Effects of ccpA gene deficiency in Lactobacillus delbrueckii subsp. bulgaricus under aerobic conditions as assessed by proteomic analysis

Background Aerobic growth provides benefits in biomass yield and stress tolerance of Lactobacillus delbrueckii subsp. bulgaricus (L. bulgaricus). Catabolite control protein A (CcpA) is a master regulator involved in the aerobic and anaerobic growth, metabolic production and stress response in L. bulgaricus, but its potential molecular mechanisms remains unclear. The aim of this study is to elucidate the role of CcpA in L. bulgaricus in aerobic growth at the proteomic perspective. Results The differential proteomic analysis was performed on the L. bulgaricus ATCC11842 and its ccpA inactivated mutant strain using iTRAQ technology. A total of 132 differentially expressed proteins were obtained, among which 58 were up-regulated and 74 were down-regulated. These proteins were mainly involved in the cellular stress response, carbohydrate and energy metabolism, amino acid transport and protein synthesis, genetic information processing. Moreover, inactivation of ccpA negatively affected the expression of key enzymes involved in glycolysis pathway, while it enhanced the expression of proteins related to the pyruvate pathway, supporting the conclusion that CcpA mediated the shift from homolactic fermentation to mixed acid fermentation in L. bulgaricus. Conclusions Overall, these results showed that the role of CcpA in L. bulgaricus as a pleiotropic regulator in aerobic metabolism and stress response. This proteomic analysis also provide new insights into the CcpA-mediated regulatory network of L. bulgaricus and potential strategies to improve the production of starter and probiotic cultures based on the metabolic engineering of global regulators.

The stress response in LAB is normally regulated by the induction of certain proteins such as heat shock proteins, cold shock proteins etc., and it is mainly mediated by the HrcA and CtsR repressors [7]. However, some common response mechanisms are triggered by different environmental stresses, thus suggesting a central role for global regulators. Among these, catabolite control protein A (CcpA) is one of the most important regulator since it is involved in the regulation of carbon catabolite repression (CCR) and various metabolic pathways in Gram-positive bacteria [8].
Some studies have demonstrated that the growth conditions under which aerobic (oxygen) and respiratory (aerobic growth in the presence of exogenous heme and/or menaquinone) growth improve the stress tolerance and biomass yield in several industrially important LAB [11,[16][17][18][19][20]. In L. lactis, inactivation of ccpA shifted homolactic fermentation to mixed acid fermentation under aeration conditions [14]. CcpA also plays a key role in respiration since it activated the repressor of the heme uptake preventing oxidative damage at the start of exponential growth of L. lactis [15]. In addition, CcpA positively regulated the expression of hrcA and groESL operons in L. plantarum [16], and aerobic growth improves stress tolerance. These results provide evidence for the role of CcpA in the regulation of aerobic metabolism, respiration and stress response. However, less is known about its role in L. bulgaricus. In addition, the scientific literature contains few studies concerning the role of the pleiotropic regulator CcpA at proteome level [21].
In previous work, we constructed a homologous ccpA deletion mutant strain of L. bulgaricus and performed a set of physiological and metabolic studies on the parental strain and the mutant strain [22]. Results showed that inactivation of ccpA significantly affected growth, metabolite production and stress tolerance.
To fully understand the central role of CcpA in modulating metabolism and stress L. bulgaricus, a thorough proteomic investigation on L. bulgaricus and its ccpA deletion mutant strain was reported in the present work. This integrative approach provides new insights into the cellular processes regulated by CcpA in L. bulgaricus.

Initial proteomic data analysis
In the present work, 1269 proteins were identified from analysis of three biological replicates. Figure 1a shows the molecular weight distribution of the identified proteins, Most had a molecular weight of 30-60 kDa, and these accounted for approximately 87% of all identified proteins. In addition, a few proteins with a molecular weight of greater than 100 kDa were identified, indicating that iTRAQ technology has a high detection sensitivity and can identify a wide range of proteins. Figure 1b shows the coverage of protein sequences. Most proteins had a peptide coverage of less than 60%, and 15-60% of proteins had a relatively uniform peptide coverage.

