Comparative transcriptomic and proteomic analyses reveal upregulated expression of virulence and iron transport factors of Aeromonas hydrophila under iron limitation

Comparative transcriptomic and proteomic analyses reveal upregulated expression of virulence and... Background: Iron plays important roles in the growth, reproduction and pathogenicity of Aeromonas hydrophila.In this study, we detected and compared the mRNA and protein expression profiles of A. hydrophila under normal and iron restricted medium with 200 μM 2,2-Dipyridyl using RNA Sequencing (RNA-seq) and isobaric tags for relative and absolute quantification (iTRAQ) analyses. Results: There were 1204 genes (601 up- and 603 down-regulated) and 236 proteins (90 up- and 146 down-regulated) shown to be differentially expressed, and 167 genes and proteins that showed consistent expression. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses revealed that the differentially expressed genes and proteins were mainly involved in iron ion transport, protein activity, energy metabolism and virulence processes. Further validation of the RNA-seq and iTRAQ results by quantitative real-time PCR (qPCR) revealed that 18 of the 20 selected genes were consistently expressed. The iron-ion absorption and concentration of A. hydrophila under iron-limited conditions were enhanced, and most virulence factors (protease activity, hemolytic activity, lipase activity, and swimming ability) were also increased. Artificial A. hydrophila infection caused higher mortality in cyprinid Megalobrama amblycephala under iron-limited conditions. Conclusion: Understanding the responses of pathogenic Aeromonas hydrophila within the hostile environment of the fish host, devoid of free iron, is important to reveal bacterial infection and pathogenesis. This study further confirmed the previous finding that iron-limitation efficiently enhanced the virulence of A. hydrophila using multi-omics analyses. We identified differentially expressed genes and proteins, related to enterobactin synthesis and virulence establishment, that play important roles in addressing iron scarcity. Keywords: Transcriptomic, Proteomic, Virulence, Iron, Aeromonas hydrophila Background and proteases, and it has the capacity to form biofilms and Aeromonas hydrophila is an opportunistic pathogenic alter metabolic pathways and gene expression under vari- bacterium that is ubiquitous in aquatic environments and ous host environments [5, 6]. Its virulence expression is causes serious infections worldwide in cultured fishes, am- closely related to the environment in which the bacteria live phibians, reptiles, and even mammals [1–4]. The pathogen- (in vivo and in vitro), nutrients, and so on [7]. For example, esis of A. hydrophila is multifactorial, causing disease with the nutrient iron deficiency in the host environment has virulence factors, such as adhesins, cytotoxins, hemolysins, been thoroughly documented as having a pronounced ef- fect on the virulence of pathogens [8]. Iron is an indispensable element of most living cells that is involved in many cellular functions, including electron * Correspondence: xiej@ffrc.cn; xup@ffrc.cn † transportation and oxygen transportation. The quantity Tao Teng and Bingwen Xi contributed equally to this work. of iron has a great impact on biological processes, for Wuxi Fisheries College, Nanjing Agricultural University, Wuxi 214081, China Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Teng et al. BMC Microbiology (2018) 18:52 Page 2 of 17 instance, iron overload will result in iron toxicity to cellu- 5 ml of normal TSB and incubated (28 °C, 24 h); bacteria lar components [9], especially for DNA damage, owing to cells were collected via centrifugation, washed three the reactions between hydroxyl radicals and other biomol- times with PBS, and then diluted to an optical density ecules [10, 11]. However, iron deficiency can also cause at 600 nm (OD 600) of 0.01 in 100 mL of normal TSB malnutrition cell death in severe cases [12]. In vivo, iron is to culture(180rpm,28°C). usually oxidized to an insoluble form due to its special physico-chemical properties, bonding with heme, ferritin, Sample collection hemoglobin, and transferrin within the cells, and thus is A. hydrophila NJ-35 cells (OD 600 ≅ 0.8) in normal and not readily accessible to bacteria [13]. In response to this iron-limited groups were collected by centrifugation iron deficiency predicament, microorganisms have evolved (5000 rpm, 4 °C, 10 mins). The pellet was rinsed twice a series of sophisticated mechanisms to compete against with saline and stored immediately at − 80 °C until further the host, such as the secretion of siderophores [14], to transcriptomic and proteomic analyses. The supernatant grab iron from transferrin, hemoglobin, and ferritin and was retained, filtered (MILLEX®GP filter unit, 0.22 μm), maintain iron dynamic balance for bacterial growth, pro- and frozen at − 20 °C, and it was used for the following liferation, and toxin secretion [15–17]. During the past de- proteolytic and hemolytic activity analyses. cades, the bacterial iron acquisition system and virulence have attracted much attention. For example, CaFTR1- Determination of iron concentration mediated iron-uptake was proven to be an important The atomic absorption spectrophotometry (GB/T 5009. virulence factor of Candida albicans [18], iron-responsive 90–2003) method [29] was used to the measure varia- transcriptional repressor PerR was required for full virulence tions in the intracellular iron of A. hydrophila NJ-35 in in Staphylococcus aureus [19], and FeoB was determined to normal and iron-limited groups, as well as the iron con- play an important role in Fe acquisition expression of viru- centration in the broth. Samples were analyzed by the lence of Helicobacter pylori [20]. Jiangsu Provincial Food Safety Testing Co., Ltd. Pathogenic bacteria virulence factors under iron- restricted growth conditions have previously been pub- Quantitative transcriptomics (RNA-seq) lished [21–24]. Proteomes and transcriptomes reflect (i) RNA isolation and mRNA purification gene expressions from two different levels, and their Total RNA was purified using an RNAqueous kit (Thermo joint analysis provides more complete expression infor- Fisher Scientific, San Jose, CA, USA) according to the mation about bacteria. Therefore, in this study, an iron manufacturer’s instructions. The RNA concentration and stress model was established to maximize the simulation integrity (RIN) were measured following the previous de- of iron deficiency environment in vivo, and the effects of scription of Wang et al. [30]. The mRNA was enriched iron-restricted stress on the growth and virulence of A. using a MICROBExpress Kit (Ambion, USA) [31], and de- hydrophila were evaluated comprehensively by combin- termined on Agilent 2100 Bioanalyzer. ing transcriptome and proteomics data. (ii) cDNA Synthesis, Illumina sequencing and library Methods construction Selection of iron chelator concentration and growth of A. Bacterial mRNA was fragmented using an RNA fragmenta- hydrophila tion kit (Illumina, San Diego, CA, USA). Double-stranded A. hydrophila (NJ-35) was isolated from dead cultured cDNA was synthesized using SuperScript II Reverse Tran- cyprinid in Jiangsu Province, China [25], and kindly pro- scriptase (Invitrogen, Carlsbad, CA) according to the man- vided by Professor Yongjie Liu from the College of Veter- ufacturer’s recommendations. Libraries were prepared with inary Medicine, Nanjing Agricultural University, P.R. the standard protocol of the TruSeq RNA Sample China. We selected 2,2’-Bipyridyl (Bip) (Sinopharm Chem- Prep v2 Low Throughput (LT) kit. Paired-end sequencing ical Reagent Co., Ltd., Shanghai, China) as the ferrous iron was processed by the Hiseq™2000 (Illumina, San Diego, chelating agent because of its high cell membrane perme- CA, USA) sequencer. ation and intracellular iron sequestering ability [26–28]. The accuracy and virulence of A. hydrophila NJ-35 were (iii) Bioinformatics Analyses confirmed by 16S rRNA gene sequencing (Biological The assembled reads were mapped to the complete gen- Engineering Technology Co., Shanghai, China) and lab in- ome of the A. hydrophila NJ-35 strain (http://www.ncbi. fection assays, respectively. Six concentrations (0, 100, 200, nlm.nih.gov/nuccore/CP006870.1). The QC of alignment 300, 400, and 500 μM Bip in normal tryptic soy broth was produced based on the standard generated by Qin medium (TSB; BD; final pH = 7.3)) were set to detect the et al. [31]. The gene expression level was calculated optimal concentration according to the growth curve of A. using the RPKM method (fragments per kb per million hydrophila NJ-35. A. hydrophila NJ-35 was inoculated in reads) [32]. Differentially expressed genes (DEGs) were Teng et al. BMC Microbiology (2018) 18:52 Page 3 of 17 identified with EdgeR software [33], and used to generate collected at a 4.5 min interval for 6–45 min, while the last statistical information such as expression level, fold segment was collected from 46 to 50 mins for a total change, p-value and FDR (false discovery rate). The spe- of 10 segments. Each segment was dried and used for cific filter conditions of DEGs were: log (fold change) ≥ 2, subsequent RPLC-MSMS analyses. p < 0.05 and bcv (biological coefficient of variation) = 0.01. GO enrichment analyses of DEGs were performed on RPLC-MSMS analyses website (http://www.geneontology.org/). The calculation In brief, samples were resuspended with Nano-RPLC buffer, method, p-value formula and enrichment score were ana- filtered through a C18 nanoLC trap column, and a lyzed according to the method reported by Yan et al. [34]. Chromxp C18 column (75 μm × 15 cm, C18, 3 μm120 Å). Additionally, the DEGs were subjected to KEGG en- The Eksigent nanoLC-Ultra™ 2D System (AB SCIEX) was richment analyses [35] to identify their main metabolic used to perform the online Nano-RPLC. Triple TOF 5600 pathways. The formula used for calculation was the system (AB SCIEX, USA) was used to analyze MS data same as that in the GO analyses. combined with Nanospray III source (AB SCIEX, USA). Quantitative proteomics (iTRAQ) (iiii) protein identification and quantification (i) Protein extraction, quantization, and SDS-PAGE Data were processed with the Protein Pilot Software v. 5. electrophoresis 0 (AB SCIEX, USA) against the NCBI database using the The extract of whole cellular protein was conducted ac- Paragon algorithm [41]. The results of protein quantifi- cording to Isaacson et al. [36] with some modification. cation were obtained by the matching of tandem mass The bacterial cells pellets were suspended in cooled spectrometry (MS) data and theoretical data, and was acetone (1 h, − 20 °C), centrifuged (15,000×g, 15 mins, performed with the search option: emphasis on bio- 4 °C), and dried with a vacuum freeze dryer. The sam- logical modifications. ples were resuspended in cold saturated-phenol (pH 7.5) An Orbitrap Elite high-resolution mass spectrometer and shaken (30 mins, 4 °C). The upper phenolic phase (Thermo Fisher Scientific, USA) was used for ITRAQ was collected by centrifugation (5000×g, 30 mins, 4 °C), quantitative proteomic analyses. Normalized high-energy 5 volumes of cold 0.1 M ammonium acetate in methanol collision dissociation (HCD) was performed, with the col- was added, and then it was stored (1 h, − 20 °C). After lision energy set at 30%. A protein database search and centrifugation (5000×g, 30 mins, 4 °C), the pellets were quantification were performed using Maxquant 1.5.1.0 washed and mixed with 2 volumes of ice-cold methanol. (Thermo Fisher Scientific, USA). The protein database The pellets were centrifuged, dried and dissolved in lysis contained 4119 proteins (https://www.ncbi.nlm.nih.gov/ solution (1 h, 30 °C). The supernatants were isolated by genome/?term=Aeromonas+hydrophila, GCF_000014805. centrifugation (15,000×g, 15 mins). The protein concen- 1_ASM1480v1_protein.faa). Oxidation (M) and acetyl trations were measured with the BCA method [37], after (protein N-term) were used as the variable modifications which they were stored at − 80 °C for iTRAQ analyses. and carbamidomethyl (C) was the fixed modification. The Additionally, 10 μg samples were subjected to 12% MS/MS tol. (FTMS) was 20 ppm. The protein quantita- SDS-PAGE, visualized and then scanned according to tion, peptides matching and the functional annotations of Candiano’sprotocol[38]. DEPs were performed according to the method reported by Yao et al. [24]. (ii) protein samples preparation and labeling The filter-aided sample preparation (FASP) method [39] Primer design, quantitative real-time PCR (qRT-PCR) was adopted for enzymatic hydrolysis of the proteins validation (100 μg). After 50 μL trypsin (50 ng/μL) digestion, pep- All of the sequence-specific primers of the target genes tides were labeled according to the manufacture’s protocol for qRT-PCR analyses were designed using Primer 5.0 for 8-plex iTRAQ reagent (AB SCIEX, USA). based on the obtained fragment (Table 3). The mRNA level of rpoB was used as an internal reference because (iii) 2D-LC-MSMS analyses of its stable expression according to Zhang et al. [42]. Total RNA from A. hydrophila was extracted using RPLC analyses RNAiso Plus (TaKaRa, Japan), and measured using a The dried samples were resuspended with 100 μL buffer Nanodrop 2000 (Thermo Fisher Scientific, USA), the A, after which reversed-phase liquid chromatography RNA concentration of each sample were diluted to (RPLC) was employed on an Agilent 1200 HPLC System 40 ng/μL, and then 2 μg of the total RNA was subjected (Agilent). Separation was conducted according to the to the following quantitative analysis with a One Step method of You et al. [40]. The first segment was collected SYBR® PrimeScript® Plus RT-PCR Kit (TaKaRa, Dalian). from 0 to 5 mins, after which each additional segment was Triplicate quantitative assays were performed on each Teng et al. BMC Microbiology (2018) 18:52 Page 4 of 17 type of cDNA using the ABI 7500 Real-time PCR System onto LB semisolid agar plates containing 0.3% agar (to (Applied Biosystems, Foster City, CA, USA) and ana- determine swimming ability) and 0.5% agar (to determine lyzed with the two-standard curve method. swarming motility). The LB plates were subsequently sealed with parafilm and incubated at 28 °C for 24 h (three Proteolytic activity parallel groups were set up for each group). At the Proteolytic activity was measured by an azocasein assay end of the culture period, the migration distance from method of Swift et al. [43] and Chu et al. [44], with some the colony edge to the colony center was determined. modifications. Briefly, 150 μL of normal group and iron- The experiment was repeated three times. limitation group NJ-35 culture supernatants were added to 1 ml of 0.3% azocasein (Sigma, St. Louis, USA) in 0. Infection assays in vivo 05 M Tris-HC1 and 0.5 mM CaCl (pH 7.5), then they A health check was conducted and healthy M. amblyce- were incubated (37 °C, 30 mins) respectively. Precooling phala (50 ± 5 g) were obtained from the Nanquan Ex- trichloroacetic acid (l0%, 0.5 ml) was then added to stop perimental Station of the Freshwater Fisheries Research the reaction, after which the samples were allowed to Center (Chinese Academy of Fishery Sciences, China) and stand for 15 mins at room temperature, then they were acclimatized in circulating water system with thermo- centrifuged (12,000 rpm, 10 mins, 4 °C) to remove the control for 2 weeks before use. Fish were given commercial precipitate. Next, 500 μL of the supernatants were added feed. The water temperature fluctuated between 27.5–28. to an equal volume of NaOH (1 mol/L). The supernatants 5 °C, with a pH between 7.2–7.8, and the DO was about 5. (200 μL) were subsequently transferred to a 96-well tissue 5mg/L. culture plate, after which the absorbance (OD400) of Strain NJ-35 was inoculated aseptically into normal the supernatant was measured. The proteolytic activity TSB medium and iron-limitation medium and then in- was calculated using the following equation: proteolytic cubated for 18 h at 28 °C while shaking at 180 rpm. The activity = OD sample – OD blank control artificial challenge experiment was performed as the pre- 400nm 400nm (normal TSB/iron limitation TSB). vious report [47]. To determine the 50% lethal dose (LD )[48], five groups of 20 M. amblycephala each Hemolytic activity were injected intraperitoneally with 150 μL of serial ten- 9 8 7 6 Hemolytic activity was determined as previously described fold diluted bacterial suspensions (1 × 10 ,10 ,10 ,10 , [45, 46], and sheep blood (Ping Rui Biotechnology, China) and 10 CFU·mL-1 measured by turbidimeter (Yue Fung was prepared by washing thrice with PBS. Washed sheep Instrument Co., Ltd., Shanghai, China)), which were blood (10 μL) was added to 490 μL of the experiment diluted with 0.9% saline. Next, an experimental group supernatants (sample), normal TSB/iron limitation TSB and a control group were injected intraperitoneally with (blank control), 1% (v/v) Trinton X-100 (positive control), 150 μL A. hydrophila (LD ) iron-limited and A. hydro- or PBS (phosphate buffer solution, negative control). After phila (LD ) basal, respectively, and the virulence was 30 mins of incubation at 37 °C, all of the samples were compared. Three replicate tanks per challenge isolate centrifuged (5000 rpm, 10 mins) at room temperature. (containing 20 fish each) were used to calculate survival The supernatants (200 μL) were then transferred to a 96- (from a total of 60 fish per isolate). The mortality of the fish well tissue culture plate, after which the absorbance of of experimental groups and control groups were monitored hemoglobin released for each solution at 540 nm was (7 days), and the activity and behavior were recorded daily; measured. The percentage of hemolysis was calculated pathogenic bacteria were isolated and identified from the using the following equation: hemolysis (%) = (OD lesion tissues of dead fish as the judging standard. 540nm sample - OD blank control)/ (OD positive con- 540nm 540nm trol Trinton X-100 - OD negative control PBS). Results 540nm Growth of A. hydrophila under different iron-limitation Lipase activity medium Bacterial cells were centrifuged and washed with PBS, after The effects of different concentrations of Bip on the growth which 5 μL of bacterial fluid was used to inoculate the LB of A. hydrophila are shown in Fig. 1.Whencompared with medium containing a 1% mass fraction of Tween 80. Sam- the control group, inhibitory effects were observed in the ples were then incubated at 28 °C for 24 h, after which they Bip addition groups, and higher Bip concentrations delayed were observed for lipase production, which was indicated the time of entering the logarithmic phase and reduced the by a white precipitate zone around the colony. maximum. When the Bip concentration was 500 μM, the growth of A. hydrophila was totally inhibited for at least Motility 24 h. Due to the significant inhibition and higher cells con- The target bacteria were centrifuged and washed with centration, 200 μMBip waschosenasthe proper iron- sterilized PBS. Next, 5 μL of bacterial fluid was dropped limitation concentration for subsequent analyses. Teng et al. BMC Microbiology (2018) 18:52 Page 5 of 17 reflecting significant changes and showing a strong correl- ation between the transcripts and proteins. Overall, 680 transcriptomes showed DEGs with no difference in pro- teins, while 35 transcriptomes showed different proteins but no difference in genes. Conversely, the expression of the following six genes and proteins was opposite (e.g., when the gene was upregulated, the protein was downregulated and vice versa): (U876_04575, YP_ 857861.1), (U876_17130, YP_855747.1), (U876_17135, YP_855746.1), (U876_19295, YP_855421.1), (U876_ Fig. 1 Effect of Bip supplementation on A. hydrophila growth. 20135, YP_855265.1), and (U876_21295, YP_855025.1). Growth curve (OD )of A. hydrophila NJ-35 grown in TSB medium This exception can be caused by regulation at several in the presence of 0, 100, 200, 300, 400, and 500 μM Bip levels, such as post transcriptional processing, degradation of the transcript, translation, post-translational processing Expression profile of iron-limited A. hydrophila and modification. In summary, most of the trends in DEP Based on the transcripts of A. hydrophila, 4327 genes abundance were consistent with the DEG data. were identified and quantified (Table 1). After filtering with FDR, 1204 genes were found to be differentially Functional classification of enriched DEGs and DEPs by expressed between the control and iron-limitation GO and KEGG groups. Detailed information for most of the DEGs is GO enrichment analyses were used to classify the enriched shown in Table 2. In comparison, the quantity of down- DEGs and DEPs between the control and iron-limitation regulated DEGs detected (603) was greater than that of groups using bioinformatics methods, and the results are the up-regulated genes (601). A total of 2244 proteins listed in Additional file 2: Excel S2 and Additional file 3: were identified; 2012 were quantified and 1946 were Excel S3, respectively. As shown in Fig. 3,the following correlated with the transcripts. Additionally, while com- three ontologies (molecular function, cellular component pared with the control group, a total of 236 DEPs (90 and biological process) were observed. up-regulated and 146 down-regulated) were identified in DEGs were distributed in up to 1460 GO terms, while the iron-limitation groups with an at least 2-fold differ- DEPs were classified into 402 GO terms. In this case, ence, and 167 of the DEPs were correlated to the corre- GO terms related to bacteria energy metabolism, iron sponding DEGs, which have the same trends. Fewer ion transport, and virulence. Based on the ‘−log Pvalue’, DEPs are probably due to the removal of some proteins most of the GO terms in the biological process category that were secreted by A. hydrophila NJ-35 in the super- were associated with energy metabolism (Fig. 3a and b). natant of the experimental design. Additionally, six genes were categorized as ‘glycerol cata- bolic process’ (GO: 0019563), three as ‘propionate cata- Integration analyses of transcriptome and proteome bolic process, 2-methylcitrate cycle’ (GO: 0019629), five To identify robust pathways that were corroborated by both as ‘oxidative phosphorylation’ (GO: 0006119), and five as datasets, we integrated the differentially expressed transcripts ‘respiratory electron transport chain’ (GO: 0022904). Re- and proteins to find the corresponding genes and proteins, garding proteomics, DEPs were mainly involved in the syn- and the results are listed in Additional file 1:Excel S1. thesis and transport of iron ions and proteins, particularly The distribution of the corresponding mRNA: protein the following GO terms: ‘iron assimilation’ (GO: 0033212), ratios is shown in a scatterplot of the log -transformed ‘ion transport’ (GO: 0006811), ‘enterobactin biosynthetic ratios. As shown in Fig. 2, almost all of the log mRNA: process’ (GO: 0009239), ‘protein secretion’ (GO: 0009306), log protein ratios are concentrated at the center of the ‘protein transport’ (GO: 0015031), and ‘electron transport plot, where mRNA and protein levels did not vary above chain’ (GO: 0022900). 