Introduction

Endocrine disrupting compounds (EDCs) are one of the most predominant environmental contaminants, which widely exist in different environments and have significant adverse effects on the reproductive system of animals and humans1,2. Natural estrogens, synthesized estrogens and estrogenic chemicals are the major components of EDCs; and 17β-estradiol is regarded as the largest contaminant in natural estrogens because of its serious adverse effects3. Although their concentrations in environment are at trace levels, environmental estrogens are difficult to be removed. They can be assimilated by humans through food intake and then threaten human health4. Sometimes, several estrogen metabolites and estrogens in non-active states discharged into environment can be converted back to the active states, adding the removal difficulty5,6.

Biodegradation using microorganisms have been considered as one efficient strategy to remove EDCs and a series of strains with estrogen degradation capabilities have been isolated from activated sludge, soil, ocean and other ecosystem environments, such as Novosphingobium, Sphingomanas, Acinetobacter, Rhodococcus, Nocardioides and Pseudomonas7,8,9,10,11,12,13,14,15,16,17. Most strains can degrade natural estrogens (E2, E1); some can utilize synthetic estrogen like EE2 or estrogenic chemicals like polycyclic aromatic hydrocarbons (PAH)15,16,17,18. In Rhodococcus sp. and Sphingomonas sp., E2 was first converted into into E1, then hydroxylate it to form 4-hydroxyestrone (4-OH-E1), with the subsequent cleavage of the 4-OH-E1 ring into the final non-estrogenic products11. Pseudomonas spp., especially P. putida strains are widely studied because of their great degradation capabilities to estrogenic chemicals. P. putida OUS82, a model organism for naphthalene remedy, can utilize phenanthrene and PAH efficiently18. Other naphthalene degradation strains, P. putida ND6, P. putida G7, P. putida KT2440 and P. putida AK5 have also been characterized16,19,20,21. P. putida KA4 and KA5 were confirmed with high degradation abilities to another EEs member, bisphenol A (BPA)22. And some Pseudomonas strains can biotransform diethylstilbestrol and 4-nonylphenol23,24.

Although a number of estrogen biodegradation strains were isolated, their entire global responses and metabolic networks to chemicals remain unclear due to limitations in technologies and methods. Recently, comparative proteomics analysis was considered a potential tool to reveal the substantial changes under different conditions. It has been used in Pseudomonas to study the whole-cell proteome composition, and unveil biochemical pathways and key effectors assisting stains to adapt special environments25,26,27,28. For example, the divergent metabolism of phenol and succinate in P. putida KT2440 was revealed by comparative proteomics29. It was found that P. putida KT2440 would induce a global response to aromatic hydrocarbon sources (phenol or benzoate) by up- or down-regulating series of enzymes for substrate uptake, transport, degradation, and product export28,29. Using iTRAQ technology, the proteome of P. fluorescens Pf5 under iron limitation conditions was studied, and some proteins involved in receptor systems, inner-membrane transporters and biosynthesis of secondary metabolites were found to be significantly altered30. Sulfur-34S stable isotope labeling of amino acids for quantification (SULAQ34) was used to investigate the proteomic changes related to naphthalene degradation in P. fluorescens ATCC 17483 and uncovered a specific oxidative-stress-like response31. These findings advance our knowledge on microbial adaptation mechanisms and fasten the biochemical pathway identification. However, proteomic studies of microorganisms adapted to estrogenic environments are rare. Quantitative proteomics was used to identify possible metabolic pathways involved in the transformation of E2 and E1 in Stenotrophomonas maltophilia ZL132. Results showed enzymes involved in certain catabolic and anabolic pathways were highly expressed, especially the lipid biosynthesis proteins32. And the proteomic analysis of Sphingomonas sp. TTNP3 to BPA and nonylphenol was performed to reveal the degradation pathways33. However, the proteomic analysis of Pseudomonas strains to estrogen stress has not been reported.

P. putida SJTE-1 isolated from sludge was able to degrade multiple estrogens efficiently, including 17β-estradiol, estrone, and other estrogenic chemicals, and bio-transform them into non-estrogenic products14. Although its genome sequence was obtained and annotated, its global response to estrogens and its metabolic mechanism are still poorly understood. In this work, we applied an iTRAQ-based quantitative proteomics technology to characterize the proteomic profiles of P. putida SJTE-1 in 17β-estradiol environment, compared to those in glucose condition. 78 proteins were found to be significantly dys-regulated, involved in the processes of stress responses, uptake and transport, energy metabolism, translation and nucleotide metabolism, and carbohydrate metabolism.