Identification of differentially expressed proteins
In this work, 132 differentially expressed proteins were significantly affected by ccpA inactivation in the mutant  strain compared to the parental strain, including 58 upregulated proteins and 74 down-regulated proteins.  Except for some unknown functional proteins, these differentially expressed proteins are listed in Table 1, and they were classified into the following categories: cellular response to stimulus, carbohydrate transport and metabolism, lipid transport and metabolism, amino acid transport and metabolism, genetic information processing, nucleotide metabolism, and other metabolism. As shown in Fig. 2, 13 differentially expressed proteins were associated with cellular response to stimulus, as well as carbohydrate transport and metabolism. 11 proteins were involved in amino acid transport and metabolism. 19 proteins were related to each of the terms genetic information processing and other metabolism. 10 proteins were involved in lipid transport and nucleotide metabolism. These data suggested that CcpA, as a pleiotropic regulator, was involved in multiple physiological and metabolic processes.
The results of the protein Gene Ontology (GO) analysis are shown in Fig. 3. Regarding cellular components, 43% were related to cell parts and 24% were related to macromolecular complexes; followed by organelles (18%) and cell membranes (13%) (Fig. 3a). Catalytic activity (43%) and binding (36%) were the dominant terms in the molecular function category. In addition, some proteins were related to transporter activity (8%) and structural molecule activity (10%) (Fig. 3b). In the biological process category, the differentially expressed proteins were related to metabolic process, cellular process, response to stimulus, immune system process, biological adhesion, and other such terms. The most enriched biological process terms were metabolic process and cellular process, which respectively accounted for 33.3% and 26.1% of the total biological process, followed by single-organism processes (18.2%) (Fig. 3c).
Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis of the differentially expressed proteins showed that 15 proteins were associated with the ribosome. 8 proteins were enriched in the biosynthesis of amino acids pathway, and 6 were annotated in the cysteine and methionine metabolism pathways, as well as the purine metabolism pathway. A total of 15 proteins were annotated in the carbon metabolism, glycolysis/ gluconeogenesis and pyruvate metabolism pathways. Additionally, lower numbers of proteins were related to pathways such as ABC transporters, aminoacyl-tRNA biosynthesis, and phosphotransferase system. (Additional file 1: Table S1).

Gene transcription analysis
To verify the proteome results, 10 key genes were evaluated by quantitative real-time PCR. The tested genes involved in carbohydrate metabolism, namely encoding phosphofructokinase (Pfk), pyruvate kinase (Pyk), D-lactic dehydrogenase (LdhA), phosphoglycerate kinase (Pgk), pyruvate oxidase (Pox) and acetate kinase (Ack), are shown in Fig. 4. The expression of pox1 was significantly up-regulated 2.87-fold, and ack was up-regulated 1.37-fold in the mutant strain compared with the parental strain (P < 0.05). In contrast, the gene expression levels of pfk, pgk, pyk and ldhA decreased by 3.57-, 1.75-, 4.55-and 2.38-fold, while the iTRAQ results showed that the protein expression levels of Pfk, Pgk, Pyk and LdhA decreased by 2.27-, 1.68-, 1.95-, and 2.28-fold, respectively (P < 0.05). In addition, genes involved in the stress response, namely encoding elongation factor Tu (EF-Tu), molecular chaperone Dnak (DnaK), chaperonin GroEL (GroEL), and heat-inducible transcription repressor HrcA (HrcA), were evaluated (Fig. 4). The expression of hrcA was up-regulated 2.17-fold, while that of Tuf, dnaK, and groL were decreased by 5.26-, 1.56-, and 2.13-fold, respectively. These results indicated that the protein expression levels were consistent with the corresponding mRNA expression levels, suggesting these proteins were regulated mainly at the transcriptional level in cells.

