 56 min switched back to 100% Buffer A to re-equilibrate for the next injection. The first 26 ml of eluant (containing all flow-through fractions) was combined into one fraction, then 14 additional 2-ml fractions were collected. All 15 of these SCX fractions were dried down completely to reduce volume and to remove the volatile ammonium formate salts, then resuspended in 9 μl of 2% (vol/vol) acetonitrile, 0.1% (vol/vol) trifluoroacetic acid and filtered prior to reverse phase C18 nanoflow-LC separation.

For the 2nd dimension separation by reverse phase nanoflow LC, each SCX fraction was auto injected onto a Chromolith CapRpd column (150 × 0.1 mm, Merck) using a 5-μl injector loop on a Tempo LC MALDI Spotting system (ABI-MDS/Sciex). Buffer C was 2% acetonitrile, 0.1% trifluoroacetic acid, and Buffer D was 98% acetonitrile, 0.1% trifluoroacetic acid. The elution gradient was 95% C/5% D (2 μl per minute flowrate from 0–3 min, then 2.5 μl per minute from 3–8.1 mon), 5% D to 38% D (8.1–40 min), 38% D to 80% D (41–44 min), 80% D to 5% D (44–49 min). Flow rate was 2.5 μl/min during the gradient, and an equal flow of MALDI matrix solution was added post-column (7 mg/ml recrystallized CHCA (α-cyano-hydrocinnamic acid), 2 mg/ml ammonium phosphate, 0.1% trifluoroacetic acid, 80% acetonitrile). The combined eluant was automatically spotted onto a stainless steel MALDI target plate every 6 seconds (0.6 μl per spot), for a total of 370 spots per original SCX fraction.

Mass Spectrometry Analysis and Protein Identification and Quantitation

After the sample spots were dried on the MALDI plate, 13 calibrant spots (ABI 4700 Mix) were added to each plate manually. MALDI target plates (15 per experiment) were analyzed in a data-dependent manner on an ABI 4800 MALDI TOF-TOF. As each plate was entered into the instrument, a plate calibration/MS Default calibration update was performed and then the MS/MS default calibration was updated. MS spectra were then acquired from each sample spot using the newly updated default calibration using 500 laser shots per spot with a laser intensity of 3200. A plate-wide interpretation was then automatically performed, choosing the highest peak of each observed m/z value for subsequent MS/MS analysis. Up to 2500 laser shots at a laser power of 4200 were accumulated for each MS/MS spectrum. When the MS and MS/MS spectra from all 15 plated in a sample set were acquired, protein identification and quantitation were performed using the Paragon algorithm as implemented in the Protein Pilot 3.0 software (version 2.01 prior to July 2009, from ABI/MDS-Sciex). The spectra were searched against the NCBInr database for Brucella abortus plus common contaminants as well as a reversed “decoy” version of itself.

For the ProteinPilot analysis, the preset Thourough (iTRAQ or Identification) search settings were used and identifications must have a ProteinPilot Unused Score greater than 1.3 (95% Confidence interval) in order to be accepted. In addition, the only protein IDs accepted were required to have a “Local False Discovery Rate” estimation of no higher than 5%, as calculated from the slope of the accumulated Decoy database hits by the PSPEP (Proteomics System Performance Evaluation Pipeline) program.

Western blot analysis for iTRAQ validation

Whole-cell lysates of the intracellular B. abortus and the in vitro growth control were subjected to 12% SDS-PAGE and transferred onto nitrocellulose membranes using a semidry transfer procedure. Immunoblotting was performed using as primary antibodies anti-UspA (1:500) (generated in this study), anti-Irr (1:4000) (kindly provided by M. Almirón), anti-PrpA (1:500) (kindly provided by JM Spera), anti-Frr (1:1000) (kindly provided by J. Cassataro), anti-Cgh (1:100) (kindly provided by MV Delpino) and anti-Omp10 (1:2000) (kindly provided by Axel Cloeckaert), and IRDye secondary anti-mouse or anti-rabbit antibodies (LI-COR, Inc.). All antibodies were diluted in TBS, 1% nonfat milk, 0.1% Tween solution. Detection was performed using the Odyssey imaging system (LI-COR, Inc.).

RNA isolation and RTqPCR

Total RNA was isolated from intracellular and in vitro Brucella with Trizol Reagent according to manufacturer’s instructions (Life Technologies) and RNA integrity was evaluated by 1% agarose gel electrophoresis. Samples were incubated with RQ1 DNAse (Promega). One microgram of RNA was reverse transcripted using a commercial kit (Superscript II RT, Invitrogen, CA, USA) with 200 ng of random primers. qRT-PCR reactions were carried out using PowerUpTm SYBR Green Master Mix (Applied Biosystems, Foster City, CA) on the Step one plus Real time PCR system (Applied Biosystems, Foster City, CA). LinRegPCR software was used to analyze raw data and determinate efficiency of PCR reactions52. The relative expression of genes was calculated using Relative expression software tool (REST©)53. Real time quantitative PCR was performed with primers described below. Irr mRNA abundance was measured with the oligonucleotides F 5′-CACTTGCGAGCCTGATTTTC-3′ and R 5′-AGACCACATTTTCCCCTTCG-3. EntC mRNA was measured with the oligonucleotides F 5′-AAGCTTGGTCTGTACTTCGG-3′ and R 5′-TCGAGCGTGGATTGTTTACC-3′. Dps mRNA was measured with the oligonucleotides F 5′-GTGAAAATATCTGCCGTGTCG-3′ and R 5′-GGTGAAGGAATCCAGACTGAAG-3′. The relative quantity of 16S rRNA using the oligonucleotides F 5′-GAGTATGGAAGAGGTGAGTGG-3′ and R 5′-CAGGCGGAATGTTTAATGCG-3′, was used as endogenous control to normalize all the values. Each assay was performed in biological triplicates.