Materials and Methods

Strains, culture media, and chemicals

Pseudomonas putida SJTE-1 used in this study was isolated from the sludge of a sewage treatment plant in Shanghai, China14. Luria-Bertani (LB) medium (tryptone 10.0 g, yeast extract 5.0 g, NaCl 8.0 g/L) and minimal medium (MM) (K2HPO4 381.5 mg, KH2PO4 50.0 mg, (NH4)2HPO4 82.5 mg, KNO3 126.25 mg, Na2SO4 20.0 mg, CaCl2 2.0 mg, FeCl3 0.2 mg, MgCl2 2.0 mg/L) were used in this study. 17β-estradiol was dissolved in anhydrous ethanol (>99.7%) to a concentration of 10 mg/mL and its working concentration was 30.0 mg/L. Glucose was used as the reference carbon source at a concentration of 2.0%. 17β-estradiol (>98%), DTT and IAA were purchased from Sigma-Aldrich (St. Louis, MO, USA). Trypsin was purchased from Promega Corporation (WI, USA). Other reagents were the products of China National Medicines Co., Ltd. (Beijing, China).

Bacteria cultivation and protein preparation

A single colony of P. putida SJTE-1 was inoculated and cultured in LB medium overnight at 30 °C with shaking at 200 rpm, followed by centrifugation at 4 °C, 5, 000 rpm for 5 min. Cells were washed three times with 1× PBS (NaCl 8.0 g, KCl 0.2 g, Na2HPO4 1.44 g, KH2PO4 0.24 g/L, pH 7.4) and the re-suspended pellets were inoculated to two flasks containing 100 mL minimal medium supplemented with 17β-estradiol or glucose to an initial OD600 = 0.1. The growth conditions were closely matched for both modes, including vessel and aeration. Cells were cultured to and harvested in the late exponential phase (OD600 = 0.8~1.0).

Cells were re-suspended into 5 mL pre-cooled lysis buffer (8 M Urea, 4% CHAPS, 40 mM Tris-HCl, pH 7.4, 1 mM PMSF, 2 mM EDTA, 0.5 mM EGTA, pH 7.4) and were sonicated on ice until the cells were completely lysed. The cell debris was removed by centrifugation at 4 °C, 12,000 rpm for 30 min. The supernatants were precipitated by adding a 5-fold volume of chilled solution (ethanol:acetone:acetic acid = 50:50:0.1, v/v/v), followed by centrifugation at 4 °C, 12,000 rpm for 60 min. The precipitate was washed three times with acetone and subjected to vacuum centrifugation (Thermo Fisher Scientific Inc., DE, USA). The dry proteins were then re-dissolved with 1 mL denaturing buffer (50 mM NH4HCO3, 6 M guanidine hydrochloride), quantified using the Bradford method and all proteins were adjusted into the same concentration34.

Protein digestion and iTRAQ labeling

Each sample of 200 μg total protein in 200 μL denaturing buffer was reduced with 2 μL 1 M DTT at 60 °C for 1 h, and then the cysteine residues were blocked by adding 10 μL 1 M IAA for 40 min at room temperate under dark conditions. The reduced and alkylated protein mixtures were subjected to the FASP protocol with spin ultra-filtration units containing a nominal molecular weight cutoff of 10,000 Da (Sartorius, Gottingen, Germany), and centrifuged at 12,000 rpm for 20 min. The bottom solution was then discarded. After washing with 100 μL dissolution buffer three times, the protein digestions were conducted by incubating the proteins and the Sequencing Grade Modified Trypsin (Promega Corporation, WI, USA) in a 1:50 ratio (trypsin-to-protein mass) at 37 °C overnight. After digestion, the liberated peptides were collected by centrifugation at 12,000 rpm for 20 min and the filtration units were washed with 50 μL of UA buffer. The resultant peptide mixture samples were labeled using an 8-Plex iTRAQ Reagent Kit from Applied Biosystems (Thermo Fisher Scientific Inc., DE, USA) as follows: E2_1: 113; E2_2: 114; Glucose_1: 117; Glucose_2: 118. The labeled peptide mixtures (E2_1: 113 and E2_2: 114, Glucose_1: 117 and Glucose_2: 118) were then pooled together and dried by SpeedVac.