Discussion
Aerobic growth significantly affect the energy metabolism and stress tolerance of LAB with important consequence in food-related application [23]. Aerobic cultivation, moreover, support the activation of respiratory chain in LAB, resulting in phenotypes with improved growth and technological performance [23]. The global regulation of CcpA in aerobic metabolism and stress response has been studied in several important LAB in the dairy industry [11,15]. However, related literature in L. bulgaricus is scarce. Therefore, the role of the global regulator CcpA in L. bulgaricus and its ccpA deletion mutant strain was investigated in the present study from the differential proteomic perspective. A physiological and metabolic study (growth, metabolite production and stress tolerance) previously performed on the same experimental system [22], appropriately integrated to provide deeper insight into the cellular processes mediatized by CcpA in L. bulgaricus.
In this work, inactivation of ccpA caused changes in the expression of some proteins related to stress response in cells. Among the differentially expressed proteins, heat shock protein (Hsp20), cold stress protein (Csp), elongation factor Tu (EF-Tu), and molecular chaperones (GroEL, GroES, and DnaK) were down-regulated in the mutant strain compared to the parental strain, while the transcriptional repressor protein HrcA, the DNA repair a Average fold-change was calculated as the ratio of the CcpA-negative mutant of L. bulgaricus ATCC11842 to the parental strain for up-regulated proteins and as the negative reciprocal values for down-regulated proteins, Proteins with the fold-change of ≥ 1.5 or ≤ − 1.5 and P-value < 0.05 were considered to be significantly up-regulated or down-regulated, respectively, based on statistic analysis for three biological replicates [19] protein RecO, the regulatory protein RecX and the SOS regulatory protein LexA were up-regulated. The expression of HrcA was up-regulated in the mutant strain, showing that it was negatively regulated by CcpA in L. bulgaricus. HrcA is a negative regulator of the class I heat shock gene (dnaK and groELS operons) that prevents these operons from inducing heat shock [24]. As expected, the expression of DnaK and GroEL was downregulated in the mutant strain, indicating that these operons were activated by CcpA. Our previous studies have found that ccpA inactivation reduced the heat tolerance of L. bulgaricus [22], which was consistent with the proteomic data. The results were in agreement with studies on L. plantarum with ccpA inactivation [11].
EF-Tu not only promotes the binding of aminoacyl-tRNA to ribosomes but also participates in protein folding and protects cells against environmental stress. Upregulation of the elongation factor EF-G in L. bulgaricus improved its tolerance to salt stress [25]. In L. plantarum, the expression of EF-Tu was found to be reduced by inactivation of ccpA [11], consistent with our data.
Our previous study showed that aerobic growth significantly enhanced tolerance to heat and oxidative stress in L. bulgaricus and its ccpA deletion mutant strain [22]. This enhancement may have practical significance for the production of starter, for instance, although it is far more economic than freeze drying, spray drying imposes oxidative damage and stress on cells [26].
Thioredoxin (Trx) is an important protein that participates in many redox reactions and regulates the function of some enzymes in microorganisms. A transcriptomic study suggested that the trxA2 and trxB1 genes play a key role in the oxidative stress response mechanism of L. plantarum [27], the expression levels of genes related to the stress response and sulfur-containing amino acid biosynthesis were significantly affected by the overexpression of trxB1. Studies on the aerobic growth of L. lactis have shown that Trx also was involved in carbon and lipid metabolism [28]. Zotta et al. [10] reported that the expression of NADH oxidase and Pox was significantly up-regulated in L. plantarum, which was considered to be the main reason for the improved oxidative tolerance of the strain after ccpA knockout. In this study, the expression of Trx and iron-sulfur (Fe-S) were significantly increased in the mutant strain due to ccpA inactivation; thus, the expression of these proteins was negatively regulated by CcpA. As previous demonstrated, the mutant strain was more tolerant than the parental strain for oxidative stress [22]. We speculate that these up-regulated proteins involved in redox reactions may enhance the oxidative tolerance of the mutant strain. Collectively, these results indicated that CcpA acts as a pleiotropic regulator in coordinating oxygen, iron and carbon metabolism [14].