2-fold. Integration analyses of transcriptome and proteome In the cellular component category (Fig. 3a and b), data revealed that 67 genes and their corresponding pro- three genes were categorized as ‘glycerol-3-phosphate teins were up-regulated, while 94 were down-regulated, dehydrogenase complex’ (GO: 0009331), five as ‘proton- Table 1 Overall features of the iron-limitation responsive expression profile Group name Type Number of genes Number of proteins Number of correlations Control-VS-Iron-Limitation Identification 4327 2244 1946 Control-VS-Iron-Limitation Quantitation 4327 2012 1733 Control-VS-Iron-Limitation Differential Expression 1204 236 167 Teng et al. BMC Microbiology (2018) 18:52 Page 6 of 17 Table 2 List of differentially expressed genes under iron restriction Accession Description Log FC U876_09860 Biosynthesis of siderophore group nonribosomal peptides 9.3945 U876_18585 ABC transporters 7.3209 U876_18590 ABC transporters 7.2995 U876_11875 Propanoate metabolism 3.6179 U876_18275 Two-component system|Bacterial chemotaxis 2.7194 U876_05565 Carbon metabolism|Glycolysis / Gluconeogenesis|Citrate cycle (TCA cycle)|Pyruvate metabolism|Butanoate metabolism|Carbon 2.3607 fixation pathways in prokaryotes U876_16675 Quorum sensing 2.3034 U876_14615 Oxidative phosphorylation|Two-component system 2.1593 U876_13185 Ribosome 1.7769 U876_13000 Cysteine and methionine metabolism|Selenocompound metabolism 1.5614 U876_17160 RNA transport 1.4601 U876_00445 Glycine, serine and threonine metabolism −1.5113 U876_10020 Purine metabolism|Drug metabolism - other enzymes −1.5726 U876_15390 Biotin metabolism −2.2592 U876_09705 Selenocompound metabolism|Aminoacyl-tRNA biosynthesis −3.4550 U876_00975 Biosynthesis of amino acids|Arginine biosynthesis −3.5694 U876_17185 Lysine degradation|Tropane, piperidine and pyridine alkaloid biosynthesis −3.8646 U876_15985 Fructose and mannose metabolism|Phosphotransferase system (PTS) −4.5035 U876_12875 Nitrogen metabolism −5.4990 U876_00965 Arginine biosynthesis −6.0546 Note: FC, Fold change, the ratio of different expression levels between the iron-limitation group and the normal TSB group transporting ATP synthase complex, catalytic core F(1)’ Enriched KEGG terms are listed under Additional file 4: (GO: 0045261), four as ‘proton-transporting ATP syn- Excel S4 and Additional file 5: Excel S5, as transcripto- thase complex, coupling factor F(o)’ (GO: 0045263), and mics and proteomics, respectively. When compared with seven as ‘bacterial-type flagellum hook’ (GO: 0009424). the whole genome, a total of 624 genes were present in Regarding proteomics, DEPs were mainly classified in the the 139 KEGG pathways as DEGs, and we selected the cell membrane and cytoplasm of GO terms, including ‘inte- 20 most critical KEGG pathways according to the enrich- gral component of membrane’ (GO: 0016021), ‘plasma mem- ment scores (Fig. 4a). The up-regulated KEGG pathways brane’ (GO: 0005886), ‘cell outer membrane’ (GO: 0009279), included 78 genes under the category of ‘ABC trans- ‘cytosol’ (GO: 0005829), and ‘cytoplasm’ (GO: 0005737). porters’ (ko02010), 20 genes under ‘TCA cycle’ (ko00020), In the molecular function category (Fig. 3a and b), 11 and 38 genes under ‘quorum sensing’ (ko02024). We in- genes were categorized as ‘receptor activity’ (GO: 0004872), ferred that transport, energy production and bacteria three as ‘energy transducer activity’ (GO: 0031992), three as interact with each other and may play important roles ‘cytochrome o ubiquinol oxidase activity’ (GO: 0008827), via stress responses that are regulated through several four as ‘siderophore uptake transmembrane transporter ac- pathways. The down-regulated KEGG pathways in- tivity’ (GO: 0015344), and three as ‘siderophore transmem- cluded 47 genes categorized as ‘Ribosome’ (ko03010), brane transporter activity’ (GO: 0015343). Regarding 71 as ‘Carbon metabolism’ (ko01200), 31 as ‘Pyruvate proteomics, DEPs were mainly related to protein activity metabolism’ (ko00620), and 35 genes as ‘Oxidative and binding capacity, including ‘siderophore transmem- phosphorylation’ (ko00190), which confirmed that bac- brane transporter activity’ (GO: 0015343), ‘receptor activity’ teria slowed down material synthesis and life activities. (GO: 0004872), ‘iron ion binding’ (GO: 0005506), ‘heme With respect to proteomics, a total of 41 proteins were binding’ (GO: 0020037), ‘metal ion binding’ (GO: 0046872), detected in the 34 KEGG pathways by DEP, while only and ‘porin activity’ (GO: 0015288). In summary, GO term eight pathways were found to be significantly enriched by enrichment analyses further explained that metabolism, filtration (Fig. 4b). The up-regulated KEGG pathways in- biosynthesis, transmembrane transport and redox homeo- cluded three that were labeled under ‘biosynthesis of sid- stasis should be tightly regulated. erophore group nonribosomal peptides’ (aha01053) and Teng et al. BMC Microbiology (2018) 18:52 Page 7 of 17 Fig. 2 Relationship patterns of all of the quantitative mRNA and protein. In the nine-quadrant diagram, the abscissa is the protein expression and the ordinate is the gene expression. Each color denotes a log mRNA ratio and a log protein ratio. Gray (filtered) represents genes and proteins 2 2 with no significant difference, red (Cor_up) indicates up-regulated genes and proteins, green (Cor_down) indicates down-regulated genes and proteins, purple (Opposite_Sig) indicates that DEGs and DEPs show opposite up- and down- regulation and blue (Single_Sig) indicates that one of the genes and proteins differ 10 that were labeled under ‘ABC transporters’ (aha02010), and generation of hemolysin (U876_04005, U876_15265, indicating clear changes in synthesis and transportation of U876_16300, and U876_16315). Heat map analyses siderophores. The down-regulated KEGG pathways in- (Fig. 5) were used to visualize genes and proteins, and cluded 11 proteins that were classified as ‘oxidative phos- the results indicated a comprehensive impact and clear phorytation’ (aha00190), six as ‘butanoate metabolism’ changes in the regulation of virulence factors. (aha00650), five proteins as ‘TCA cycle’ (aha00020), five as ‘pyruvate metabolism’ (aha00620), seven as ‘carbon Validation of selected DEGs/DEPs by qRT-PCR analyses metabolism’ (aha01200), and six as ‘two-component sys- To further evaluate the expression of genes in an iron- tem’ (aha02020), indicating the bacteria repress energy limited environment, 20 virulence genes (13 up-regulated metabolize to adaptive constraint environment. Conversely, and seven down-regulated genes) together with reference the total number of DEPs among them was far smaller genes (rpoB) were selected for investigation based on their than that of the DEGs, and most DEGs and DEPs were expressions, which were measured by real-time quantitative down-regulated. PCR (RT-qPCR) (Table 3) according to the results of the GO analyses. These selected genes were involved in Clustering of virulence genes and proteins in A. hydrophila virulence factors, hemolysis, secretion systems, lipases, in iron-limited medium phospholipids, serine-type peptidases, metallopeptidases, According to the bioinformatics analyses, we found that flagella, polysaccharides, siderophore transporters, quorum there were 60 virulence factors in the differential genes, sensing, and outer membrane production. which mainly fell under the category of synthesis of iron The results of qPCR showed that the majority of the carriers (U876_01620, U876_18555, U876_21285, U876_ selected virulence factors (90%, 18/20) were consistent 21455, U876_23515, and U876_24445), motility of flagella with the transcriptome data. Notably, five virulence- (U876_20435, U876_07265, U876_07270, and U876_07305), related factors, U876_15265 (hemolysin, log FC = 3.80), 2 Teng et al. BMC Microbiology (2018) 18:52 Page 8 of 17 Fig. 3 GO enrichment analyses of DEGs and DEPs Control group vs Iron-Limitation group. GO term analyses of transcriptomics (a) and proteomics (b) that were catalogued as Biological Process, Cellular Component, and Molecular Function U876_15575 (secretin, log FC = 5.00), U876_18585 (hemin log FC = 3.33), and U876_09860 (2,3-dihydroxybenzoate- 2 2 ABC transporter substrate-binding protein, log FC = 4.71), AMP ligase, log FC = 6.20) were shown to be significantly 2 2 U876_20975 (transcriptional activator protein AhyR/AsaR, up-regulated (log FC > 3.00) under iron-limited conditions. 2 Teng et al. BMC Microbiology (2018) 18:52 Page 9 of 17 Fig. 4 KEGG enrichment analyses of DEGs and DEPs Control group vs Iron-Limitation group. KEGG enrichment analyses of transcriptomics (a) and proteomics (b) Teng et al. BMC Microbiology (2018) 18:52 Page 10 of 17 Fig. 5 Clustering of 60 mainly related virulence genes and proteins. Numbers are listed as the log value of difference multiples. Expression differences are shown in different colors; red indicates up-regulation, while green indicates down-regulation. A heatmap was used to visualize the genes and proteins that were related to virulence factor (hemolysis, secretion system, lipase, phospholipid, serine-type peptidase, metallopeptidase, flagellum, polysaccharides, siderophore transporter, quorum sensing, and outer membrane) Moreover, two selected genes, U876_07270 (flagellar hook concentration of 0.664 mg/100 g in the normal TSB protein FlgE) and U876_12225 (murein transglycosy- group strain cell was lower than 0.998 mg/100 g in lase A), showed appositive results to the RNA-seq the iron-limitation group strain cell. All of the results data, which might have been due to differences in the are shown in Table 4. analyses methods. Effect of iron-limitation on virulence factors production in Determination of iron concentration A. hydrophila Atomic absorption spectrophotometry revealed that the As shown in Table 5, the total protease activity in super- medium iron concentration of 0.44 mg/100 g in the normal natants from A. hydrophila NJ-35 growing without Bip TSB group was higher than 0.28 mg/100 g in the iron- was 0.105 (OD400 nm), whereas the presence of Bip re- limitation group, indicating that iron scavenger 2,2- bipyri- sulted in a significant increase in protease activity to 0.36 dine has a higher efficiency. After bacterial growth, the (OD400 nm) (Fig. 6a). When compared with the control medium iron content of the normal TSB group was higher group, the hemolytic activity of A. hydrophila NJ-35 was than that of the iron-limitation group. Surprisingly, the significantly enhanced under iron limitation, indicating Teng et al. BMC Microbiology (2018) 18:52 Page 11 of 17 Table 3 Primers and sequences used in this study for q-PCR Name Gene product Primer Sequence (5′➔3′) qRT-PCR Illumina FC FC Log Regulated Log Regulated 2 2 U876_04005 RTX toxin F GCCAAGAACCTGACCTAC 0.78 Up 1.06 Up R TAACTACCGTCCGACCAT U876_15265 hemolysin F TGCTCGTACTTGCTGTTG 3.85 Up 3.80 Up R GACTACCTGCTGCTGGAT U876_15575 secretin F CGATGCGTACCGATATGT 5.00 Up 5.33 Up R AGACTAACAACCAGGATGAG U876_16325 type I secretion system F GCTCATCGCCTCAATACC 1.42 Up 1.02 Up permease/ATPase R TAGCCAGTGTGAGTCAGG U876_02495 phosphatidylcholine-sterol F TTCGGTGTTCCAGCCATA 2.34 Up 1.87 Up acyltransferase R CCAAGTATCAGGTCATCAAC U876_00180 lysophospholipase L2 F AGCACATAATCGTCAAACTG 1.23 Up 1.51 Up R GCCATCCTCATCGTCAAC U876_12850 FAD/NAD(P)-binding F CGATTACCACAAGATTGACC 2.34 Down −5.25 Down oxidoreductase R TGATCCAGCAGCACTATG U876_06295 HPr family phosphocarrier F CGGAGACCACAGTGATCT −0.86 Down −2.07 Down protein R TGTACGAGAAGTCTGTTGTT U876_06300 cysteine synthase A F CAGAGCAATACCCGTGTT 1.69 Up 1.99 Up R TCAACCGTGTTACCAAGG U876_14760 peptidase T F CCGAGGATCAAACCCATTC −1.23 Down −6.23 Down R CTTGCCGTGGAAGTTGTG U876_07265 flagellar hook capping protein F CAATGTCGGTTACCTGGAA −0.86 Down −1.05 Down R GTCCTTGTCCTTGCCATC U876_07270 flagellar hook protein FlgE F TCAGCGACCTACAGCAAT 0.25 Up −1.25 Down R CACCAGACAGCAGAGACT U876_12225 murein transglycosylase A F CCAGACTGATGCCGTAAC 1.25 Up −1.16 Down R CAAGATGACTCGTCGCTAC U876_03850 PAP2 family protein F GATGGTGCCGTTGTTCTC 2.54 Up 