High pH reverse phase liquid chromatography

The peptide mixture was re-dissolved with buffer A (20 mM ammonium formate, pH 10.0), and fractionated using a Survey HPLC system (Thermo Fisher Scientific Inc., DE, USA) equipped with a reverse phase column (Durashell-C18 column, 2.1 mm × 250 mm, 5 μm, 100 Å, Agela Technologies, Wilmington, DE, USA). The peptides were eluted with gradient 5~30% buffer B (20 mM ammonium formate in 80% ACN, pH 10.0) in 25 min, 15~38% buffer B in 15 min, 90% buffer B hold for 10 min, with a constant flow rate of 0.8 mL/min. The absorbance at 214 nm was monitored, and a total of twenty four fractions were collected and dried in the vacuum concentrator.

Nano LC-MS/MS

The protein fractions were analyzed using an LC system (Eksigent 1D) coupled with an ESI-Q-TOF mass spectrometer (Triple TOF 4600, SCIEX Pte. Ltd. Framingham, MA, USA). Each peptide sample was re-dissolved in 2% acetonitrile with 0.1% formic acid, and then loaded onto a Peptide trap column (0.1 mm × 2 cm, 5 μm, Dionex, Thermo Fisher Scientific Inc., DE, USA) with the auto-sampler of the LC system. To desalt and concentrate the sample, the trap column was washed with 2% acetonitrile with 0.1% formic acid for 10 min at a flow rate of 5 μL/min. The trapped peptides were released and separated with a C18 capillary column (75 μm × 150 mm, 3 μm, Dionex, Thermo Fisher Scientific Inc., DE, USA). The peptides were subsequently eluted with mobile phase B (98% ACN with 0.1% formic acid) using a gradient system, with solvent A (99.9% water with 0.1% formic acid) and solvent B (98% ACN with 0.1% formic acid) from 5% to 45% B (5–100 min), with a constant flow rate of 300 nL/min. The Triple TOF 4600 mass spectrometer was operated in information-dependent data acquisition mode to switch automatically between MS and MS/MS acquisition. The electrospray voltage was set at 2.5 KV; MS spectra were acquired across the mass range of 350–1250 m/z in high sensitivity mode with rolling collision energy. The 25 most intense precursors were selected for fragmentation per cycle with a dynamic exclusion time of 25 s.

Database Search

Tandem mass spectra were extracted, and charge states de-convoluted and de-isotoped by the MS Data Converter software from SCIEX Pte. Ltd. (Framingham, MA, USA). The peak list was directly generated from wiff data using a centroid algorithm with peak width set as 0.1 m/z and intensity above 100. No peak smoothing or filter process was applied. After the charge states were calculated, the de-isotoped peak list was exported as an.mgf file for further database searching. Mascot (Matrix Science, London, UK; version 2.3.02) was set up to search the NCBI_SJTE-1 database (4,698 entries), which was established according to gene homology between P. putida F1 and P. putida SJTE-1, assuming digestion with the enzyme trypsin. Mascot was searched with a fragment ion mass tolerance of 0.1 Da and a parent ion tolerance of 25 ppm. Oxidation of methionine, iTRAQ 8-Plex of tyrosine and carbamylation of lysines were specified as variable modifications, iTRAQ (N-terminal, +304 Da), iTRAQ (Lys, +304 Da) and MMTS (Cys, +46 Da) were specified as fixed modifications.

Quantitative data analysis

Scaffold Q+ (version Scaffold_4.3.2, Proteome Software Inc., Portland, OR) was used to validate MS/MS based isobaric tag peptide and protein identifications. The setting was as followed: Model type was Intensity based normalization, the Calculation type was median and the Quant uniqueness model was unique peptides. The Reference type was average protein reference, and the Normalization between samples was on. Protein identifications were accepted if the peptides probabilities were greater than 92.0%, an FDR less than 1.0% by the Scaffold Local FDR algorithm and at least two identified peptides. Protein quantifications were accepted if they could be established at greater than 99.0% probability to achieve an FDR less than 1.0% and contained at least 1 identified peptide. Proteins that contained similar peptides and could not be differentiated on the basis of MS/MS analysis alone were grouped to satisfy the principles of parsimony. Proteins sharing significant peptide evidence were grouped into clusters. Acquired intensities in the experiment were globally normalized across all acquisition runs. Individual quantitative samples were normalized within each acquisition run. Intensities of peptide identification were normalized within the assigned protein. The reference channels were normalized to produce a 1:1 fold change. Differentially expressed proteins were determined using Mann Whitney Test analysis (p-value). The final list of protein ratios was an average of the protein ratios. Protein-protein interaction networks were built using the Search Tool for the Retrieval of Interacting Genes/Proteins (STRING, version 10.0) with a medium confidence level (0.4) and all available prediction methods (http://string-db.org/). COG classifications are obtained according to P. putida, and P. putida SJTE-1 project in IMG database (https://img.jgi.doe.gov/cgibin/er/main.cgi, Project ID: Gi23653). A wide range of heterogeneous annotation content, such as GO terms, KEGG pathways, into term or gene classes were organized and condensed via DAVID (http://david.abcc.ncifcrf.gov/summary.jsp)35,36.