The role for CcpA in control of carbohydrate metabolism has been reported in some LAB [12,29]. The regulatory pathways differ among microorganisms, although all are based on PTS/CcpA-mediated signal transduction [30]. In L. lactis and L. plantarum, the pfk-pyk operon is positively regulated by CcpA, which activates gene transcription [9,14]. However, in L. casei, CcpA negatively regulates the expression of the pfk and pyk genes [12]. The pfk-pyk operon in L. bulgaricus is regulated by CcpA, but the regulatory mechanism is unclear. Our previous work showed that the activities of key enzymes (Ldh, Pyk, and Pfk) in the glycolytic pathway are significantly decreased by ccpA inactivation [22]. In this study, Pfk, Pyk and LdhA were found to be down-regulated at the protein expression level in the mutant strain compared with the parental strain; additionally, changes in their gene transcription levels were consistent with those in the protein expression levels. Thus, our results demonstrated that CcpA positively regulated the pfk-pyk operon in L. bulgaricus. CcpA directly or indirectly controls the expression levels of enzymes involved in the pyruvate pathway, which facilitate the utilization of available carbohydrates [9,30] Pyruvate plays a very important role in growth and metabolism since it affects the energy and redox status of the cells [30]. In this work, inactivation of ccpA caused pyruvate flux away from lactate production in L. bulgaricus under aerobic conditions. As shown in Fig. 5, the expression of some enzymes involved in glycolysis was significantly down-regulated while that of Pox and Ack was significantly up-regulated. Previous work on growth and metabolite production also showed decreases in the growth rate, the utilization of glucose, the production of lactic acid, and increases in the yield of acetic acid in the mutant strain [22], which verified our proteomic results. Collectively, these findings clearly showed that CcpA mediated the shift from homolactic fermentation to mixed acid fermentation in L. bulgaricus, which was in agreement with the findings of Zotta et al. [10] and Mazzeo et al. [11] for L. plantarum. Fig. 4 Relative expression levels of carbohydrate metabolism and stress response related genes as measured by qRT-PCR. Relative expression fold changes were calculated relative to the transcript levels in the ccpA-inactivated mutant strain compared to the parental strain for up-regulated genes and as the negative reciprocal values for down-regulated genes. The data were normalized to the transcription level of 16S rRNA and are expressed as the mean ± SD of three biological and technical replicates. Statistical significance is identified as *P < 0.05 Fig. 5 Overview of the metabolic pathway of lactate and acetate production in L. bulgaricus. The red arrows indicated proteins were down-regulated by ccpA inactivation in L. bulgaricus, as observed at the protein level and transcriptomic level. The green arrows indicated proteins that were up-regulated. Pfk 6-phosphofructokinase, Pgk phosphoglycerate kinase, Pyk pyruvate kinase, LdhA d-lactate dehydrogenase, Pox pyruvate oxidase, Ack acetate kinase Inactivation of ccpA altered the expression of some proteins involved in protein synthesis and translation in L. bulgaricus. For example, the expression levels of various ribosomal genes were up-regulated in the mutant strain. Many studies have showed that upregulation of ribosomal in LAB under stress conditions, indicating that these ribosomes not only are involved in protein synthesis but also may be sensors for environmental changes [31]. Clp proteases belong to the family of ATPases, which plays a key role in refolding or degrading damaged proteins in low-GC Gram-positive bacteria [32]. In this study, the expression of multiple Clp proteases was down-regulated in the mutant strain compared to the parental strain, similar to the study of the ccpA mutant strain of L. plantarum [10].
Remarkable changes in the proteome of L. bulgaricus were caused by the ccpA inactivation, thus confirming its role as a pleiotropic regulator under acerbic conditions. Moreover, the role of CcpA in aerobic growth was also reported by Mazzeo et al. [11] in L. plantarum compared with anaerobic growth at proteomic level, which brought new insights into the regulatory mechanism of CcpA response to oxidative stress. Therefore, comparison with anaerobic growth could provide a novel direction for our future research.

