Multi-omics based changes in response to cadmium toxicity in Bacillus licheniformis A

Abstract

Cadmium (Cd), a widespread substance with high toxicity and persistence, is known to cause a broad range of adverse effects in all living organisms. However, many microorganisms have special detoxification mechanisms for Cd. In this study, we investigated the response to Cd in Bacillus licheniformis A at the omics level. Proteomic biomarkers were identified, including peroxiredoxin, nucleoside diphosphate kinase, trigger factor, and elongation factor Tu. These proteins suggested that Cd could induce oxidative stress, disturbance in energy metabolism, and protein synthesis in B. licheniformis A. Furthermore, metabolites (glutamate, glycine, and S-adenosylhomocysteine) involved in the glutathione synthesis pathway were also affected. In addition, some metabolic biomarkers were validated by relating to respective proteins, including 50S ribosomal protein L7/L12, trigger factor, peroxiredoxin and superoxide dismutase in B. licheniformis A. This study provides new insights into the utilization of an omics approach in bacteria to assess the effects of Cd pollution in the environment.

---

1. Introduction

Environmental pollution caused by cadmium (Cd) has received much attention in recent years because of its toxicity and stability. Sources of Cd include metallurgy, mining, fossil fuel combustion, the manufacture and disposal of rechargeable nickel–cadmium batteries, and the utilization of pesticides and fertilizers.^1,2 It binds to organic compounds by forming bonds with nitrogen and sulfur, inactivating proteins, and potentially leading to negative side-effects, including extreme toxicity, carcinogenicity, and mutagenicity.³ It is easily absorbed and accumulated by lower organisms and subsequently transferred to higher trophic levels in food chains. Due to Cd's severe toxic effects, fast accumulation through the food chains has caused serious damage to ecosystems and led to economic losses and negative impacts on food chains and human health. For this reason, the International Agency for Research on Cancer has classified Cd as a human carcinogen belonging to group 1 agents in 1993.⁴

Bioremediation with microorganisms has become an attractive alternative to conventional techniques for toxin removal. Furthermore, there are some studies that have demonstrated the efficiency of metal removal by bacteria.^5–7 Some microorganisms can tolerate heavy metals by using strategies such as exclusion by permeability barriers, intra- and extra-cellular sequestration, resistance to oxidative stress, extracellular biosorption, active transport efflux pumps, enzymatic detoxification, and a reduction in the sensitivity of cellular targets to metal ions.^8–11 For example, as a Cd²⁺-resistance determinant the CadA gene in Listeria monocytogenes encodes an integral protein which belongs to the P-type ATPase family.¹² These ATPases are membrane proteins coupling ATP hydrolysis and active transport of cations or anions across the membrane in which they are embedded.

Toxicity of various heavy metals to bacteria in the environment has been extensively studied for many years.^13–17 Recently, omics technologies have increasingly been used in a variety of different contexts to gain a comprehensive understanding of nearly all components within a biological entity. They include transcriptomics, proteomics and metabolomics, which refer to the pool of RNA transcripts, proteins and metabolites in a cell, tissue biofluid, or indeed whole organisms.^3,18 These technologies and related data processing techniques provide powerful tools which have the potential to be of great value in ecotoxicological research.^18–23 Proteomics has been widely used to study the mechanisms of heavy metal toxicity in bacteria. Mass spectrometry (MS), two-dimensional electrophoresis (2-DE), protein affinity techniques, high-throughput immunochemistry and bioinformatics are among the major proteomics technologies used for understanding metal toxicity. Kılıç et al. used a proteomics approach to study the Cr(VI) stress response of Pseudomonas aeruginosa.⁹ Additionally, Siripornadulsil et al. identified and characterized protein expression in Cupriavidus taiwanensis KKU2500-3 during growth under Cd stress by using 2-DE and MS.²⁴ Metabolites, as the end products of cellular regulatory processes, can be regarded as the ultimate response of biological systems to genetic or environmental changes.³ Metabolomic analyses have multiple advantages in that results obtained at a molecular biological level may also provide insight into mechanisms of toxicity.²⁵

Bacillus licheniformis, a Gram-positive soil bacterium, classified as GRAS (generally recognized as safe), was shown to be an efficient plant growth promoting rhizobacterium.²⁶ Earlier study showed B. licheniformis is able to improve the growth and development of the host plant in Cd contaminated soils.²⁷ However, little information is available regarding the mechanisms of B. licheniformis resistance to Cd.

In this study, proteomic, transcriptomic and metabolomic analyses were combined to explore the effects of Cd exposure in B. licheniformis A. Bacterial proteins were isolated by 2-DE and identified by peptide mass fingerprinting (PMF). Differential gene transcription of some annotated candidates was further assessed by Real Time PCR while proton (¹H) nuclear magnetic resonance (NMR) spectroscopy was utilized for the metabolomic analyses.

2. Materials and methods

2.1. Bacterial strain

B. licheniformis A (CGMCC no. 4324) is isolated and characterized in our laboratory. A single colony was propagated in beef extract peptone medium at 37 °C overnight. The cultures were separately used to determine the minimum inhibitory concentrations (MICs) for an ancestor strain by Cd and for resistant strains by progressive Cd²⁺ gradient concentration domestication. Growth was monitored spectrophotometrically (measurement of the optical density (OD)) at 560 nm at intervals of 6 h for 48 h.