RNA extraction and transcriptional level analysis (RT-qPCR)

P. putida SJTE-1 was cultured in minimal medium with E2 or glucose to the late exponential phase as described above, and RNA was extracted using an RNeasy Mini kit (QIAGEN, CA, USA) following the manufacturer’s instructions. The total RNA samples were pre-treated with DNase I to exclude the contamination of genomic and plasmid DNA. The quality and integrity of extracted RNA were detected by 2% agarose electrophoresis, and the RNA concentration was estimated with a NanoDrop UV spectrometer (Thermo Scientific, DE, USA).

20 μL of RT reactions were preformed in mixture containing contained 1 μL (1 μg/μL) of total RNAs (all the RNAs of different samples were adjusted to 1 μg/μL), 1 μL of Gene Specific Primer (2 μM), 10 μL of 2× PrimeScript Reverse Transcriptase Master Mix (TaKaRa, Dalian, China), and 8 μL of RNase-free ddH2O. The mixture was incubated at 37 °C for 15 min, and then incubation was continued at 50 °C for 5 min. The reaction was inactivated by heating at 85 °C for 5 min. The cDNA quantification was performed NanoDrop UV spectrometer (Thermo Scientific, DE, USA).

Quantitative real-time PCR was performed using the cDNAs of different samples (adjusted to the same concentration), the gene-specific primers (Table S1) and the IQSYBR Green Super-mix (TaKaRa, Dalian, China) in IQTM 5 Multicolor Real-time PCR Detection System (Bio-Rad Laboratories, Inc., CA, USA). The reactions were incubated at 95 °C for 5 min, followed by 40 cycles of 95 °C for 10 s, and 68 °C for 30 s. At least three independent experiments were conducted for each RNA sample. The 16S rRNA gene was used as the reference gene. The relative fold change in mRNA quantity was calculated using the DDCt method. SPSS 21.0 software was used for statistical analysis and P < 0.05 was considered statistically significant change.

Results

Detection and relative quantification of proteins in P. putida SJTE-1 with 17β-estradiol or glucose as carbon sources

To profile the expression of proteins induced by 17β-estradiol in P. putida SJTE-1, quantitative proteomic analysis based on an iTRAQ labeling method was executed. To assure biological reproducibility, duplicate protein samples in minimal medium with 17β-estradiol and minimal medium with glucose in two independent experiments were prepared. iTRAQ labels (113 and 114) were separately used to label the samples from 17β-estradiol group samples, and the glucose group samples were separately labeled by 117 and 118.

As to the 17β-estradiol group, a total of 3,069 unique peptides and 1,000 proteins were identified in two biological replicates (Tables S2, S3). 787 proteins were both detected in two experiments (Fig. 1). The identified proteins accounted for about 21% of the 4,698 predicted proteins encoded by P. putida SJTE-1. The proteomics data has been deposited into iProx database (www.iprox.org), with the subject ID IPX00080001. Mann Whitney Test analysis was performed to determine the biological reproducibility. To determine the cut-off for up- or down-regulation, the Coefficient of Variation (CV) for all proteins in each quantitative sample was determined. The average %CV was around 22%. Based on this, the regulation threshold was set at 1.5 fold. Ratios of ≥1.5 or ≤0.67 and a P-value of less than 0.05 were considered as significant change of protein level. Total of 78 proteins were considered as the dys-expressed proteins; 45 and 33 proteins were significantly up- and down- regulated (Table 1, Fig. 2).

Figure 1: Proteins identified using iTRAQ and LC-MS methods.
figure 1

The black circles and grey circles represent the number of proteins detected from the first and second MS experiment. The overlapped number of identified proteins was 787.

Full size image
Table 1 The differentially-expressed proteins of P. putida SJTE-1 in 17β-estradiol environment.
Full size table
Figure 2: Distribution of the differentially-expressed proteins.
figure 2

The differentially expressed proteins with a fold change of ≥1.5-fold or ≤0.67-fold and p ≤ 0.05 are shown in the Volcano plot. The Y-axis represents p-value in the form of −log 10; the greater numerical value means the smaller p-value and the higher credibility. X-axis represents fold change of protein expression in the form of log 2 ratio; the negative value means the positive change and the positive value means negative change. Therefore, the green dots represent the up-regulated proteins, and the red dots are on behalf of down-regulated proteins.