2.07 Up R ACAGCAGTGGTAGACAGAG U876_17510 outer membrane protein F GGTGAGTGGAACGGTTAC −0.99 Down −2.14 Down R ATCGGAGTGCCAGTAGATA U876_18585 hemin ABC transporter F CGATCTGGTGCTGGTTAG 4.71 Up 7.32 Up substrate-binding protein R CTTGATCCACTTGGCGAT U876_21455 TonB-dependent siderophore F CGTCTCAGTCACCAGTCT 2.61 Up 6.07 Up receptor R ATCCAGGTTGTTGTTCTTGT U876_20975 transcriptional activator F TTGAACAGCACCACCTTG 3.33 Up 1.22 Up protein AhyR/AsaR R GCTTGAGTACCTCGAACAT U876_23540 LuxR family transcriptional F GAAGGAGTGCCTGTTCTG 1.14 Up 1.45 Up regulator R TATGATGCCGCTGGAGAT U876_09860 2,3-dihydroxybenzoate-AMP F TACAGGATGCCGATGGTTA 6.20 Up 9.23 Up ligase R ATCCGTGCTGACGATGAA U876_01300 DNA-directed RNA F GGATCACGGTGCCTACAT (rpoB) polymerase subunit beta R TAACGCTCGGAAGAGAAGA Teng et al. BMC Microbiology (2018) 18:52 Page 12 of 17 Table 4 Determination of iron concentration under two culture bacterial multiplication was enhanced after injecting ex- conditions ogenous iron into experimentally infected animals, and the Group Medium before Medium after Strain cell/ virulence of pathogens including Vibrio cholerae, Pseudo- culture/(mg/100 g) culture/(mg/100 g) (mg/100 g) monas aeruginosa, Klebsiella pneumoniae,and Mycobacter- b a Normal TSB 0.44 ± 0.032 0.27 ± 0.029 0.664 ± 0.019 ium tuberculosis was established with sufficient iron [8, group 53]. A. hydrophila establishes virulence through many a b Iron-Limitation 0.28 ± 0.021 0.20 ± 0.030 0.998 ± 0.012 mechanisms [54], including iron-binding systems, secre- group tion systems, biofilm formation, flagella and pili adhe- Note: Means with different lowercase letters within the same column were sion, structural proteins, phospholipids, polysaccharides, significantly different (P < 0.05) hemolysis, collagenase, serine protease, metalloprotease, that NJ-35 produced 83.8% more hemolysin (Fig. 6b). To enolase, lipase, and nucleases [5, 6]. The pathogenesis of observe the hemolysis ability, sheep blood agar plates were diseases involves most virulence factors [1], beginning with used for rough detection. A. hydrophila NJ-35 under iron molecular changes at the micro level and progressing to limitation generated a large hemolytic zone on the blood phenotypic changes at the macro level [55]. Under iron- agar plates compared to the control group, but the lipase limited conditions, virulence genes and proteins were up- activity and swarming motility did not differ significantly regulated more than down-regulated (Fig. 5), suggesting (Table 5). Interestingly, the swimming ability of the bac- that virulence expression was enhanced in A. hydrophila to teria was strong under iron limiting conditions, which compensate for iron insufficiency, which was confirmed in could reflect attempts to move to areas with more suitable F. tularensis [56]. These virulence factors exerted synergis- conditions (Table 5). tic effects [57] and contributed to the production of toxins. The results of the infection assays further confirmed this Infection assays conclusion (Fig. 7). Theferric uptakeregulator (Fur) is a The isolated pathogenic bacteria were A. hydrophila after negative regulator in iron acquisition systems [58]that con- morphological, physiological and biochemical, molecular trols the expression of 90 virulence and metabolic genes [7, identification. Megalobrama amblycephala injected with 15, 59]. For example, the biosynthesis of rhizoferring, an A. hydrophila NJ-35 showed distinct mortality rates under iron siderophore in F. tularensis, is regulated by operon iron and non-iron limited conditions (Fig. 7). Although fslABCDEF [60]. In this study, the expression of the fur the difference was not significant, the survival rate in the gene (U876_15170) was up-regulated (log FC = 0.3187). group injected with A. hydrophila was substantially higher This phenomenon could be explained by the higher iron (by 19.77%) than that of the iron-limitation group at four concentration in bacterial cells of the iron-limited group. days post-challenge. At the sampling time-point, more iron was stored in the iron-limited group, after which fur was up-regulated to re- Discussion duce the iron absorption [61]. Comparative transcriptomic and proteomic analyses Iron homeostasis was coordinated by the absorption, The survival and proliferation of bacteria was sensitive transport, utilization, and storage of iron ions [62]. A. to environment factors. Many environmental stress fac- hydrophila utilized multiple iron sequestration systems tors, e.g., pH, temperature, oxygen, acidity and salinity to hijack host iron ions [63]. Under an iron deficient en- [49, 50] significantly affected the expression of virulence. vironment, A. hydrophila secreted large amounts of iron Iron limitation is an important external stimulus [51] transporters and iron-specific scavenger-siderophores. that has profound impacts on almost all bacteria. The The same results were confirmed by transcriptome ana- culturability and growth rate of A. hydrophila were lyses of Bacillus cereus ATCC 10987, which showed the reduced under iron-limited conditions [52]; however, upregulation of predicted iron transporters in the pres- ence of 2,2-Bipyridine [64]. As an important virulence characteristic of pathogens to both animals and plants Table 5 Effect of iron limitation on A. hydrophila extracellular [65], siderophores were formed and played a major role enzyme activity and motility in microbial iron acquisition. Siderophore-assisted iron Virulence NJ-35 uptake and reductive iron assimilation are both induced Control Iron-Limitation upon iron starvation [58]. In previous studies, A. hydro- Lipase (cm) 0.95 ± 0.16 1.06 ± 0.21 phila was found to secrete siderophores to compete with Blood-plate hemolysis (cm) 0.90 ± 0.07 1.07 ± 0.09 transferrin in vivo to meet the iron demand required for a b Swimming ability (cm) 1.02 ± 0.01 1.17 ± 0.01 growth and virulence [66]. Measurement of the iron b a concentration confirmed that the iron chelating ability of Swarming motility (cm) 0.89 ± 0.12 0.84 ± 0.07 bacterial siderophores was notable (Table 4), because E. coli Note: Means with different lowercase letters within the same row were significantly different (P < 0.05) [67]and A. hydrophila synthesize and secrete enterobactin Teng et al. BMC Microbiology (2018) 18:52 Page 13 of 17 Fig. 6 Effect of control and iron-limitation conditions on A. hydrophila NJ-35. a Total protease, and (b) hemolytic activity. The data represent the mean values of three independent experiments and are presented as the means ± SD siderophores [68] in response to iron starvation. Enterobac- iron storage protein in A. hydrophila [72]. Thedatadem- tin synthase subunit E (entE), which is encoded by entA, onstrated that ferritin (U876_00270, log FC = − 0.4088) entB, and entC genes, is a key enzyme involved in the syn- and bacterioferritin (U876_02285, log FC = 13.4043) partic- thesis of isochorismate synthase. In both E. coli and A. ipated in iron ion transport and storage, which may benefit hydrophila, a 22 kB gene cluster including entD-fepA- the survival of bacteria. The up-regulation of this protein fes-entD-fepE-fepC-fepG-fepD-fepB-entC-entE-entB-entA- may be responsible for the increased intracellular iron ybdA genes encodes proteins responsible for the synthesis concentration in A. hydrophila. The expression levels and transport of enterobactin [69]. During this process, the of bacterioferritin in different isolates, including F. entE polypeptide is responsible for activating the DHBA tularensis, also varied [73, 74]. The TonB mechanism carboxylate group with ATP by forming the enzyme-bound is essential to the virulence of avian pathogenic E. coli 2,3-dihydroxybenzoyadenylate as an intermediary in the [75], indicating that a specific TonB-dependent outer biosynthetic pathway [70]. Genes with similar enterobactin membrane receptor might be involved in the transport of transport functions (iroN, fepC, cirA, fepC, and iroC) were iron from transferrin [76]. TonB-dependent outer mem- also found in Salmonella enterica [71]. After differential brane receptors TonB-2 (U876_00270, log FC = 5.9844), analyses of the genes and proteins, we found that the entE AHA_4249 (YP_858666.1, log FC = 6.1718), AHA_4250 expression level of gene U876_09860 (log FC = 9.39) and (YP_858667.1, log FC = 7.4891), and AHA_4251 (YP_ 2 2 protein YP_856992.1 (log FC = 15.46) had increased signifi- 858668.1, log FC = 10.7778) were found to be required for 2 2 cantly during the biosynthesis of the siderophore subunits the transfer of iron chelators and heme to the periplasm, (ko01053). Upon GO term analyses of the DEGs, the entE followed by transport to the cytoplasm by ATP-binding gene and protein expression levels were not increased sig- cassette (ABC)-type transporters. Inorganic iron in the nificantly, which may have been because of differences in periplasm is transported to the cytoplasm by membrane the analyses methods and software. Ferritin is the major transporters, such as Sfu ABC [77]. Iron influences a number of catalytic reactions involv- ing cell energy metabolism in vivo, including respiration and nucleic acid replication [78]. Overall, when iron de- mand is not met, some enzymes related to metabolism, the regulation of protein synthesis, and the ability of A. hydrophila to utilize nutrients, such as carbohydrates, decreased. It has been hypothesized that decreased viru- lence might be caused by the loss of metabolic activity and the lack of toxin production [79, 80]. According to bioinformatic analyses conducted in this study, the energy generation system and electron respiration chain appeared to be depressed under iron starvation, which is consistent with previous quantitative proteomic analyses of A. hydro- phila [24]. Iron scarcity reduces iron utilization in iron Fig. 7 Kaplan-Meier survival analyses of Megalobrama amblycephala nonessential pathways, and limited iron is used for the syn- challenged with A. hydrophila NJ-35 from normal and iron-limited media. thesis of iron-containing enzymes involved in the citric acid Data represent accumulative fish mortality in three replicates cycle and the electron transport chain [81]. For example, Teng et al. BMC Microbiology (2018) 18:52 Page 14 of 17 the expression of NADP-dependent glyceraldehyde-3- isolated from protease deficient strains of A. hydrophila phosphate significantly altered the antioxidant activity were found to lead to the death of catfish [88]. Blood-plate of bacteria, and NADPH is involved in the transform- hemolysis results showed that the hemolytic ability of A. 3+ 2+ ation of Fe into Fe in some of the identified hydrophila under iron deficiency was stronger than that bacteria [16]. Similar to S. pneumoniae in manganese under normal conditions, and it caused greater toxicity limited environments [82], the metabolic activity of and damage to the host. The results also showed that iron bacteria will become inert, so bacteria can survive in these exerted an inhibitory effect on extracellular hemolysin and environments for a long time [83]. Based on the high- protease activity. Notability, the presence of hemolysin throughput data analyses, it is apparent that 969 genes de- alone does not cause disease [89]. creased, 905 genes increased, 146 proteins decreased, 90 The invasion of pathogenic bacteria was found to be proteins increased, the gene and protein ratio was down- significantly correlated with the level of corresponding regulated, and the regulation of bacteria itself was also enzyme production, and protease activity [90], which is used to interpret iron starvation. consistent with the results of this trial. Not only can pro- teases degrade a variety of proteins to provide amino Virulence evaluation of A. hydrophila under iron-limited acids for bacterial survival and growth, but