2.2. Transmission electron microscopy analysis

Transmission electron microscopy (TEM) analysis of the bacteria before and after coming in contact with Cd²⁺ was carried out as follows: firstly, the cell pellets were fixed using 2.5% (v/v) of glutaraldehyde in 0.1 M phosphate buffered saline (PBS, pH 7.4) for 2–3 h at 4 °C followed by washing with 0.1 M PBS, post-fixation was carried out with 2% (v/v) osmium tetroxide for 2 h in the same buffer at 4 °C. After several washes in buffer, the cells were dehydrated using increasing concentrations of 99.9% (v/v) ethanol (30, 50, 70, 80, 90, and 100%) at 10 min intervals followed by 5 min washing in 100% (v/v) acetone. The cells were then embedded in 50% and 100% (v/v) of epoxy resin in acetone for 15 min each. The cell pellets were again infiltrated with 100% (v/v) epoxy resin and cured at 60 °C overnight. Specimens of 100 nm thickness were sectioned from the embedded blocks using a microtome (HM505E, Carl Zeiss International, Germany) and mounted on 200-mesh copper grids. The specimens were stained with uranyl acetate, post-stained with lead-citrate and viewed using a transmission electron microscope (TEM) (H-7650, Hitachi Ltd., Japan).

2.3. Proteomic analysis

2.3.1. Protein extraction. Protein extracts of B. licheniformis A were prepared from cells obtained from 500 mL of cultures grown in the absence or presence of 1000 ppm Cd²⁺. After 24 h incubation, the cells were harvested by centrifugation at 8000g for 20 min at 4 °C. Proteins were isolated as described previously with slight modifications.¹⁰ Briefly, after harvesting the cells, the pellets were washed in 20 mL water three times and resuspended in 10 mL water containing a protease inhibitor cocktail (Amresco). The cells were disrupted by intermittent ultrasonic treatment on ice. Intact cells and cellular debris were removed by centrifugation at 8000g for 10 min and the supernatants were collected. The sample was mixed with ice-cold acetone (1 : 4) and incubated at −20 °C overnight. The protein pellet was further centrifuged at 12 000g for 30 min at 4 °C and resuspended in 7 M urea, 2 M thiourea, 4% CHAPS, 65 mM dithiothreitol (DTT), and 0.2% Bio-Lyte (Bio-Rad) sample buffer (for isoelectric focusing (IEF)). Protein concentrations were determined using a non-interference protein assay kit (Sangon Biotech Co., Ltd, Shanghai, China) against a bovine serum albumin standard curve. Protein samples were stored at −80 °C for further analysis.

2.3.2. 2-DE analysis. 2-DE was performed according to a previously-described protocol with slight modifications.²² Protein extracts containing 300 μg of protein were dissolved in IEF sample buffer and this mixture was used to rehydrate a pH 4–7 linear immobilized pH gradient (IPG) strip (7 cm length, Bio-Rad). IEF was performed in a Bio-Rad IEF cell using a 7 cm focusing tray. The total focusing was set at 20 000 V h and a maximum voltage of 4000 V. After the strip was equalized in 2% (w/v) DTT and 2.5% (w/v) iodoacetamide (IAA) (6 M urea, 20% glycerol, 2% SDS, 0.375 M pH 8.8 Tris-HCl) for 15 min, respectively, proteins were separated in the second dimension by using a 12% acryl amide SDS-PAGE gel and electrophoresed with voltages of 80 V and 120 V. All 2-DE experiments were carried out in triplicate.

2.3.3. Visualization of protein spots and image analysis. Protein spots were visualized by staining with coomassie brilliant blue (CBB) R 250. The stained gels were scanned in an Image Scanner (Bio-Rad) and differential analysis was performed using the PDQuest 2-DE analysis (Version 8.0.1) software. Following background subtraction and spot detections, the gel patterns were matched to each other by visual comparison. Altered spots were standardized and then compared based on their volume percentages in the total spot volume over the whole gel image. Significantly changed spots were selected if their intensity increased/decreased ≥1.5-fold or after complete appearance and disappearance.²³

2.3.4. Preparation of peptide samples and mass spectrometry analysis. Protein spots of interest were manually excised from the gel and cut into 1–2 mm³ pieces. They were destained with 50 mM ammonium bicarbonate (ABC) in 50% (v/v) methanol for 30 min at 37 °C. The gel pieces were dried in a SpeedVac vacuum drier and rehydrated at 4 °C for 30 min in 10 μL of digestion solution (50 mM ABC/10% (v/v) acetonitrile (ACN), 20 ng μL⁻¹ of trypsin [Promega]). Digestion was carried out overnight at 37 °C with shaking. The peptides were extracted from the gel with 60 μL 2.5% (v/v) trifluoroacetic acid (TFA)/90% (v/v) ACN. The supernatants were dried in a SpeedVac and the pellets dissolved in 1.5 μL of 30% (v/v) ACN/0.1% (v/v) TFA. The samples were hand-spotted on a ground steel target plate (Applied Biosystems) using a matrix solution that contained 5 mg ml⁻¹ a-cyano-4-hydroxycinnamic acid dissolved in 50% (v/v) ACN/0.1% (v/v) TFA. The mixture was allowed to dry under ambient conditions.