Full size image

Functional categories of the differentially expressed proteins

The identified up- and down- regulated proteins were noted and were classified into six categories by their functions and roles in biological processes on basis of COG functional classification (Fig. 3). Proteins involved in process of stress response, chemotaxis and motility, energy metabolism, carbonate metabolism, transport systems and cell division responded to the 17β-estradiol condition. Most differentially expressed proteins were clustered into carbohydrate metabolism process, implying strain’s significant response to the two different carbon sources. Also proteins in the translation and nucleotide metabolic clusters and proteins involved in membrane transport and ABC transport systems also accounted for large portions of the dys-expressed proteins, indicating 17β-estradiol would drive an active cellular transportation and metabolic support for the uptake and utilization of this difficult-to-be-used carbon source. In addition, the expression changes of each protein were analyzed and displayed in heat-maps (Fig. 4). Although some errors may exist in the individual biological replication or technological replication, these maps could still clearly reflect the credible variation trend. For example, the TonB-dependent receptor involved in cell transport and the cytochrome C (class I) related to electron transfer were significantly over-expressed upon 17β-estradiol treatment, while the ATP-dependent protease ATPase subunit HslU was significantly down-regulated in this condition, implying the elevated electron transport and energy requirements of cells in the 17β-estradiol biotransformation process. Gene ontology analysis of the differentially expressed proteins showed that 52 proteins were significantly enriched in the translation, oxidation reduction, the generation of precursor metabolites and energy and the Acetyl-CoA metabolic and protein metabolic processes (Table 2). Further enrichment degrees of the dys-expressed proteins in KEGG pathways analyzed by DAVID showed that 46 covered proteins in pathways of ribosome, butanoate metabolism, pentose phosphate pathway, amino acid degradation and ABC transporters were with significant change in estrogen environment (Table 3). This implied that proteins in the several pathways of carbohydrate metabolism process play an important role in the utilization of 17β-estradiol, consistent with previous work34. Besides, the protein-protein interaction networks of differentially expressed proteins were built and analyzed using STRING. Proteins were grouped into three large units: unit associated with or involved in translation (top), oxidation reduction and carbon metabolism (left bottom), and Acetyl-CoA and fatty acid metabolism (right bottom) (Fig. 5). It demonstrated that in estrogenic environment, bacterial cells started an active cellular metabolic processes to participate and support the transportation and the utilization of 17β-estradiol. This revealed the relations hidden behind the changes of protein levels by means of a computer-assisted analytical approach, and shed a light on the study of bacterial global effect under an estrogen environment.

Figure 3: Functional categories of the differentially expressed proteins in P. putida SJTE-1 with 17β-estradiol as its sole carbon source.
figure 3

The protein annotation and classification of the differentially expressed proteins were organized and condensed via DAVID. The percentage of each category represents the proteins’ ratio in this category in all differentially-expressed proteins.

Full size image
Figure 4: The heatmaps of each differentially expressed protein.
figure 4

A heat map of the log 2 relative abundance of proteins under estradiol environment compared to the glucose control condition was created using Genesis V1.7 with the iTRAQ-derived quantitative data. Proteins were grouped according to their known or putative role in metabolic pathways or cellular processes.

Full size image
Table 2 The Gene ontology analysis of the differentially expressed proteins.
Full size table
Table 3 List of the enrichment degrees of dys-expressed protein in metabolic pathways in KEGG.
Full size table
Figure 5: The protein and protein interaction networks of all differentially expressed proteins.
figure 5

The protein-protein interaction networks were built using STRING 10.0 with a medium confidence level (0.4) and all available prediction methods.

Full size image

Quantitative RT-PCR validation of the genes transcription levels

To validate the results from proteomic analysis, we selected some genes with significant differential expression profiles to perform RT-qPCR validation. In the meantime, several Short-chain dehydrogenase/reductase (SDR) genes with little variance in the iTRAQ analysis were also chosen as targets. As shown in Table 4, the expression change trends of most genes were consistent in the two experiments, despite minor differences in the fold change levels. Some genes’ transcription change and expression variation were different, probably related to the gene abundance. It indicates that the proteomics results are mostly consistent with those of RT-qPCR, which could reflect the changes occurring in a 17β-estradiol environment.