they can also environment directly cause tissue injury, resulting in the spread Many studies have shown that the virulence of A. hydro- through the defense mechanism and evasion of the im- phila increased in response to iron deficiency [52]. Two mune system of the host [91]. In addition, the A. hydro- aspects may contribute to the establishment of bacterial phila family of extracellular proteases can cooperate pathogenicity: invasiveness and toxin production [84]. with other virulence factors [92] to activate other patho- The invasive ability of A. hydrophila is closely related to genic factors. In this study, A. hydrophila NJ-35 under low- their motility, as well as the secretion of toxins, including iron growth conditions were detected with higher protease aerolysin, hemolysin, and enterotoxin, and extracellular activity than the control, demonstrating that iron scarcity protease. To evaluate the virulence of A. hydrophila more can promote NJ-35 virulence factor expression. comprehensively, we conducted an encompassing study of A. hydrophila hemolytic and enzymatic activity in vitro Conclusion and lethality rate in vivo. In this paper, we simulated the iron restriction environ- A. hydrophila pilus is an important coagulation factor ment in the fish host, coalition analyzed the transcriptome and a major colonization factor that enables bacteria to and proteomics data of A. hydrophila, and identified the adhere to host digestive epithelial cells during the invasion changes of enzyme activity, comprehensively revealed the process. In terms of virulence establishment, pili-assisted pathogenicity of A. hydrophila increased. This study also adhesion bacteria were 10–200 times more effective than provide a profound theoretical basis for the effect of ex- bacteria that do not express pili [85]. Flagella-mediated ogenous iron preparation on the toxicity of bacteria. motility also promotes the initial stages of adhesion [86]. In this study, although the swimming ability of the control Additional files group was significantly stronger than that of the iron re- striction group, swimming ability was enhanced under Additional file 1: Excel S1. The results of differentially expressed transcripts iron-limited conditions, indicating that A. hydrophila can andproteinsto findthe correspondinggenes andproteins. (XLSX87kb) overcome unfavorable conditions by accelerating their Additional file 2: Excel S2. The enriched DEGs GO terms between control and iron-limitation groups using bioinformatics methods. (XLSX 42 kb) swimming and adhesion abilities, thereby enhancing Additional file 3: Excel S3. The enriched DEPs GO terms between control their resilience to environmental restraints. Alterna- and iron-limitation groups using bioinformatics methods. (XLSX 42 kb) tively, these findings demonstrate the complexity of Additional file 4: Excel S4. The enriched DEGs KEGG terms between Aeromonas sp. virulence. control and iron-limitation groups using bioinformatics methods. (XLSX 25 kb) Lethal pathogenic extracellular products (ECPs) of A. Additional file 5: Excel S5. The enriched DEPs KEGG terms between hydrophila are produced to compete with rivals for lim- control and iron-limitation groups using bioinformatics methods. (XLSX 13 kb) ited iron resources [87]. After removal of ECPs by re- peated washing with normal saline, the invasion and Abbreviations pathogenicity of the pathogenic bacteria to the host cells ABC: ATPbinding cassette; Bip: 2,2’-Bipyridyl; BLAST: Basic Local Alignment Search Tool; CA: Citric acid; CID: Collision-induced dissociation; was reduced or even completely lost. As a typical ECP, DEGs: Differentially expressed genes; DEPs: Differentially expressed proteins; hemolysin that is synthesized and secreted into the organ- ECP: Extracellular products; FDR: False discovery rate; GO: Gene Ontology; ism’s environment can dissolve various sources of iron by HCD: High-energy collision dissociation; IDA: Information dependent acquisition; iTRAQ: Isobaric tags for relative and absolute quantification; destroying intracellular red blood cells or hydrolyzing KEGG: Kyoto Encyclopedia of Genes and Genomes; LC MS/MS: Liquid hemoglobin. Hemolytic activity was detected both in vivo chromatography tandem mass spectrometry; LD : 50% lethal dose; and in vitro in septic animals, and beta hemolysins MS: Mass spectrometry; NCBI: National Centre for Biotechnology Information; Teng et al. BMC Microbiology (2018) 18:52 Page 15 of 17 PBS: Phosphate buffer solution; PRIDE: PRoteomics IDEntifications; QC: Quality 9. Halliwell B, Gutteridge JM. Oxygen toxicity, oxygen radicals, transition control; qPCR: Real-time Quantitative polymerase chain reaction; RIN: RNA metals and disease. Biochem J. 1984;219:1–14. integrity value; RNA-seq: RNA Sequencing; SRA: Sequence Read Archive; 10. Braun V. Avoidance of iron toxicity through regulation of bacterial iron TCA: Tricarboxylic acid; TSB: Tryptic soy broth transport. Biol Chem. 1997;378:779–86. 11. Miller RA, Britigan BE. Role of oxidants in microbial pathophysiology. Clin Microbiol Rev. 1997;10:1–18. Funding This study was supported by The earmarked fund for China Agriculture Research 12. Zhang YC, Shen YY, Yan XH, Wang FD. Molecular mechanisms of System (CARS-45), Natural Science Foundation of China (31572662), Jiangsu Natural mammalian iron homeostasis. Chin. J Cell Biol. 2011;33:1179–90. Science Foundation (BK20171152), and Postgraduate Research & Practice Innovation 13. Teng T, Xi BW, Xie J, Chen K, Pao X, Pan LK. Molecular cloning and Program of Jiangsu Province (KYLX16_1082). The funding agencies have not been expression analysis of Megalobrama amblycephala transferrin gene and effects of involved in the design of research and collection, analysis and interpretation of data exposure to iron and infection with Aeromonas hydrophila. Fish Physiol Biochem. and in writing of manuscripts. We also thank OEbiotech.co.ltd for technical support 2017;43:987–97. in our transcriptome and proteomics analyses. 14. Telford JR, Raymond KN. Amonabactin: a family of novel siderophores from a pathogenic bacterium. J Biol Inorg Chem. 1997;2:750–61. 15. Litwin CM, Calderwood SB. Role of iron in regulation of virulence genes. Availability of data and materials Clin Microbiol Rev. 1993;6:137–49. The RNA-seq data and analyses discussed in this publication were deposited 16. Ratledge C, Dover LG. Iron metabolism in pathogenic bacteria. Annu Rev in the NCBI Sequence Read Archive (SRA) database under accession number Microbiol. 2003;54:881–941. SRR5894319. The mass spectrometry proteomics data have been deposited 17. Miethke M, Marahiel M. Siderophore-based iron acquisition and pathogen to the ProteomeXchange Consortium via the PRIDE partner repository with control. Microbiol Mol Biol Rev. 2007;71:413–51. the dataset identifier PXD007641. 18. Ramanan N, Wang Y. A high-affinity iron permease essential for candida albicans virulence. Science. 2000;288:1062–4. Authors’ contributions 19. Horsburgh MJ, Clements MO, Crossley H, Ingham E, Foster SJ. Perr controls BX, JX and PX conceived and designed the experiments; BX guided the oxidative stress resistance and iron storage proteins and is required for experiments, TT performed the experiments, analyzed the data; TT and BX wrote virulence in staphylococcus aureus. Infect Immun. 2001;69:3744. the paper, revised the paper, they contributed equally to this work; LP, KC 20. Velayudhan J, Hughes NJ, Mccolm AA, Bagshaw J, Clayton CL, Andrews SC, participated in the collection of samples, planning and coordination of the study, Kelly DJ. Iron acquisition and virulence in helicobacter pylori :a major role for provided general supervision. All authors read and approved the final manuscript. feob, a high-affinity ferrous iron transporter. Mol Microbiol. 2000;37:274–86. 21. Deng K, Blick RJ, Liu W, Hansen EJ. Identification of Francisella tularensis Ethics approval and consent to participate genes affected by iron limitation. Infect Immun. 2006;74:4224–36. The study protocol was granted by the Research Ethics Committee, Wuxi 22. Lenco J, Hubálek M, Larsson P, Fucíková A, Brychta M, Macela A, Stulík J. Fisheries College of Nanjing Agriculture University (Permit No. NJYY20160929–1), Proteomics analysis of the Francisella tularensis LVS response to iron and all methods were performed in accordance with the approved guidelines restriction: induction of the F. tularensis pathogenicity island proteins and regulations. IglABC. FEMS Microbiol Lett. 2007;269:11–21. 23. Folsom JP, Parker AE, Carlson RP. Physiological and proteomic analysis of Competing interests Escherichia coli iron-limited chemostat growth. J Bacteriol. 2014;196:2748–61. The authors declare that they have no competing interests. 24. Yao Z, Wang Z, Sun L, Li W, Shi Y, Lin L, Lin WX, Lin XM. Quantitative proteomic analysis of cell envelope preparations under iron starvation stress in Aeromonas hydrophila. BMC Microbiol. 2016;16:161. Publisher’sNote 25. Pang MD, Jiang JW, Xie X, Wu YF, Dong YH, Kwok AH, Zhang W, Yao HC, Lu Springer Nature remains neutral with regard to jurisdictional claims in published CP, Leung FC, Liu YJ. Novel insights into the pathogenicity of epidemic maps and institutional affiliations. Aeromonas hydrophila ST251 clones from comparative genomics. Sci Rep. 2015;5:9833. Author details 26. Caliaperumal J, Wowk S, Jones S, Ma YL, Colbourne F. Bipyridine, an iron chelator, Wuxi Fisheries College, Nanjing Agricultural University, Wuxi 214081, China. does not lessen intracerebral iron-induced damage or improve outcome after Key Laboratory of Freshwater Fisheries and Germplasm Resources intracerebral hemorrhagic stroke in rats. Transl Stroke Res. 2013;4:719–28. Utilization, Ministry of Agriculture, Freshwater Fisheries Research Center, 27. Alencar TD, Wilmart-Gonçalves TC, Vidal LS, Fortunato RS, Leitão AC, Lage C. Chinese Academy of Fishery Sciences, Wuxi 214081, China. Bipyridine (2,2′-dipyridyl) potentiates Escherichia coli lethality induced by nitrogen mustard mechlorethamine. Mutat Res. 2014;765:40. Received: 28 September 2017 Accepted: 5 April 2018 28. Lee P, Tan KS. Effects of epigallocatechin gallate against Enterococcus faecalis biofilm and virulence. Arch Oral Biol. 2015;60:393. 29. GB/T 5009.90–2003. Determination of iron, magnesium and manganese in foods. References Beijing: Standardization Administration of the People’s republic of China; 2003. 1. Janda JM, Abbott SL. The genus Aeromonas: taxonomy, pathogenicity, and 30. Wang XK, Yang RQ, Zhou YL, Gu ZX. A comparative transcriptome infection. Clin Microbiol Rev. 2010;23:35–73. and proteomics analysis reveals the positive effect of supplementary 2. Feelders RA, Vreugdenhil G, Eggermont AM, Kuiper-Kramer PA, van Eijk HG, 2+ Ca on soybean sprout yield and nutritional qualities. J Proteome. Swaak AJ. Regulation of iron metabolism in the acute-phase response: 2016;143:161. interferon gamma and tumour necrosis factor alpha induce hypoferraemia, 31. Qin N, Tan X, Jiao Y, Liu L, Zhao W, Yang S, Jia AQ. RNA-Seq-based ferritin production and a decrease in circulating transferrin receptors in transcriptome analysis of methicillin-resistant Staphylococcus aureus biofilm cancer patients. Eur J Clin Investig. 1998;28:520–7. inhibition by ursolic acid and resveratrol. Sci Rep. 2014;4:5467. 3. Reines HD, Cook FV. Pneumonia and bacteremia due to Aeromonas 32. Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ. Basic local alignment hydrophila. Chest. 1981;80:264–7. search tool. J Mol Biol. 1990;215:403–10. 4. Brenden RA, Huizinga HW. Pathophysiology of experimental Aeromonas hydrophila infection in mice. J Med Microbiol. 1986;21:311–7. 33. Robinson MD, Mccarthy DJ, Smyth GK. Edger: a Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics. 5. Rasmussen-Ivey CR, Figueras MJ, Mcgarey D, Liles MR. Virulence factors of 2010;26:139–40. Aeromonas hydrophila: in the wake of reclassification. Front Microbiol. 