The samples were analyzed on an Autoflex™ speed matrix-assisted laser desorption ionization time-of-flight mass spectrometer (Bruker Daltonics, Bremen, Germany) using a 200 Hz, 355 nm nitrogen laser for desorption and ionization. Default mode was used with an acquisition mass range from 700 to 3200 Da and the total acceleration voltage was 20 kV. The instrument was internally calibrated using tryptic autolytic peaks. All spectra were smoothed, noise filtered, and peak lists generated using flexAnalysis version 2.0 (Bruker). Then, the measured tryptic peptide masses were transferred to the BioTools (Bruker) program to be searched against the taxonomy of Bacteria in the protein database using the Mascot search engine (http://www.matrixscience.com). For protein searches, monoisotopic masses were used, considering a peptide tolerance of 50 ppm and allowance of one missed tryptic cleavage site. When MS/MS was carried out, a tolerance of 0.5 Da was acceptable. Fixed modifications were set for carbamidomethylation of cysteine and variable modifications were set for oxidation of methionine. Protein identifications were considered as reliable when the Mascot score was ≥84 (Mascot score was calculated as −10 × log P, where P is the probability that the observed match is a random event). This is the lowest score indicated by the program as significant (p < 0.05) and indicated by the probability of incorrect protein identification. For searches done with identified proteins were only considered if having at least two peptides with 95% confidence. They were further validated with inspection of the assigned sequence. Functional characterization of the proteins was predicted with Expert Protein Analysis System (Expasy) proteomics server of the Swiss Institute of Bioinformatics (SIB) (http://www.expasy.org/) based on Gene Ontology (GO) annotation under the category of biological process in UniProtKB (http://www.uniprot.org).

2.4. Confirmation of differentially expressed proteins by real time PCR

Real time PCR (RT-PCR) was performed using three replicates for different concentrations of Cd treatment groups. Total RNA was extracted from log-phase B. licheniformis A with Trizol reagent (Takara), according to the manufacturer's protocol. Single-strand cDNA synthesis was performed according to the M-MLV cDNA Reverse Transcription Kit (Sangon) Usage information. The reactions were incubated at 42 °C for 1 h, and terminated by heating at 70 °C for 10 min. RT-PCR was performed for 4 genes differentially expressed during Cd treatment using gene specific primers (see Table 1). Primers were chosen to have melting temperatures of 59–60 °C and the amplified product size was approximately 200 base pairs long. 16s rRNA of B. licheniformis A was used as a reference housekeeping gene for normalization. The RT-PCR amplifications were carried out in a total volume of 20 μL containing 10 μL 2 × SYBR Green Mix (Takara), 4 μL of the 1 : 20 diluted cDNA, 1 μL of each primer (20 mM) and 4 μL PCR grade water. RT-PCR amplification was performed using a Rotor-Gene 6000 real-time PCR detection system (Qiagen, Germany). The RT-PCR parameters were as follows: denaturation at 95 °C for 10 min, followed by 40 cycles of 95 °C for 15 s, 60 °C for 20 s, 72 °C for 20 s. Melting analysis of the amplified products was performed at the end of each PCR run to confirm that a single PCR product was produced and detected. Quantification of the PCR product was performed using the −△△C_t method. The value obtained denoted the n-fold difference relative to the calibrant (control samples). The data were presented as relative mRNA expression levels (means ± S.D., n = 3), and the results subjected to paired t-test analysis and p ≤ 0.05 were considered statistically significant.

Table 1 Primers used in this study

Primer	Sequence (5′–3′)	Name
P1	F-CGGCGTGCCTAATACATGC	16s rRNA (reference gene)
P2	R-GGCAGGTTACCCACGTGT
P3	F-ACGGTGTTGCAATCGATCCT	PTS system glucose-specific transporter subunit
P4	R-CGTGAGCTTCGAAACCTTCAC
P5	F-GAAGACGCTCAAAAACGCGT	Trigger factor
P6	R-CGCTTCCAAGAGCTTGCTTG
P7	F-ATTCGGTAGCTTCGATGCGT	Superoxide dismutase
P8	R-CGTTTTTCAGCAGCGTTCCA
P9	F-CGCTTGAAGCTGGCCTTG	Universal stress protein
P10	R-CAAGGACATCGCATTTCGCA

---

2.5. Metabolomics analysis

2.5.1. Metabolite extraction. Metabolites were extracted using a methanol-water based extraction as described by previously-described protocols^28,29 with slight modifications. Cells were collected at their log phase (identical to the protein extraction conditions) and flash frozen in liquid nitrogen to quench metabolic activity before metabolomic extraction.³ Pellets were subsequently mixed with a methanol-water mixture (2 : 1, v/v) and sonicated for 15 min. Cell debris was removed by centrifugation at 8000 g for 10 min. The supernatants were evaporated to dryness and stored at −80 °C prior to NMR analysis.