Table 4 Correlation of the transcriptional and the expressional changes by quantitative PCR and iTRAQ methods.
Full size table

Discussion

17β-estradiol, known as an uncommon carbon source, has potential stress for bacterial growth and metabolism compared with glucose; cells may induce specific proteins to bio-transform it and adapt this environment. In this study, the proteomic data suggested that proteins involved in the metabolic processes including stress response, chemotaxis and motility, uptake and transportation, electron transfer and energy metabolism and carbonate metabolism had significant changes in the utilization of 17β-estradiol by P. putida SJTE-1.

Stress response, chemotaxis and motility

As a recalcitrant carbon source, 17β-estradiol brings potential stress to bacterial growth in many aspects, although it can be eventually mineralized as a carbon and energy source. These growth stresses can be defined as an overload of some metabolic pathways, regional over-accumulation of internal toxicants, and regeneration deficiency of some cycling factors for electron transfer or co-factor supply. According to our data, seven proteins involved in stress response were found to be significantly differentially expressed, and most of them are responsible for the protection of protein activities (Table 1, Fig. 4). LPS-assembly protein LptD is an outer membrane transport protein involved in the lipopolysaccharide transport (Lpt) system and plays an essential role in impermeable outer membrane (OM) biogenesis. It can form a complex with LptE to transport lipopolysaccharide like lipid A, which is responsible for permeability of the OM to the outer cell surface37,38. The OM can protect cells from environmentally toxic molecules and is important for cell survival under stress conditions. At the same time, the OM may have functions in conjunction with multiple efflux pumps to decrease assimilation of toxic substances39. When 17β-estradiol was used as carbon source, the over-expressed LptD protein could promote the synthesis of OM to support a protection for cells in this uncommon cultivation condition and decrease the toxic effect caused by 17β-estradiol stress. Another damaging effect of environmental stress to cells is the generation of a hyperosmotic and oxidative cellular micro-environment, which results in the aggregation of a large number of denatured proteins and influences cellular metabolism. Ribonuclease PH belongs to heat shock dnaK gene cluster extended subsystem, which participates actively in the response to hyper-osmotic stress and heat shock reactions, by preventing the aggregation of stress-denatured proteins and causing disaggregation of denatured proteins. It is noteworthy that many proteins in this system are predicted to be involved in various tRNA or rRNA modifications. Since many tRNA modifications are believed to improve reading frame maintenance, it is tempting to speculate that the role of these subsystem proteins is protecting ribosomal function (e.g. accuracy of translation) during heat shock and other stresses40. Ribonuclease PH is a 3′-5′ exoribonuclease and nucleotidyltransferase involved in tRNA processing. Up-regulation of Ribonuclease PH may be essential for the protection of ribosomal function under 17β-estradiol stress by correcting the mistakes in tRNA procedure. In addition, a flagellar motor switch protein FliG playing a role in cell motility was also found to have 1.67-fold up-regulated expression in 17β-estradiol environment. When under starvation or stress environments, bacterial cells will take advantage of chemotaxis and motility systems to reach to carbon attractants in the environment and support their growth. Up-regulation of FliG can enhance the cells’ motile ability and may contribute to the cells’ capture of 17β-estradiol.

Uptake and transportation

Microorganisms can change their uptake and transport systems to adapt to different nutrient conditions. Under iron limitation conditions, the Ton-B-dependent receptor system and some inner-member transporters of P. fluorescens Pf-5 were significantly up-regulated41. Proteomic characteristics of P. putida KT2440 in response to benzoate showed that 21 proteins involved in ABC transporters were up-regulated, including periplasmic binding proteins of amino acid ABC transporters and extracellular ligand-binding receptors42. Assimilation of carbon sources from the environment is the basic activity for bacterial growth, and the ATP-binding cassette (ABC) transporters are the most important routes for bacteria to acquire these carbon sources. 17β-estradiol is soluble in ethanol, but not in water, hampering its uptake by microorganisms. Some specific receptors and special transporters may be needed to assist its uptake and utilization. Based on the proteomic data, twelve proteins related to uptake and transport systems were found with significant changes in their expression levels and six of them were found in the ABC transporters term of KEGG pathway.

One TonB-dependent receptor was over-expressed 3.3-fold in 17β-estradiol environment. TonB protein is an energy transducer to receptors in outer membrane using cytoplasmic membrane proton motive force (PMF), to facilitate the active transport of substrates through the outer membrane43,44. That means an entire TonB-dependent transport system must include a special outer membrane receptor; and there are at least seven outer membrane receptors corresponding to TonB45. Although the detailed information of this up-regulated TonB-dependent receptor is unclear, we hypothesize it may be involved in the transportation of 17β-estradiol or its intermediates across the outer membrane.