Conclusions
This work is the first report of the role of CcpA in regulating carbohydrate metabolism and stress response of L. bulgaricus under acerbic conditions at proteomic level. The proteomic results coherently match with physiological and metabolic features previous demonstrated by Li et al. [22] on the same experimental system. A total of 132 proteins were identified to be differentially expressed between the mutant strain and the parental strain. Inactivation of ccpA negatively affected the expression of key enzymes involved in glycolysis, supporting the conclusion that CcpA mediated the shift from homolactic fermentation to mixed acid fermentation in L. bulgaricus. This proteomic results provide new knowledge in the role of global regulator CcpA in L. bulgaricus, and may have practical significance for the production of starter and probiotic cultures.

Bacterial strains and growth conditions
Lactobacillus bulgaricus ATCC11842 strain was purchased from the American Type Culture Collection (Manassas, VA, USA). Inactivation of the ccpA gene was achieved by homologous recombination as described previously by Li et al. [23]. Both strains were stored at − 80 °C in 20% glycerol in this experiment. L. bulgaricus and its mutant ccpA deletion strain were cultivated at 37 °C in MRS broth under aerobic conditions in Erlenmeyer flasks (the medium volume was 1/10 of the flask volume) with shaking (150 rpm). Bacterial samples were harvested during exponential growth phase (final OD 600 = 1.0) and immediately frozen in liquid nitrogen until further use [23].

Protein preparation
Total protein extraction was performed as previously reported with slight modifications [33]. Briefly, all strains were cultivated overnight in MRS broth at 37 °C and harvested by centrifugation at 8000×g for 20 min at 4 °C. After the cell pellets were washed with 50 mM Tris buffer, 1 mL of lysis buffer (7 M urea, 2 M thiourea, 4% CHAPS, 1 mM DTT, 1 mM PMSF, and 2 mM EDTA) was added, and the samples were sonicated in an ice bath until the bacterial solution was clarified. Then, the samples were transferred into tubes and centrifuged at 20,000×g for 20 min at 4 °C. Then, the supernatant was collected for SDS-PAGE on 5% stacking gels and 12.5% resolving gels. Gels were stained with Coomassie Blue. Protein concentrations were quantified using the Bradford method.

Protein digestion and iTRAQ labeling
Protein digestion was performed using the FASP method [34]. Briefly, 100 μg of each sample was added to precooled acetone (volume ratio of acetone:sample = 6:1), and the mixture was precipitated for 1 h at − 20 °C. The protein precipitate was fully dissolved using dissolution buffer (50 μL) and 1% SDS (1 μL) from the iTRAQ kit (Applied Biosystems, Foster City, CA, USA). Next, 4 μL of reducing reagent was added and reacted at 60 °C for 1 h. Then, 2 μL of cysteine-blocking reagent was added at room temperature for 10 min in the dark. The samples were transferred to new ultrafiltration tubes and centrifuged at 12,000×g for 20 min. After the lower layer of solution in the collection tubes was discarded, 100 μL of dissolution buffer was added and centrifuged under the same conditions described above. This step was repeated three times. The collection tubes were replaced with new tubes, and 4 μg of trypsin was added to the ultrafiltration tubes and incubated overnight at 37 °C. The enzymatically decomposed peptides were collected by centrifugation, and 50 μL of dissolution buffer was added to the ultrafiltration tubes. The lower layer of solution was then collected and combined with the previously obtained solution. The peptides from each sample were desalted, and the protein concentration was determined by measuring the OD at 280 nm. iTRAQ reagent was dissolved in 60 μL of isopropanol according to the manufacturer's instructions and then mixed with the samples at room temperature for 2 h. All samples were combined after labeling and dried by vacuum centrifugation for strong cation exchange (SCX) fractionation.

SCX fractionation
iTRAQ-labeled peptides were fractionated by SCX chromatography according to the experimental conditions described previously [34]. Configured transfer buffer A (10 mM KH 2 PO 4 and 25% acetonitrile, pH 3.0) and elution butter B (500 mM KCl, 10 mM KH 2 PO 4 , and 25% acetonitrile, pH 3.0) were used for liquid chromatography. The samples were dissolved in 4 mL of buffer A and loaded into a 4.6 × 100 mm polysulfoethyl column containing 5-µm particles (PolyLC Inc., Columbia, MD, USA) in a sampler for separation. The column was equilibrated in buffer A for 10 min. Peptides were eluted with buffer B at a flow rate of 1 mL/min as the following gradient: 0-8% buffer B for 20 min, 8-52% buffer B for 25 min, 52-100% buffer B for 3 min, 100% buffer B for 7 min and then buffer B was reset to 0% for 5 min, for a total of 60 min. Based on the UV absorbance at 214 nm, the fractions were collected in 1 min intervals using a collector and then vacuum dried for subsequent analysis.