2.5.2. ¹H NMR parameters. The dried extracts were rehydrated in 600 μL D₂O with adequate mixing, then passed through an ultrafiltration membrane (Millipore Amicon® ULTRA 3 KDa) by centrifugation at 13 000 rpm for 30 min at 4 °C. 450 μL of the supernatant and 50 μL 5 mM Anachro Certified DSS Standard Solution were mixed together then transferred into 5 mm NMR tube. The samples were analyzed using a Bruker AV III 600 MHz spectrometer equipped with an inverse cryoprobe operating at 600.13 MHz for ¹H frequency at 298 K. Spectra were collected using a solvent suppression pulse sequence based on a 1-dimensional NOESY sequence with presaturation during relaxation delay and mixing time. Generally, 32 free induction decays (FIDs) and 128 transients were collected for each sample over a spectral width of 12 019.231 Hz at 25 °C according to Jamers et al.³

2.5.3. Data analysis of ¹H NMR spectra. The signal of free induction decays (FIDs) were imported to Chenomx NMR suite (version 7.6, Edmonton, Canada). All NMR data were Fourier-transformed, phased, referenced and baseline-corrected in the Chenomx NMR suite. DSS-d₆ served as an internal standard for both chemical shifts and quantification. The peak of DSS-d₆ was utilized as an internal reference set to 0.0 ppm. Spectral identification and quantification of metabolites was carried out by spectral deconvolution using the Chenomx NMR suite database supplement.^30,31 Finally, all spectral data were normalized using a Pareto scaling method. Principle component analysis (PCA) based multivariate data analysis processing was performed with the R software package.

3. Results

3.1. Effect of Cd²⁺ on the growth of B. licheniformis A strain and TEM analysis

To study the effect of Cd²⁺ on the growth of B. licheniformis A, the cells were cultivated in the presence of increasing concentrations of Cd²⁺. The measured optical densities (at 560 nm) of control and Cd²⁺ treated cells over time are shown in Fig. 1(a). Growth rate were significantly affected by Cd²⁺ decreasing from 0.258 h⁻¹ in the control sample to less than 0.081 h⁻¹ in the presence of 1000 ppm Cd²⁺. B. licheniformis A cells treated with 500 ppm Cd²⁺ did not show obvious differences in growth rate relative to the control sample. However, the lag phase for 1000 ppm Cd²⁺ contaminated cultures was longer and its maximum growth rate was slower than the control. It can be seen from Fig. 1(a) that the stress-free cells reached the mid-log phase faster than the stressed cells which needed a longer time period for adaptation. Finally, we chose the strains that were cultured in a beef extract peptone medium with 1000 ppm Cd²⁺ as the experimental group and the samples lacking Cd²⁺ as the control group for the proteomic studies.

Fig. 1 Growth curve of B. licheniformis A cultured without or with different concentrations of Cd²⁺ (a), and related control (b), 500 ppm Cd²⁺ (c), and 1000 ppm Cd²⁺ (d) TEM micrographs.

Transmission electron micrographs of varying concentrations of Cd²⁺-treatment cells showed that the surface of the normal cell is smooth, with even edges (Fig. 1(b)). However, while the bacterial cell structure remained intact with the increase of Cd²⁺ concentration (Fig. 1(c) and (d)), separation between the bacterial cell wall and membrane was noticeable (Fig. 1(d)). A noticeable strongly electron-dense area between the cell wall and cytoplasm clearly indicates the deposition of Cd²⁺ onto the bacterial cells functional groups with further translocation into the cytoplasmic region (Fig. 1).

3.2. Differential protein expression

A comparative proteomic analysis was used to detect significantly up- or down-regulated proteins in B. licheniformis A in response to the presence of 1000 ppm Cd²⁺ stress. An average of 245 protein spots were detected in the control group gel and 327 protein spots were detected in the experimental group. From these, a total of 145 spots were matched in all gels. A total of 18 spots were selected for identification by MS/MS (Fig. 2). Of these, 13 proteins were up-regulated and 5 proteins down-regulated upon Cd²⁺ stress (Table 2). These proteins were found to be related to energy and primary metabolic processes (PTS system glucose-specific transporter subunit IIA, nucleoside diphosphate kinase, dihydrolipoamide dehydrogenase, acetoin dehydrogenase complex), stress and defense (universal stress protein, trigger factor, peroxiredoxin, superoxide dismutase, manganese superoxide dismutase, hypothetical protein IC5_00055), cell growth/division (naphthoate synthase, putative delta-1-pyrroline-5-carboxylate dehydrogenase, 1-pyrroline-5-carboxylate dehydrogenase, spore coat-associated protein N, response regulator aspartate phosphatase), and signal transduction (elongation factor Tu, 50S ribosomal protein L7/L12).

Fig. 2 Representative 2-DE images of proteins from B. licheniformis A cultured without (a) or with 1000 ppm Cd²⁺ (b). The spots in black boxes were selected to be identified by MS/MS and listed in Table 2.