2016;7:1337. 6. Toma'S JM. The main Aeromonas pathogenic factors. ISRN Microbiol. 2012; 34. Yan MX, Dai WJ, Cai EP, Deng YZ, Chang CQ, Jiang ZD, Zhang LH. Transcriptome 2012:256261. analysis of Sporisorium scitamineum reveals critical environmental signals for 7. Mekalanos JJ. Environmental signals controlling expression of virulence fungal sexual mating and filamentous growth. BMC Genomics. 2016;17:354. determinants in bacteria. J Bacteriol. 1992;174:1–7. 35. Kanehisa M, Araki M, Goto S, Hattori M, Hirakawa M, Itoh M, Katayama T, 8. Sritharan M. Iron as a candidate in virulence and pathogenesis in mycobacteria Kawashima S, Okuda S, Tokimatsu T, Yamanishi Y. KEGG for linking genomes and other microorganisms. World J Microbiol Biotechnol. 2000;16:769–80. to life and the environment. Nucleic Acids Res. 2008;36:480–4. Teng et al. BMC Microbiology (2018) 18:52 Page 16 of 17 36. Isaacson T, Damasceno CM, Saravanan RS, He Y, Catalá C, Saladié M, Rose 60. Girija R. Iron and virulence in Francisella tularensis. Front Cell Infect JKC. Sample extraction techniques for enhanced proteomic analysis of plant Microbiol. 2017;7:107. tissues. Nat Protoc. 2006;1:769–74. 61. Skaar EP. The battle for Iron between bacterial pathogens and their 37. Smith PK, Krohn RIG, Hermanson G, Mallia AKFD, Gartner FJH, Provenzano vertebrate hosts. PLoS Pathog. 2010;6:e1000949. MD, Fujimoto EK, Goeke NM, Olson BJ, Klenk DC. Measurement of protein 62. Wrighting DM, Andrews NC. Iron homeostasis and erythropoiesis. Curr Top using Bicinchoninic acid. Anal Biochem. 1985;150:76–85. Dev Biol. 2008;82:141–67. 38. Candiano G, Bruschi M, Musante L, Santucci L, Ghiggeri GM, Carnemolla B, 63. Maltz M, Levarge BL, Graf J. Identification of iron and heme utilization genes Orecchia P, Zardi L, Righetti PG. Blue silver: a very sensitive colloidal in Aeromonas and their role in the colonization of the leech digestive tract. Coomassie G-250 staining for proteome analysis. Electrophoresis. 2004; Front Microbiol. 2015;6:763. 25:1327–33. 64. Hayrapetyan H, Siezen R, Abee T, Groot MN. Comparative genomics of Iron- 39. Wisniewski J, Zougman A, Nagaraj N, Mann M. Universal sample preparation transporting systems in Bacillus cereus strains and impact of Iron sources on method for proteome analysis. Nat Methods. 2009;6:359–62. growth and biofilm formation. Front Microbiol. 2016;7:842. 40. You C, Lin C. He H, et al. iTRAQ-based proteome profile analysis of superior 65. Neilands JB. Siderophores: structure and function of microbial iron transport and inferior Spikelets at early grain filling stage in japonica Rice. BMC Plant compounds. J Biol Chem. 1995;270:26723–6. Biol. 2017;17(1):100. 66. Long H, Zeng Y. Studies on resistance property of fish serum Transferrins 41. Shilov IV, Seymour SL, Patel AA, Loboda A, Tang WH, Keating SP, Hunter CL, against Aeromonas hydrophila. Journal of Hubei Agricultural College. Nuwaysir LM, Schaeffer DA. The paragon algorithm, a next generation 2004;24:119–23. search engine that uses sequence temperature values and feature 67. Gehring AM, Mori I, Walsh CT. Reconstitution and characterization of the probabilities to identify peptides from tandem mass spectra. Mol Cell Escherichia coli Enterobactin Synthetase from EntB, EntE, and EntF. Proteomics. 2007;6:1638–55. Biochemistry. 1998;37:2648–59. 42. Zhang MC, Cao YN, Yao B, Bai DQ, Zhou ZG. Characteristics of quenching 68. Neilands JB. Molecular aspects of regulation of high affinity iron absorption enzyme AiiO-AIO6 and its effect on Aeromonas hydrophila virulence factors in microorganisms. Adv Inorg Biochem. 1990;8:63–90. expression. J Fish China. 2011;35:1720–8. 69. Crosa JH, Walsh CT. Genetics and assembly line enzymology of Siderophore 43. Swift S, Lynch MJ, Fish L, Kirke DF, Tomás JM, Stewart GSAB, Williams P. biosynthesis in Bacteria. Microbiol Mol Biol Rev. 2002;66:223–49. Quorum sensing-dependent regulation and blockade of exoprotease 70. Franza T, Enard C, Van GF, Expert D. Genetic analysis of the Erwinia chrysanthemi production in Aeromonas hydrophila. Infect Immun. 1999;67:5192–9. 3937 chrysobactin iron-transport system: characterization of a gene cluster 44. Chu W, Zhou S, Zhu W, Zhuang X. Quorum quenching bacteria Bacillus sp. involved in uptake and biosynthetic pathways. Mol Microbiol. 1991;5:1319–29. QSI-1 protect zebrafish (Danio rerio) from Aeromonas hydrophila infection. 71. Bearson BL, Bearson SM, Uthe JJ, Dowd SE, Houghton JO, Lee I, Toscano MJ, Sci Rep. 2014;4:5446. Lay Jr DC. Iron regulated genes of Salmonella enterica serovar typhimurium 45. Gang L, Huang L, Su Y, Qin Y, Xu X, Zhao L, Yan Q. Flra, flrb and flrc regulate in response to norepinephrine and the requirement of fepDGC for adhesion by controlling the expression of critical virulence genes in Vibrio norepinephrine-enhanced growth. Microbes Infect. 2008;10:807–16. alginolyticus. Emerging Microbes Infect. 2016;5:e85. 72. Bou-Abdallah F. The iron redox and hydrolysis chemistry of the ferritins. 46. Tsou AM, Zhu J. Quorum sensing negatively regulates hemolysin transcriptionally Biochim Biophys Acta. 2010;1800:719–31. and posttranslationally in Vibrio cholerae. Infect Immun. 2010;78:461–7. 73. Hubálek M, Hernychová L, Havlasová J, Kasalová I, Neubauerová V, Stulík J, 47. Teng T, Liang LG, Chen K, Xi BW, Xie J, Xu P. Isolation, identification and Macela A, Lundqvist M, Larsson P. Towards proteome database of Francisella phenotypic and molecular characterization of pathogenic Vibrio vulnificus tularensis. J Chromatogr B. 2003;787:149–77. isolated from Litopenaeus vannamei. PLoS One. 2017;12:e0186135. 74. Hubálek M, Hernychová L, Brychta M, Lenco J, Zechovská J, Stulík J. 48. Saganuwan SA. A modified arithmetical method of reed and Muench for Comparative proteome analysis of cellular proteins extracted from highly determination of a relatively ideal median lethal dose (LD 50). Afr J Pharm virulent Francisella tularensis ssp. tularensis and less virulent F. tularensis ssp. Pharmacol. 2011;5:1543–6. holarctica and F. tularensis ssp. mediaasiatica. Proteomics. 2004;4:3048–60. 49. Liu W, Dong H, Li J, Ou Q, Lv Y, Wang X, Xiang Z, He Y, Wu Q. RNA-seq 75. Holden KM, Browning GF, Noormohammadi AH, Markham PF, Marenda MS. reveals the critical role of OtpR in regulating Brucella melitensis metabolism TonB is essential for virulence in avian pathogenic Escherichia coli.Comparative and virulence under acidic stress. Sci Rep. 2015;5:10864. Immunology Microbiology and Infectious Diseases. 2012;35:129–38. 50. Nagar V, Bandekar JR, Shashidhar R. Expression of virulence and stress 76. Dong Y, Liu J, Pang M, Du H, Wang N, Awan F, Lu C, Liu Y. Catecholamine- response genes in Aeromonas hydrophila under various stress conditions. J stimulated growth of Aeromonas hydrophila requires the TonB2 energy Basic Microbiol. 2016;56:1132–7. transduction system but is independent of the Amonabactin Siderophore. 51. Teixeiragomes AP, Cloeckaert A, Zygmunt MS. Characterization of heat, Front Cell Infect Microbiol. 2016;6:183. oxidative, and acid stress responses in Brucella melitensis. Infect Immun. 77. Angerer A, Gaisser S, Braun V. Nucleotide sequences of the sfuA, sfuB, and 2000;68:2954–61. sfuC genes of Serratia marcescens suggest a periplasmic-binding-protein- 52. Casabianca A, Orlandi C, Barbieri F, Sabatini L, Cesare AD, Sisti D, Pasquaroli S, dependent iron transport mechanism. J Bacteriol. 1990;172:572–8. Magnani M, Citterio B. Effect of starvation on survival and virulence expression of 78. Andrews NC. Iron homeostasis: insights from genetics and animal models. Aeromonas hydrophila from different sources. Arch Microbiol. 2015;197:431–8. Nat Rev Genet. 2000;1:208–17. 53. Payne SM, Lawlor KM. Chapter 11 – molecular studies on iron acquisition by 79. Rahman MH, Suzuki S, Kawai K. Formation of viable but non-culturable state non- Escherichia coli, species. Bacteria. 1990:225–48. (VBNC) of Aeromonas hydrophila, and its virulence in goldfish, Carassius auratus. Microbiol Res. 2001;156:103–6. 54. Allan BJ, Stevenson RM. Extracellular virulence factors of Aeromonas hydrophila in fish infections. Can J Microbiol. 1981;27:1114–22. 80. Maalej S, Gdoura R, Dukan S, Hammami A, Bouain A. Maintenance of 55. Lv J, Yu GC, Sun ZH, Wang NJ, Zhu Y, Wang HC, Sun XS. Proteomic analysis pathogenicity during entry into and resuscitation from viable but of effects of iron depletion on Streptococcus pyogenes MGAS5005. nonculturable state in Aeromonas hydrophila, exposed to natural seawater Microbiology China. 2012;39:515–25. at low temperature. J Appl Microbiol. 2004;97:557–65. 56. Bhatnagar, Elkins, Fortier. Heat stress alters the virulence of a rifampin-resistant 81. Mchugh JP, Rodríguezquinoñes F, Abdultehrani H, Svistunenko DA, Poole mutant of Francisella tularensis LVS. I nfect Immun 1995; 63:154–159. RK, Cooper CE, Andrews SC. Global iron-dependent gene regulation in 57. Ellis AE, Burrows AS, Stapleton KJ. Lack of relationship between virulence of Escherichia coli. A new mechanism for iron homeostasis J Biol Chem. 2003; Aeromonas salmonicida and the putative virulence factors: A-layer, extracellular 278:29478–86. proteases and extracellular haemolysins. J Fish Dis. 1988;11:309–23. 82. He X. Proteomic analysis of biological affects on Streptococcus pneumoniae induced by manganese depression. Jinan University 2011. 58. Brickman TJ, Cummings CA, Liew SY, Relman DA, Armstrong SK. 83. Cunningham AB, Sharp RR, Jr FC, Gerlach R. Effects of starvation on bacterial Transcriptional profiling of the iron starvation response in Bordetella pertussis transport through porous media. Adv Water Resour. 2007;30:1583–92. provides new insights into siderophore utilization and virulence gene expression. J Bacteriol. 2011;193:4798–812. 84. Elgaml A, Miyoshi SI. Regulation systems of protease and hemolysin 59. Sheikh MA, Taylor GL. Crystal structure of the Vibrio cholerae, ferric uptake production in Vibrio vulnificus. Microbiol Immunol. 2017;61:1–11. regulator (Fur) reveals insights into metal co-ordination. Mol Microbiol. 85. Tang TS, Lu CP. An Acinetobacter baumannii strain isolated from mandarin 2009;72:1208–20. fish possesses type 4 pili. J Nanjing Agric Univ. 1997;20:114–6. Teng et al. BMC Microbiology (2018) 18:52 Page 17 of 17 86. Chen XQ, Cai HY, Zhang W, Yan MY, Gao H, Duan GX, Yan ZY. The Role of Fur Involved in Bioflim Formation of Vibrio Cholerae. Prog Mod Biomed. 2013;13:2841–5. 87. Vasil ML, Ochsner UA. The response of Pseudomonas aeruginosa, to iron: genetics, biochemistry and virulence. Mol Microbiol. 1999;34:399–413. 88. Thune RL, Johnson MC, Graham TE, Amborski RL. Aeromonas hydrophila B- haemolysin: purification and examination of its role in virulence in 0-group channel catfish, Ictalurus punctatus (Rafinesque). J Fish Dis. 1986;9:55–61. 89. Zhu DL, Li AH, Wang JG, Li M, Cai TZ, Hu G. The correlation between the distribution pattern of virulence genes and the virulence of Aeromonas hydrophila strains. Acta Sci Nat Univ Sunyatseni. 2006;45:82–5. 90. Chu WH. Invasion mechanism and extracellular protease of Aeromonas hydrophila. Nanjing Agric Univ. 2002; 91. Cao Q, Zhang XY, Pang MD, Wang NN. Identification and molecular typing of the epidemic Aeromonas hydrophila strains in one farm of Nanjing. J Fish China. 2017;41:134–41. 92. Cascón A, Yugueros J, Temprano A, Sánchez M, Hernanz C, Luengo JM, Naharro GN. A major secreted elastase is essential for pathogenicity of Aeromonas hydrophila. Infect Immun. 2000;68:3233–41. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png BMC Microbiology Springer Journals

Comparative transcriptomic and proteomic analyses reveal upregulated expression of virulence and iron transport factors of Aeromonas hydrophila under iron limitation

Free
17 pages

Loading next page...