In addition, three amino acid/amide ABC transporter substrate-binding proteins including both the PAAT (polar amino acid transporter) and HAAT (hydrophobic amino acid transporter) families displayed significant up-regulated expression ranging from 2.5 to 7.6 folds. Substrate-binding protein (SBP)-dependent transporters are extensively present in bacteria, which are involved in diverse processes, such as nutrient uptake, quorum sensing and multidrug resistance, which was considered as the primary choices for high-affinity uptake of nutrients46,47. As to the characteristics of 17β-estradiol, SBP-dependent transporters may be needed for its uptake and transportation. These three transporters are probably the members of the specific transportation systems for 17β-estradiol and other estrogenic chemicals. Furthermore, the poly (Hydroxyalkanoate) granule-associated protein PhaF was found with an approximate 1.97-fold increase of its expression level when 17β-estradiol was used as carbon source. This protein has been reported to direct granules to the center of the cell without specificities during the cell division process48. It may act as the director of 17β-estradiol for its equal distribution during cell division and cell growth processes; and its up-regulation can probably facilitate the efficient utilization and bio-transformation of this chemical in cells.

On the other hand, in the 17β-estradiol-supplied environment, proteins relative to the uptake and transportation of glucose were also influenced. The down-regulation of glucose ABC transporter ATP-binding protein just demonstrates that it is not required in glucose-absent environments. As estradiol is not soluble in water, it cannot be transported through the water-filled channels. Porin proteins can form trans-membrane water-filled channels to allow the diffusion of substrates in outer membrane of gram-negative bacteria; and OprB is a glucose-selective porin which has selectivity for glucose and xylose, produced by P. aeruginosa and P. putida49. With 17β-estradiol supplied, the role of porin OprB was confirmed with a 0.47-fold reduction, consistent with its known functions.

Electron transfer and energy metabolism

Enough energy and oxygen supply are the key points to support normal cell growth and reproduction processes. In stress environments, cells initiate protection strategies like biofilm formation to ensure sufficient energy supply and maintain basic cell metabolism. Pseudomonas strains can generate large amounts of biofilms, especially in the late growth phase to adapt the nutrient starvation or other stress. Biofilm dispersion is an energy-requiring process, and the proton motive force is essential for this process50. Therefore, when 17β-estradiol was used as sole carbon source, more efficient energy supply and electron transfer systems were required to support its metabolism and the cell growth of P. putida SJTE-1.

The proteomics data showed that there were several proteins in P. putida SJTE-1 involved in electron transfer and energy metabolism which were significantly up-regulated in the 17β-estradiol environment (Table 1, Fig. 4). Compared to those of the glucose culture condition, the coenzyme pyrrolo-quinoline quinone (PQQ) synthesis protein E and the cytochrome C (class I) were over-expressed 4.8-fold and 1.9-fold, respectively. Cytochrome C (cytC) is a well-known electron-transfer protein and a member of the respiratory chain, functioning in different redox processes51. PQQ, acting as a prosthetic group, is non-covalently attached to dehydrogenase to form quinoproteins. The quinoproteins usually catalyze the first step of the oxidation reaction in the bacterial periplasm, contributing to the formation of a proton motive force and the formation of ATP52. There is a special ‘periplasmic oxidation system’ in gram-negative bacteria, mainly initiated by the quinoprotein dehydrogenases coupled with the respiratory chains; it is crucial for the generation of membrane potential without making toxic products to enhance cells’ growth advantage in various environments53. In the meantime, the expression of ATP synthase subunit alpha and beta increased 1.57-fold and 1.48-fold, which could promote the oxidative phosphorylation pathway and synthesized more ATP to support the large energy requirement in 17β-estradiol environment. Overall, the over-expression of all these proteins could generate high-efficient electron transfer and respiratory chains, give more energy supply to overcome the growth stress and potential toxicity, and ensure normal cellular metabolism.

Carbonate metabolism

There were twenty enzymes relating to different metabolic pathways which were found to be differentially expressed in 17β-estradiol utilization process (Table 1, Fig. 4). Among them, four proteins belonged to TCA pathway and four proteins of pentose phosphate pathways were all down-regulated. Pyruvate dehydrogenase catalyzes the formation of acetyl-CoA to accelerate the TCA pathway; isocitrate dehydrogenase functions as the regulatory enzyme in TCA cycle, and the glucose-6-phosphate-dehydrogenase and 6-phosphogluconate dehydratase are the key enzymes in pentose phosphate pathway. Under glucose-starvation conditions, the two pathways may be weakened.