Table 2 List of protein spots which were differentially expressed in B. licheniformis A induced by Cd²⁺

Spot no^a	Accession no.	Protein name	Mr/pI^b	Protein score^c	Sequence coverage	Expect	Matched peptides	Fold change^d
a Assigned spot no. as indicated in Fig. 2.b Experimental mass and pI.c Mascot score reported.d Fold changes with significances (>1.5 folds and p < 0.05) were calculated using PDQuest 2DE analysis (Version 8.0.1) software.
1	gi|423370999	Hypothetical protein IC5_00055	87411/5.92	444	41%	5.6 × 10^−38	22	3.58
2	gi|206974153	Peroxiredoxin	20864/4.81	425	56%	4.4 × 10^−36	10	2.34
3	gi|301055460	Dihydrolipoamide dehydrogenase	49395/5.13	274	44%	5.6 × 10^−21	13	2.05
4	gi|30265343	PTS system glucose specific transporter subunit IIA	17464/5.41	343	73%	7 × 10^−28	9	2.60
5	gi|217961965	Trigger factor	47185/4.53	283	52%	7 × 10^−22	19	New
6	gi|206977024	Naphthoate synthase	29886/5.56	403	52%	7 × 10^−34	14	1.69
7	gi|30018370	50S ribosomal protein L7/L12	12510/5.72	92	21%	8.1 × 10^−3	8	3.56
8	gi|254739504	Elongation factor Tu	41570/4.77	250	55%	1.4 × 10^−18	17	New
9	gi|30264347	Superoxide dismutase	22650/5.30	337	34%	2.8 × 10^−27	5	2.06
10	gi|306447497	Manganese superoxide dismutase	22616/5.33	168	59%	2.2 × 10^−10	8	1.53
11	gi|52140879	Universal stress protein	16842/5.68	189	84%	1.8 × 10^−12	11	1.55
12	gi|301053152	Nucleoside diphosphate kinase	18849/5.26	315	60%	4.4 × 10^−25	11	1.59
13	gi|206977864	Spore coat-associated protein N	21901/5.76	292	43%	8.9 × 10^−23	8	−2.78
14	gi|30265465	Superoxide dismutase	24122/5.40	156	49%	3.5 × 10^−9	7	1.63
15	gi|206973646	Acetoin dehydrogenase complex	49619/5.65	151	25%	1.1 × 10^−8	7	−2.89
16	gi|229035348	Response regulator aspartate phosphatase	43565/5.73	446	49%	3.5 × 10^−38	17	−1.52
17	gi|196040213	Putative delta-1-pyrr Oline-5-carboxylate dehydrogenase	56446/5.43	249	29%	1.8 × 10^−18	12	−3.86
18	gi|229089424	1-Pyrroline-5-carboxylate dehydrogenase	56445/5.51	432	48%	8.9 × 10^−37	20	−1.92

---

3.3. mRNA-level expression

To further verify the results of protein responses and compare the correlation between protein abundances and gene expressions, four representative proteins, including PTS system glucose specific transporter subunit IIA (spot 4), trigger factor (spot 5), superoxide dismutase (spot 9) and universal stress protein (spot 11), were quantified using qRT-PCR. As shown in Fig. 3, the gene expressions levels of spot 4 (34-fold), spot 9 (26-fold) and spot 11 (5-fold) were positively correlated with the corresponding proteins (2.60-fold, 2.06-fold and 1.55-fold) during Cd²⁺ treatment. The expression of spot 5 (16-fold), as a new protein compared with control, was also observed on the mRNA level.

Fig. 3 qRT-PCR analysis of gene expression level in different Cd²⁺ concentrations treated B. licheniformis A. The mRNA expression levels of the ratio of treatment groups to the control are plotted (*P < 0.05, **P < 0.01).

3.4. Metabolite profiling

Fig. 4 shows the representative ¹H NMR spectra of B. licheniformis A, including amino acids and derivatives (alanine, asparagine, glutamate, glycine, isoleucine, leucine, methionine, valine, S-adenosylhomocysteine, etc.), ammoniums compounds (choline, O-phosphocholine), nucleic acid components (adenosine, cytidine, guanosine, inosine, oxypurinol, uracil, uridine) and organic acids (acetate, formate, lactate, succinate).

Fig. 4 Representative 1-dimensional 600 MHz ¹HNMR spectra of B. licheniformis A extracts. Keys: (1) isoleucine, (2) leucine, (3) valine, (4) ethanol, (5) lactate, (6) alanine, (7) lysine, (8) acetate, (9) glutamate, (10) methionine, (11) succinate, (12) aspartate, (13) asparagine, (14) phenylalanine, (15) betaine, (16) threonine, (17) uridine, (18) adenosine, (19) uracil, (20) guanosine, (21) cytidine, (22) S-adenosylhomocysteine, (23) inosine, (24) tyrosine, (25) oxypurinol, (26) formate, (27) 4-aminobutyrate, (28) ethylene glycol, (29) glycine.

Principal component analysis (PCA) was performed on the NMR spectral datasets of B. licheniformis A extracts from control and Cd²⁺-treated groups. Examination of the score plot (Fig. 5) in the area defined by the first two principal components (97.8% of the total variance) revealed two clusters of B. licheniformis A extracts with a clear separation along the first principal component (PC1, 95.9%). The second principal component (PC2) explained 1.9% of the variance. The PCA resulted in separations between control and Cd²⁺-treated groups, as shown in Fig. 5.