 
/lp/springer_journal/comparative-transcriptomic-and-proteomic-analyses-reveal-upregulated-tir4NM49Oy
Publisher
Springer Journals
Copyright
Copyright © 2018 by The Author(s).
Subject
Life Sciences; Microbiology; Biological Microscopy; Mycology; Parasitology; Virology; Life Sciences, general
eISSN
1471-2180
D.O.I.
10.1186/s12866-018-1178-8
Publisher site
See Article on Publisher Site

Abstract

Background: Iron plays important roles in the growth, reproduction and pathogenicity of Aeromonas hydrophila.In this study, we detected and compared the mRNA and protein expression profiles of A. hydrophila under normal and iron restricted medium with 200 μM 2,2-Dipyridyl using RNA Sequencing (RNA-seq) and isobaric tags for relative and absolute quantification (iTRAQ) analyses. Results: There were 1204 genes (601 up- and 603 down-regulated) and 236 proteins (90 up- and 146 down-regulated) shown to be differentially expressed, and 167 genes and proteins that showed consistent expression. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses revealed that the differentially expressed genes and proteins were mainly involved in iron ion transport, protein activity, energy metabolism and virulence processes. Further validation of the RNA-seq and iTRAQ results by quantitative real-time PCR (qPCR) revealed that 18 of the 20 selected genes were consistently expressed. The iron-ion absorption and concentration of A. hydrophila under iron-limited conditions were enhanced, and most virulence factors (protease activity, hemolytic activity, lipase activity, and swimming ability) were also increased. Artificial A. hydrophila infection caused higher mortality in cyprinid Megalobrama amblycephala under iron-limited conditions. Conclusion: Understanding the responses of pathogenic Aeromonas hydrophila within the hostile environment of the fish host, devoid of free iron, is important to reveal bacterial infection and pathogenesis. This study further confirmed the previous finding that iron-limitation efficiently enhanced the virulence of A. hydrophila using multi-omics analyses. We identified differentially expressed genes and proteins, related to enterobactin synthesis and virulence establishment, that play important roles in addressing iron scarcity. Keywords: Transcriptomic, Proteomic, Virulence, Iron, Aeromonas hydrophila Background and proteases, and it has the capacity to form biofilms and Aeromonas hydrophila is an opportunistic pathogenic alter metabolic pathways and gene expression under vari- bacterium that is ubiquitous in aquatic environments and ous host environments [5, 6]. Its virulence expression is causes serious infections worldwide in cultured fishes, am- closely related to the environment in which the bacteria live phibians, reptiles, and even mammals [1–4]. The pathogen- (in vivo and in vitro), nutrients, and so on [7]. For example, esis of A. hydrophila is multifactorial, causing disease with the nutrient iron deficiency in the host environment has virulence factors, such as adhesins, cytotoxins, hemolysins, been thoroughly documented as having a pronounced ef- fect on the virulence of pathogens [8]. Iron is an indispensable element of most living cells that is involved in many cellular functions, including electron * Correspondence: xiej@ffrc.cn; xup@ffrc.cn † transportation and oxygen transportation. The quantity Tao Teng and Bingwen Xi contributed equally to this work. of iron has a great impact on biological processes, for Wuxi Fisheries College, Nanjing Agricultural University, Wuxi 214081, China Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Teng et al. BMC Microbiology (2018) 18:52 Page 2 of 17 instance, iron overload will result in iron toxicity to cellu- 5 ml of normal TSB and incubated (28 °C, 24 h); bacteria lar components [9], especially for DNA damage, owing to cells were collected via centrifugation, washed three the reactions between hydroxyl radicals and other biomol- times with PBS, and then diluted to an optical density ecules [10, 11]. However, iron deficiency can also cause at 600 nm (OD 600) of 0.01 in 100 mL of normal TSB malnutrition cell death in severe cases [12]. In vivo, iron is to culture(180rpm,28°C). usually oxidized to an insoluble form due to its special physico-chemical properties, bonding with heme, ferritin, Sample collection hemoglobin, and transferrin within the cells, and thus is A. hydrophila NJ-35 cells (OD 600 ≅ 0.8) in normal and not readily accessible to bacteria [13]. In response to this iron-limited groups were collected by centrifugation iron deficiency predicament, microorganisms have evolved (5000 rpm, 4 °C, 10 mins). The pellet was rinsed twice a series of sophisticated mechanisms to compete against with saline and stored immediately at − 80 °C until further the host, such as the secretion of siderophores [14], to transcriptomic and proteomic analyses. The supernatant grab iron from transferrin, hemoglobin, and ferritin and was retained, filtered (MILLEX®GP filter unit, 0.22 μm), maintain iron dynamic balance for bacterial growth, pro- and frozen at − 20 °C, and it was used for the following liferation, and toxin secretion [15–17]. During the past de- proteolytic and hemolytic activity analyses. cades, the bacterial iron acquisition system and virulence have attracted much attention. For example, CaFTR1- Determination of iron concentration mediated iron-uptake was proven to be an important The atomic absorption spectrophotometry (GB/T 5009. virulence factor of Candida albicans [18], iron-responsive 90–2003) method [29] was used to the measure varia- transcriptional repressor PerR was required for full virulence tions in the intracellular iron of A. hydrophila NJ-35 in in Staphylococcus aureus [19], and FeoB was determined to normal and iron-limited groups, as well as the iron con- play an important role in Fe acquisition expression of viru- centration in the broth. Samples were analyzed by the lence of Helicobacter pylori [20]. Jiangsu Provincial Food Safety Testing Co., Ltd. Pathogenic bacteria virulence factors under iron- restricted growth conditions have previously been pub- Quantitative transcriptomics (RNA-seq) lished [21–24]. Proteomes and transcriptomes reflect (i) RNA isolation and mRNA purification gene expressions from two different levels, and their Total RNA was purified using an RNAqueous kit (Thermo joint analysis provides more complete expression infor- Fisher Scientific, San Jose, CA, USA) according to the mation about bacteria. Therefore, in this study, an iron manufacturer’s instructions. The RNA concentration and stress model was established to maximize the simulation integrity (RIN) were measured following the previous de- of iron deficiency environment in vivo, and the effects of scription of Wang et al. [30]. The mRNA was enriched iron-restricted stress on the growth and virulence of A. using a MICROBExpress Kit (Ambion, USA) [31], and de- hydrophila were evaluated comprehensively by combin- termined on Agilent 2100 Bioanalyzer. ing transcriptome and proteomics data. (ii) cDNA Synthesis, Illumina sequencing and library Methods construction Selection of iron chelator concentration and growth of A. Bacterial mRNA was fragmented using an RNA fragmenta- hydrophila tion kit (Illumina, San Diego, CA, USA). Double-stranded A. hydrophila (NJ-35) was isolated from dead cultured cDNA was synthesized using SuperScript II Reverse Tran- cyprinid in Jiangsu Province, China [25], and kindly pro- scriptase (Invitrogen, Carlsbad, CA) according to the man- vided by Professor Yongjie Liu from the College of Veter- ufacturer’s recommendations. Libraries were prepared with inary Medicine, Nanjing Agricultural University, P.R. the standard protocol of the TruSeq RNA Sample China. We selected 2,2’-Bipyridyl (Bip) (Sinopharm Chem- Prep v2 Low Throughput (LT) kit. Paired-end sequencing ical Reagent Co., Ltd., Shanghai, China) as the ferrous iron was processed by the Hiseq™2000 (Illumina, San Diego, chelating agent because of its high cell membrane perme- CA, USA) sequencer. ation and intracellular iron sequestering ability [26–28]. The accuracy and virulence of A. hydrophila NJ-35 were (iii) Bioinformatics Analyses confirmed by 16S rRNA gene sequencing (Biological The assembled reads were mapped to the complete gen- Engineering Technology Co., Shanghai, China) and lab in- ome of the A. hydrophila NJ-35 strain (http://www.ncbi. fection assays, respectively. Six concentrations (0, 100, 200, nlm.nih.gov/nuccore/CP006870.1). The QC of alignment 300, 400, and 500 μM Bip in normal tryptic soy broth was produced based on the standard generated by Qin medium (TSB; BD; final pH = 7.3)) were set to detect the et al. [31]. The gene expression level was calculated optimal concentration according to the growth curve of A. using the RPKM method (fragments per kb per million hydrophila NJ-35. A. hydrophila NJ-35 was inoculated in reads) [32]. Differentially expressed genes (DEGs) were Teng et al. BMC Microbiology (2018) 18:52 Page 3 of 17 identified with EdgeR software [33], and used to generate collected at a 4.5 min interval for 6–45 min, while the last statistical information such as expression level, fold segment was collected from 46 to 50 mins for a total change, p-value and FDR (false discovery rate). The spe- of 10 segments. Each segment was dried and used for cific filter conditions of DEGs were: log (fold change) ≥ 2, subsequent RPLC-MSMS analyses. p < 0.05 and bcv (biological coefficient of variation) = 0.01. GO enrichment analyses of DEGs were performed on RPLC-MSMS analyses website (http://www.geneontology.org/). The calculation In brief, samples were resuspended with Nano-RPLC buffer, method, p-value formula and enrichment score were ana- filtered through a C18 nanoLC trap column, and a lyzed according to the method reported by Yan et al. [34]. Chromxp C18 column (75 μm × 15 cm, C18, 3 μm120 Å). Additionally, the DEGs were subjected to KEGG en- The Eksigent nanoLC-Ultra™ 2D System (AB SCIEX) was richment analyses [35] to identify their main metabolic used to perform the online Nano-RPLC. Triple TOF 5600 pathways. The formula used for calculation was the system (AB SCIEX, USA) was used to analyze MS data same as that in the GO analyses. combined with Nanospray III source (AB SCIEX, USA). Quantitative proteomics (iTRAQ) (iiii) protein identification and quantification (i) Protein extraction, quantization, and SDS-PAGE Data were processed with the Protein Pilot Software v. 5. electrophoresis 0 (AB SCIEX, USA) against the NCBI database using the The extract of whole cellular protein was conducted ac- Paragon algorithm [41]. The results of protein quantifi- cording to Isaacson et al. [36] with some modification. cation were obtained by the matching of tandem mass The bacterial cells pellets were suspended in cooled spectrometry (MS) data and theoretical data, and was acetone (1 h, − 20 °C), centrifuged (15,000×g, 15 mins, performed with the search option: emphasis on bio- 4 °C), and dried with a vacuum freeze dryer. The sam- logical modifications. ples were resuspended in cold saturated-phenol (pH 7.5) An Orbitrap Elite high-resolution mass spectrometer and shaken (30 mins, 4 °C). The upper phenolic phase (Thermo Fisher Scientific, USA) was used for ITRAQ was collected by centrifugation (5000×g, 30 mins, 4 °C), quantitative proteomic analyses. Normalized high-energy 5 volumes of cold 0.1 M ammonium acetate in methanol collision dissociation (HCD) was performed, with the col- was added, and then it was stored (1 h, − 20 °C). After lision energy set at 30%. A protein database search and centrifugation (5000×g, 30 mins, 4 °C), the pellets were quantification were performed using Maxquant 1.5.1.0 washed and mixed with 2 volumes of ice-cold methanol. (Thermo Fisher Scientific, USA). The protein database The pellets were centrifuged, dried and dissolved in lysis contained 4119 proteins (https://www.ncbi.nlm.nih.gov/ solution (1 h, 30 °C). The supernatants were isolated by genome/?term=Aeromonas+hydrophila, GCF_000014805. centrifugation (15,000×g, 15 mins). The protein concen- 1_ASM1480v1_protein.faa). Oxidation (M) and acetyl trations were measured with the BCA method [37], after (protein N-term) were used as the variable modifications which they were stored at − 80 °C for iTRAQ analyses. and carbamidomethyl (C) was the fixed modification. The Additionally, 10 μg samples were subjected to 12% MS/MS tol. (FTMS) was 20 ppm. The protein quantita- SDS-PAGE, visualized