It is noteworthy that acetyl-CoA hydrolase and acetyl-CoA acetyltransferase, enzymes in the pyruvate metabolism pathway are up-regulated 1.94- and 2.24-fold, respectively. In Comamonas testosterone, the B, C, D-rings of steroids were speculated to undergo β-oxidation after meta-cleavage with the support of acetyl-CoA54. The major product of β-oxidation is acetyl-CoA. The similar molecular structure of 17β-estradiol to testosterone suggests that the metabolic pathway of 17β-estradiol may also have similarity to that of testosterone, and therefore the production of acetyl-CoA may be critical for 17β-estradiol utilization. Acetyl-CoA acetyltransferase supports the acetyl transfer and hence the –S-CoA could be added to the B, C, and D-rings. It may be coupled with acetyl-CoA acetyltransferase to build an equilibrium state between consumption and synthesis of acetyl-CoA.

Short-chain dehydrogenase/reductases (SDR) were considered important for 17β-estradiol metabolism. One SDR protein was found down-regulated a little slightly to 0.58-fold. In fact, we also detected the expression of four other SDRs in the two conditions, and no significant change was noted. Besides the protein instability and technological limitations, we suppose that there are two possibilities. One is that a specific SDR for 17β-estradiol metabolism may not exist, as a previous report hypothesized that bacteria may choose a co-metabolic degradation mode to use the existing enzymes to degrade steroidal hormones without generating novel specific proteins7. It is an adaption strategy to reduce the metabolic and energetic consumption and enhance cell survival in restricted environments. However, it’s still a hypothesis and more studies are needed to verify this. Another possibility is that the obvious expression changes of SDR genes may occur at the very early phase of cell growth as a fast response to estradiol. Because we used the cells at the logarithmic to exponential growth phase (0.8-1.0) to perform the proteomic analysis, cell densities and cell adaptability are relatively improved; therefore, it was unnecessary for cells to change the expression of SDR genes.

Transcription, translation and macromolecule metabolism

According to the data, fourteen ribosomal proteins and other proteins involved in translation and nucleotide metabolism displayed significant changes, such as 50S ribosomal protein L10, 30S ribosomal protein S20, nucleoside diphosphate kinase, which belonged to the ribosome pathway. This was expected, as cells have to make relative changes in transcription and translation to generate differential gene expression to fit the requirements of different environments. In many proteomic studies, changes in ribosomal protein expression are always a big part. Strain SJTE-1 modifies the expression levels of different ribosomal proteins can help cells adapt to the restricted estradiol conditions, reduce the toxic effects and guarantee cellular metabolism. Besides, proteins involved in fatty acid metabolism, amino acid metabolism, cell division and cell envelope biosynthesis and some with unclear functions were also dysregulated to accommodate 17β-estradiol environment. These may work with the transcription and translation process and assist the degradation of estradiol.

In conclusion, this work compared the protein expression levels of P. putida SJTE-1 in 17β-estradiol and glucose environments using iTRAQ labeling and LC-MS/MS technology, and tried to identify the key metabolic pathways or proteins involved in the microbial estrogen degradation process. 78 proteins were identified with significant changes, and are mainly involved in the process of stress response, energy metabolism, transportation systems, chemotaxis and cell motility, and carbon metabolism. The up-regulation of proteins involved in electron transfer, energy generation and transport systems implies they may be crucial for the efficient uptake, translocation and utilization of 17β-estradiol. Additionally, the over-expression of proteins involved in carbon metabolism indicated cells launch the members of steroid related pathways for the valid biotransformation of 17β-estradiol. On the other hand, proteins involved in glucose capture and metabolism were mostly down-regulated. P. putida SJTE-1 has great degradation capability to estrogenic chemicals, and it contains several putative estrogen degradation genes14. Although some estrogen degradation strains have been found, up to now, only several genome sequences of the estrogen-degrading strains have been obtained, and the comparative proteomics analysis of bacteria to estrogenic chemicals has not been reported. P. putida has a great advantage in bioremediation and several strains have been confirmed with the utilization capabilities to estrogenic chemicals. This work can give a light on the biodegradation mechanism of 17β-estradiol in Pseudomonas and promote further environment application.

Additional Information

How to cite this article: Xu, J. et al. iTRAQ-based quantitative proteomic analysis of the global response to 17β-estradiol in estrogen-degradation strain Pseudomonas putida SJTE-1. Sci. Rep. 7, 41682; doi: 10.1038/srep41682 (2017).

Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.