Fig. 5 Principal component analysis (PCA) score plot of PC1 versus PC2 for ¹H NMR spectra of B. licheniformis A extracts (n = 5). The score plot in the area defined by the first two principal components (97.8% of the total variance) revealed two clusters according to Cd²⁺-treatment.

Further analysis was performed on the NMR spectral data and quantified metabolite concentrations to identify potential metabolic biomarkers induced by Cd²⁺ (Fig. 6). The concentrations of almost all of identified metabolites were significantly decreased in Cd²⁺-treatment samples, especially in two different sulfur-containing amino acids. In contrast, the concentration of lactate increased upon Cd²⁺ stress. Compared to these metabolites in control group, choline and O-phosphocholine were uniquely detected in Cd²⁺-treatment group. In agreement with B. licheniformis from Genes and Genomes (http://www.genome.jp/kegg/) and Uniprot (http://www.uniprot.org/), Fig. 7 displays the schematic presentation of whole proteomic and metabolomic responses in B. licheniformis A challenged by Cd²⁺.

Fig. 6 Changes in metabolite concentrations in B. licheniformis A between control and Cd²⁺-treatment groups.

Fig. 7 Schematic representation of molecular response mechanisms in B. licheniformis A with Cd²⁺-treatment according to B. licheniformis of Genes and Genomes (http://www.genome.jp/kegg/) and Uniprot (http://www.uniprot.org/). Abbreviation: ala, Alanine; Asp, aspartate; Asn, aspsrsgine; Cys, S-adenosylhomocysteine; DLD, dihydrolipoamide dehydrogenase; Gly, glycine; GSH, glutathione; Ile, isoleucine; Leu, leucine; Lys, lysine; Met, methionine; NDK, nucleoside diphosphate kinase; NS, naphthoate synthase; Phe, phenylalanine; PTS, PTS system glucose specific transporter subunit IIA; RP, 50S ribosomal protein L7/L12; Thr, threonine; Tyr, tyrosine; Val, valine.

4. Discussion

4.1. Effects of Cd²⁺ on the proteome of B. licheniformis A

B. licheniformis A modulated the expression of several proteins when exposed to high concentrations of Cd²⁺. Fig. 7 demonstrated the pathways involved in the response of B. licheniformis A to Cd²⁺ challenge. In Cd²⁺-treated samples, the responsive proteins (Table 2) related to metabolism were involved in energy and primary metabolisms. In this study, the significant up-regulation of the PTS system glucose specific transporter subunit IIA might imply that this protein contributes to the defense system of B. licheniformis A. The phosphoenolpyruvate-dependent sugar phosphotransferase system (sugar PTS), which is a major carbohydrate active transport system, catalyzes the phosphorylation of incoming sugar substrates concomitantly with their translocation across the cell membrane.³² Therefore, the up-regulation of PTS system glucose specific transporter subunit IIA probably indicated that bacteria up-regulate their energy transportation system components to meet the needs of their growth under heavy metal stress. Dihydrolipoamide dehydrogenase (DLD) belongs to the class-I pyridine nucleotide-disulfide oxidoreductase family. It serves to re-oxidize lipoic acid residues in the formation of acyl-CoA or succinyl-CoA in the citric acid (TCA) cycle.³³ Nucleoside diphosphate kinases (NDKs) are enzymes which catalyze the exchange of phosphate groups between various nucleoside diphosphates.¹⁸ In the TCA cycle, NDKs can convert guanosine triphosphate (GTP) to adenosine triphosphate (ATP). Because the bacteria required ATP as an energy source under Cd stress, the NDK was probably up-regulated in B. licheniformis A to produce more ATP to compensate for the energy loss, and stressed growth conditions.

Some stress proteins, including molecular chaperones, heat shock proteins and starvation proteins, are easily induced in response to a wide range of stress challenges, such as heavy metals, starvation, oxidation, anaerobiosis, unsuitable temperatures and different pH value.⁹ Rince et al. demonstrated that the general stress proteins DnaK and GroEL were induced in Enterococcus faecalis in response to heat, ethanol, bile salts, NaCl, H₂O₂ and pH shifts.³⁴ Under Cd stress, the up-regulation of universal stress protein might imply that this protein contributes to the defense system of B. licheniformis A. Molecular chaperones are wildly distributed stress proteins conserved among almost all organisms. In addition to helping newly synthesized proteins, molecular chaperones play a key role in protein export. Trigger factor which is reported to function as a peptidyl-prolyl cis–trans isomerase, acts as a chaperone by maintaining the newly synthesized protein in an open conformation.³⁵ This protein was solely expressed in the Cd-treatment group, possibly reflecting a series of adaptive changes when faced with environmental stress.

There is a mass of literature showing that Cd can induce oxidative stress in various species.^2,3,8,36 The up-regulation of peroxiredoxin, superoxide dismutase, manganese superoxide dismutase was observed in this study suggesting that antioxidant enzymes are useful in protection against Cd toxicity.