Liquid chromatography/tandem mass spectrometry (LC-MS/MS) analysis based on Q Exactive technology
Mass spectrometry analysis was carried out as described by Luo et al. [35] with moderate modifications. Each fraction was dissolved in buffer C (0.1% formic acid) and then analyzed using the EASY-nLC 1000 HPLC system (Thermo Fisher Scientific, Waltham, MA, USA). A total of 10 μL of the peptide mixture was injected onto a Pep-Map C18 trap column (100 μm × 2 cm, 3 μm, 100 Å, Thermo Fisher Scientific, Rockford, IL, USA) via a sampler and were then separated on an Acclaim PepMap C18 column (75 μm × 10 cm, 3 μm, 100 Å, Thermo Fisher Scientific) at a flow rate of 300 nL/min with an elution gradient of 0-50% buffer D (84% acetonitrile and 0.1% formic acid) for 55 min and 50-100% buffer D for 55-57 min. Buffer D was maintained at 100% after 57 min.
MS data were acquired using a data-dependent top10 method in the mass spectrometer, which was operated in positive ion mode with a mass range of 300-1800 m/z for high-energy collisional dissociation (HCD) fragmentation. The MS spectra and HCD spectral scans were acquired at a resolution of 70,000 and 17,500 at m/z 200, and the maximum injection times were set to 10 and 60 ms, respectively. The instrument was operated using a peptide recognition pattern. The other identification parameters were as follows: a normalized collision energy of 30 eV, an isolation window of 2 m/z, a dynamic exclusion duration of 40 s, and an underfill ratio of 0.1%.

Protein identification and data analysis
The raw mass spectrometry data were submitted to Mascot 2.2 software (Matrix Science, London, UK) through Proteome Discoverer 1.4, and searches were performed against the L. bulgaricus database, which contains 24,177 protein sequences, for protein identification and quantification. The screening conditions for the trusted proteins included a false discovery rate of no more than 1%, which was calculated with a decoy database model, and the protein ratios were quantified based on the median of only unique peptide ratios in Mascot. The t-test was used to evaluate statistical significance. Proteins with the fold-change of ≥ 1.5 or ≤ −1.5 and P-value < 0.05 were considered significantly up-regulated or down-regulated, respectively. The Blast2GO program was used to annotate the molecular function, biological process and cellular component information for the target proteins. Metabolic pathway analysis of the differentially expressed proteins was performed using KEGG.

RNA extraction and transcriptional analysis (qRT-PCR)
Bacterial cells were harvested by centrifugation at 8000×g for 20 min at 4 °C as described above. RNA extraction and complementary DNA (cDNA) synthesis were separately performed using the RNeasy Midi Kit (Tiangen, Beijing, China) and cDNA RT reagent kit (Taraka, Dalian, China) according to the manufacturer's protocol. The sequences of the primers are listed in Additional file 1: Table S2. qRT-PCR was performed with Taq SYBRGreen qPCR Premix in a LightCycler instrument (ABI PRISM 7500 System, Applied Biosystems, Foster City, CA, USA) according to the manufacturer's instructions. The thermal profile suggested by the manufacturer was 30 s of denaturation at 95 °C followed by 40 cycles of 10 s at 94 °C and 30 s at 60 °C. Three independent replicates of each sample were tested and the 2 −ΔΔCt method was used to calculate the expression levels of the target genes. For gene expression normalization, 16S ribosomal RNA (16S rRNA) was used as the internal standard for mRNA expression [36].
Additional file 1: Table S1. KEGG pathway analysis of differentially expressed proteins. Table S2. Primers used for qRT-PCR.