Elongation factor Tu (EF-Tu) is a protein which promotes the GTP-dependent binding of aminoacyl-tRNA to the A-site of ribosome during protein biosynthesis.³⁷ The new induction of EF-Tu was also showed in P. fluorescens ATCC 948 treated with 0.1 mM cobalt.³⁸ Ribosomal proteins are also over-expressed at high heavy metal concentrations. The up-regulation of proteins involved in protein biosynthesis is a common response to heavy metals. This result was consistent with the finding by Manara et al.¹⁴ EF-Tu and 50S ribosomal protein L7/L12 are intimately involved in protein biosynthesis, reflecting the need for a vigorous synthesis of protective proteins against Cd. Interestingly, Schroeter et al. found many down-regulated proteins involved in translation (e.g. most aminoacyl-tRNA synthetases and some translation elongation factors) after H₂O₂ treatment in B. licheniformis.³⁹ This difference might be caused by the different oxidative stress between Cd and H₂O₂.

4.2. Effects of Cd²⁺ on the metabolome of B. licheniformis A

In Cd²⁺-treated samples, the concentrations of most identified metabolites were down-regulated (Fig. 6). Amino acids are not only building-blocks in protein synthesis but some are precursors and carbon sources for the TCA cycle (Fig. 7). The consistency between the down-regulation noted in amino acids and proteins including EF-Tu, 50S ribosomal protein L7/L12, trigger factor, peroxiredoxin and superoxide dismutase confirmed the disturbance in protein synthesis. In addition, two kinds of sulfur-containing amino acids may be combined with Cd through thiol groups. Furthermore, a reduction was observed in the levels of glutamate, glycine and S-adenosylhomocysteine, compounds which are necessary for the synthesis of glutathione (GSH). It is a tripeptide, which is synthesized from the amino acids cysteine, glutamate and glycine.³ GSH is the most abundant intracellular thiol in all living cells and is known to be involved in many biological processes, including the synthesis of proteins and DNA, transport, enzyme activity modulation and metabolism as well as defense against reactive oxygen species (ROS).^40,41 Bianucci et al. demonstrated that Cd increases GSH concentrations as cells respond by producing increased antioxidant capacity.⁴²

In our research, the concentration of lactate was increased. Lactate can be converted to pyruvate. And, pyruvate as an end product of glycolysis can repair the cell damage through antioxidation and modulation of energy metabolism.⁴³

In comparison with the metabolites detected in the control group, choline and O-phosphocholine were uniquely detected in the Cd²⁺-treated group. However, the Gram-positive model organism Bacillus subtilis are not known to naturally contain phosphocholine as a cellular membrane phospholipids.⁴⁴ Therefore, choline and O-phosphocholine might be used to resist Cd toxicity.

5. Conclusions

In summary, we have studied the effects of Cd on B. licheniformis A adopting an omic approach. The results showed that Cd could induce the synthesis of proteins that counteract oxidative stress, including enzymes that can scavenge and detoxify ROS induced by Cd. Moreover, B. licheniformis A can up-regulate energy transportation system components to meet the needs of their growth under heavy metal stress and achieve heavy metal tolerance. These indicate that B. licheniformis A is a promising microorganism for the experimental bioremediation of soils contaminated with Cd.

Acknowledgements

This work was financially supported by State Oceanic Administration (201105007).

References

1. J. Shim, J.-M. Lim, P. J. Shea and B.-T. Oh, J. Hazard. Mater., 2014, 272, 129–136.
2. M. Filipič, Mutat. Res., 2012, 733, 69–77.
3. A. Jamers, R. Blust, W. De Coen, J. L. Griffin and O. A. H. Jones, Aquat. Toxicol., 2013, 126, 355–364.
4. IARC, Beryllium, cadmium, mercury, and exposures in the glass manufacturing industry, World Health Organization, Lyon, France, 1993.
5. G. O. Oyetibo, M. O. Ilori, O. S. Obayori and O. O. Amund, Environ. Monit. Assess., 2013, 185, 6809–6818.
6. T. B. Ozer, I. A. Erkaya, A. U. Udoh, D. Y. Duygu, A. Akbulut, G. Bayramoglu and M. Y. Arica, Environ. Sci. Pollut. Res., 2012, 19, 2983–2993.
7. M. Cristani, C. Naccari, A. Nostro, A. Pizzimenti, D. Trombetta and F. Pizzimenti, Environ. Sci. Pollut. Res., 2012, 19, 161–168.
8. I. Poirier, P. Hammann, L. Kuhn and M. Bertrand, Aquat. Toxicol., 2013, 128-129, 215–232.
9. N. K. Kılıç, A. Stensballe, D. E. Otzen and G. Dönmez, Bioresour. Technol., 2010, 101, 2134–2140.
10. R. Andreazza, S. Pieniz, L. Wolf, M.-K. Lee, F. A. O. Camargo and B. C. Okeke, Sci. Total Environ., 2010, 408, 1501–1507.
11. J. Y. Lee, J. G. Yang, D. Zhitnitsky, O. Lewinson and D. C. Rees, Science, 2014, 343, 1133–1136.
12. C. C. Wu, A. Gardarin, P. Catty, F. Guillain and E. Mintz, Biochimie, 2006, 88, 1687–1692.
13. Y. Song, S. Swift, P. J. Swedlund and N. Singhal, Appl. Geochem., 2011, 26, 898–906.
14. A. Manara, G. DalCorso, C. Baliardini, S. Farinati, D. Cecconi and A. Furini, J. Protein Res., 2012, 11, 4169–4179.
15. A. L. C. Mattos, V. R. L. Constantino, R. A. A. de Couto, D. M. L. Pinto, T. M. Kaneko and B. P. Espósito, J. Trace Elem. Med Biol., 2013, 27, 103–108.
16. S. Pandey, P. Ghosh, S. Ghosh, T. De and T. Maiti, J. Microbiol., 2013, 51, 11–17.
17. I. V. N. Rathnayake, M. Megharaj, G. S. R. Krishnamurti, N. S. Bolan and R. Naidu, Chemosphere, 2013, 90, 1195–1200.
18. H. Wu, C. Ji, L. Wei, J. Zhao and H. Lu, J. Proteomics, 2013, 94, 54–67.
19. D. J. Spurgeon, O. A. H. Jones, J.-L. C. M. Dorne, C. Svendsen, S. Swain and S. R. Stürzenbaum, Sci. Total Environ., 2010, 408, 3725–3734.
20. T. Monsinjon and T. Knigge, Proteomics, 2007, 7, 2997–3009.
21. M. Aivaliotis, W. Haase, M. Karas and G. Tsiotis, Proteomics, 2006, 6, 217–232.
22. C. Bar, R. Patil, J. Doshi, M. J. Kulkarni and W. N. Gade, J. Biotechnol., 2007, 128, 444–451.
23. F. Zakeri, M. Sadeghizadeh, M. R. Kardan, H. Shahbani Zahiri, G. Ahmadian, F. Masoumi, H. Sharafi, G. Rigi, H. Vali and K. Akbari Noghabi, J. Proteomics, 2012, 75, 4820–4832.
24. S. Siripornadulsil, L. Thanwisai and W. Siripornadulsil, Can. J. Microbiol., 2014, 60, 121–131.
25. D. Mahle, P. Anderson, N. DelRaso, M. Raymer, A. Neuforth and N. Reo, Metabolomics, 2011, 7, 206–216.
26. G. Brunetti, K. Farrag, P. Soler-Rovira, M. Ferrara, F. Nigro and N. Senesi, Geoderma, 2012, 170, 322–330.
27. C. Dimkpa, T. Weinand and F. Asch, Plant, Cell Environ., 2009, 32, 1682–1694.
28. O. H. Jones, M. Maguire, J. Griffin, Y.-H. Jung, J. Shibato, R. Rakwal, G. Agrawal and N.-S. Jwa, Eur. J. Plant Pathol., 2011, 129, 539–554.
29. J. E. Le Belle, N. G. Harris, S. R. Williams and K. K. Bhakoo, NMR Biomed., 2002, 15, 37–44.
30. C. M. Slupsky and B. D. Sykes, Metabolomics, 2006, 2, 113–123.
31. M. Neerathilingam, D. E. Volk, S. Sarkar, T. M. Alam, M. K. Alam, G. A. S. Ansari and B. A. Luxon, Toxicol. Lett., 2010, 199, 10–16.
32. N. D. Meadow, D. K. Fox and S. Roseman, Annu. Rev. Biochem., 1990, 59, 497–542.
33. M. J. Danson, R. Eisenthal, S. Hall, S. R. Kessell and D. L. Williams, Biochem. J., 1984, 218, 811–818.
34. A. Rince, S. Flahaut and Y. Auffray, Int. J. Food Microbiol., 2000, 55, 87–91.
35. M. Gamerdinger and E. Deuerling, Science, 2014, 344, 590–591.
36. S. Tang, Q. Cai, H. Chibli, V. Allagadda, J. L. Nadeau and G. D. Mayer, Toxicol. Appl. Pharmacol., 2013, 272, 443–452.
37. D. N. Wilson, Nat. Rev. Microbiol., 2014, 12, 35–48.
38. S. Sharma, C. S. Sundaram, P. M. Luthra, Y. Singh, R. Sirdeshmukh and W. N. Gade, J. Biotechnol., 2006, 126, 374–382.
39. R. Schroeter, B. Voigt, B. Jürgen, K. Methling, D.-C. Pöther, H. Schäfer, D. Albrecht, J. Mostertz, U. Mäder, S. Evers, K.-H. Maurer, M. Lalk, T. Mascher, M. Hecker and T. Schweder, Proteomics, 2011, 11, 2851–2866.
40. M. J. Penninckx and M. T. Elskens, Adv. Microb. Physiol., 1993, 34, 239–301.
41. L.-L. Lin, Y.-Y. Chen, M.-C. Chi and A. Merlino, Biochim. Biophys. Acta, Proteins Proteomics, 2014, 1844, 1523–1529.
42. E. Bianucci, A. Fabra and S. Castro, Curr. Microbiol., 2011, 62, 96–100.
43. A. Bignucolo, V. P. Appanna, S. C. Thomas, C. Auger, S. Han, A. Omri and V. D. Appanna, J. Biotechnol., 2013, 167, 309–315.
44. C. Sohlenkamp, I. M. López-Lara and O. Geiger, Prog. Lipid Res., 2003, 42, 115–162.

---

Footnote

† Present address: Fenghua Road, Ningbo City, Zhejiang Province 315211, P. R. China.

---