and then scanned according to tion, peptides matching and the functional annotations of Candiano’sprotocol[38]. DEPs were performed according to the method reported by Yao et al. [24]. (ii) protein samples preparation and labeling The filter-aided sample preparation (FASP) method [39] Primer design, quantitative real-time PCR (qRT-PCR) was adopted for enzymatic hydrolysis of the proteins validation (100 μg). After 50 μL trypsin (50 ng/μL) digestion, pep- All of the sequence-specific primers of the target genes tides were labeled according to the manufacture’s protocol for qRT-PCR analyses were designed using Primer 5.0 for 8-plex iTRAQ reagent (AB SCIEX, USA). based on the obtained fragment (Table 3). The mRNA level of rpoB was used as an internal reference because (iii) 2D-LC-MSMS analyses of its stable expression according to Zhang et al. [42]. Total RNA from A. hydrophila was extracted using RPLC analyses RNAiso Plus (TaKaRa, Japan), and measured using a The dried samples were resuspended with 100 μL buffer Nanodrop 2000 (Thermo Fisher Scientific, USA), the A, after which reversed-phase liquid chromatography RNA concentration of each sample were diluted to (RPLC) was employed on an Agilent 1200 HPLC System 40 ng/μL, and then 2 μg of the total RNA was subjected (Agilent). Separation was conducted according to the to the following quantitative analysis with a One Step method of You et al. [40]. The first segment was collected SYBR® PrimeScript® Plus RT-PCR Kit (TaKaRa, Dalian). from 0 to 5 mins, after which each additional segment was Triplicate quantitative assays were performed on each Teng et al. BMC Microbiology (2018) 18:52 Page 4 of 17 type of cDNA using the ABI 7500 Real-time PCR System onto LB semisolid agar plates containing 0.3% agar (to (Applied Biosystems, Foster City, CA, USA) and ana- determine swimming ability) and 0.5% agar (to determine lyzed with the two-standard curve method. swarming motility). The LB plates were subsequently sealed with parafilm and incubated at 28 °C for 24 h (three Proteolytic activity parallel groups were set up for each group). At the Proteolytic activity was measured by an azocasein assay end of the culture period, the migration distance from method of Swift et al. [43] and Chu et al. [44], with some the colony edge to the colony center was determined. modifications. Briefly, 150 μL of normal group and iron- The experiment was repeated three times. limitation group NJ-35 culture supernatants were added to 1 ml of 0.3% azocasein (Sigma, St. Louis, USA) in 0. Infection assays in vivo 05 M Tris-HC1 and 0.5 mM CaCl (pH 7.5), then they A health check was conducted and healthy M. amblyce- were incubated (37 °C, 30 mins) respectively. Precooling phala (50 ± 5 g) were obtained from the Nanquan Ex- trichloroacetic acid (l0%, 0.5 ml) was then added to stop perimental Station of the Freshwater Fisheries Research the reaction, after which the samples were allowed to Center (Chinese Academy of Fishery Sciences, China) and stand for 15 mins at room temperature, then they were acclimatized in circulating water system with thermo- centrifuged (12,000 rpm, 10 mins, 4 °C) to remove the control for 2 weeks before use. Fish were given commercial precipitate. Next, 500 μL of the supernatants were added feed. The water temperature fluctuated between 27.5–28. to an equal volume of NaOH (1 mol/L). The supernatants 5 °C, with a pH between 7.2–7.8, and the DO was about 5. (200 μL) were subsequently transferred to a 96-well tissue 5mg/L. culture plate, after which the absorbance (OD400) of Strain NJ-35 was inoculated aseptically into normal the supernatant was measured. The proteolytic activity TSB medium and iron-limitation medium and then in- was calculated using the following equation: proteolytic cubated for 18 h at 28 °C while shaking at 180 rpm. The activity = OD sample – OD blank control artificial challenge experiment was performed as the pre- 400nm 400nm (normal TSB/iron limitation TSB). vious report [47]. To determine the 50% lethal dose (LD )[48], five groups of 20 M. amblycephala each Hemolytic activity were injected intraperitoneally with 150 μL of serial ten- 9 8 7 6 Hemolytic activity was determined as previously described fold diluted bacterial suspensions (1 × 10 ,10 ,10 ,10 , [45, 46], and sheep blood (Ping Rui Biotechnology, China) and 10 CFU·mL-1 measured by turbidimeter (Yue Fung was prepared by washing thrice with PBS. Washed sheep Instrument Co., Ltd., Shanghai, China)), which were blood (10 μL) was added to 490 μL of the experiment diluted with 0.9% saline. Next, an experimental group supernatants (sample), normal TSB/iron limitation TSB and a control group were injected intraperitoneally with (blank control), 1% (v/v) Trinton X-100 (positive control), 150 μL A. hydrophila (LD ) iron-limited and A. hydro- or PBS (phosphate buffer solution, negative control). After phila (LD ) basal, respectively, and the virulence was 30 mins of incubation at 37 °C, all of the samples were compared. Three replicate tanks per challenge isolate centrifuged (5000 rpm, 10 mins) at room temperature. (containing 20 fish each) were used to calculate survival The supernatants (200 μL) were then transferred to a 96- (from a total of 60 fish per isolate). The mortality of the fish well tissue culture plate, after which the absorbance of of experimental groups and control groups were monitored hemoglobin released for each solution at 540 nm was (7 days), and the activity and behavior were recorded daily; measured. The percentage of hemolysis was calculated pathogenic bacteria were isolated and identified from the using the following equation: hemolysis (%) = (OD lesion tissues of dead fish as the judging standard. 540nm sample - OD blank control)/ (OD positive con- 540nm 540nm trol Trinton X-100 - OD negative control PBS). Results 540nm Growth of A. hydrophila under different iron-limitation Lipase activity medium Bacterial cells were centrifuged and washed with PBS, after The effects of different concentrations of Bip on the growth which 5 μL of bacterial fluid was used to inoculate the LB of A. hydrophila are shown in Fig. 1.Whencompared with medium containing a 1% mass fraction of Tween 80. Sam- the control group, inhibitory effects were observed in the ples were then incubated at 28 °C for 24 h, after which they Bip addition groups, and higher Bip concentrations delayed were observed for lipase production, which was indicated the time of entering the logarithmic phase and reduced the by a white precipitate zone around the colony. maximum. When the Bip concentration was 500 μM, the growth of A. hydrophila was totally inhibited for at least Motility 24 h. Due to the significant inhibition and higher cells con- The target bacteria were centrifuged and washed with centration, 200 μMBip waschosenasthe proper iron- sterilized PBS. Next, 5 μL of bacterial fluid was dropped limitation concentration for subsequent analyses. Teng et al. BMC Microbiology (2018) 18:52 Page 5 of 17 reflecting significant changes and showing a strong correl- ation between the transcripts and proteins. Overall, 680 transcriptomes showed DEGs with no difference in pro- teins, while 35 transcriptomes showed different proteins but no difference in genes. Conversely, the expression of the following six genes and proteins was opposite (e.g., when the gene was upregulated, the protein was downregulated and vice versa): (U876_04575, YP_ 857861.1), (U876_17130, YP_855747.1), (U876_17135, YP_855746.1), (U876_19295, YP_855421.1), (U876_ Fig. 1 Effect of Bip supplementation on A. hydrophila growth. 20135, YP_855265.1), and (U876_21295, YP_855025.1). Growth curve (OD )of A. hydrophila NJ-35 grown in TSB medium This exception can be caused by regulation at several in the presence of 0, 100, 200, 300, 400, and 500 μM Bip levels, such as post transcriptional processing, degradation of the transcript, translation, post-translational processing Expression profile of iron-limited A. hydrophila and modification. In summary, most of the trends in DEP Based on the transcripts of A. hydrophila, 4327 genes abundance were consistent with the DEG data. were identified and quantified (Table 1). After filtering with FDR, 1204 genes were found to be differentially Functional classification of enriched DEGs and DEPs by expressed between the control and iron-limitation GO and KEGG groups. Detailed information for most of the DEGs is GO enrichment analyses were used to classify the enriched shown in Table 2. In comparison, the quantity of down- DEGs and DEPs between the control and iron-limitation regulated DEGs detected (603) was greater than that of groups using bioinformatics methods, and the results are the up-regulated genes (601). A total of 2244 proteins listed in Additional file 2: Excel S2 and Additional file 3: were identified; 2012 were quantified and 1946 were Excel S3, respectively. As shown in Fig. 3,the following correlated with the transcripts. Additionally, while com- three ontologies (molecular function, cellular component pared with the control group, a total of 236 DEPs (90 and biological process) were observed. up-regulated and 146 down-regulated) were identified in DEGs were distributed in up to 1460 GO terms, while the iron-limitation groups with an at least 2-fold differ- DEPs were classified into 402 GO terms. In this case, ence, and 167 of the DEPs were correlated to the corre- GO terms related to bacteria energy metabolism, iron sponding DEGs, which have the same trends. Fewer ion transport, and virulence. Based on the ‘−log Pvalue’, DEPs are probably due to the removal of some proteins most of the GO terms in the biological process category that were secreted by A. hydrophila NJ-35 in the super- were associated with energy metabolism (Fig. 3a and b). natant of the experimental design. Additionally, six genes were categorized as ‘glycerol cata- bolic process’ (GO: 0019563), three as ‘propionate cata- Integration analyses of transcriptome and proteome bolic process, 2-methylcitrate cycle’ (GO: 0019629), five To identify robust pathways that were corroborated by both as ‘oxidative phosphorylation’ (GO: 0006119), and five as datasets, we integrated the differentially expressed transcripts ‘respiratory electron transport chain’ (GO: 0022904). Re- and proteins to find the corresponding genes and proteins, garding proteomics, DEPs were mainly involved in the syn- and the results are listed in Additional file 1:Excel S1. thesis and transport of iron ions and proteins, particularly The distribution of the corresponding mRNA: protein the following GO terms: ‘iron assimilation’ (GO: 0033212), ratios is shown in a scatterplot of the log -transformed ‘ion transport’ (GO: 0006811), ‘enterobactin biosynthetic ratios. As shown in Fig. 2, almost all of the log mRNA: process’ (GO: 0009239), ‘protein secretion’ (GO: 0009306), log protein ratios are concentrated at the center of the ‘protein transport’ (GO: 0015031), and ‘electron transport plot, where mRNA and protein levels did not vary above chain’ (GO: 0022900). 2-fold. Integration analyses of transcriptome and proteome In the cellular component category (Fig. 3a and b), data revealed that 67 genes and their corresponding pro- three genes were categorized as ‘glycerol-3-phosphate teins were up-regulated, while 94 were down-regulated, dehydrogenase complex’ (GO: 0009331), five as ‘proton- Table 1 Overall features of the iron-limitation responsive expression profile Group name Type Number of genes Number of proteins Number of correlations Control-VS-Iron-Limitation Identification 4327 2244 1946 Control-VS-Iron-Limitation Quantitation 4327 2012 1733 Control-VS-Iron-Limitation Differential Expression 1204 236 167 Teng et al. BMC Microbiology (2018) 18:52 Page 6 of 17 Table 2 List of differentially expressed genes under iron restriction Accession Description Log FC U876_09860 Biosynthesis of siderophore group nonribosomal peptides 9.3945 U876_18585 ABC transporters 7.3209 U876_18590 ABC transporters 7.2995 U876_11875 Propanoate metabolism 3.6179 U876_18275 Two-component system|Bacterial chemotaxis 2.7194 U876_05565 Carbon metabolism|Glycolysis / Gluconeogenesis|Citrate cycle (TCA cycle)|Pyruvate metabolism|Butanoate metabolism|Carbon 2.3607 fixation pathways in prokaryotes U876_16675 Quorum sensing 2.3034 U876_14615 Oxidative phosphorylation|Two-component system 2.1593 U876_13185 Ribosome 1.7769 U876_13000 Cysteine and methionine metabolism|Selenocompound metabolism 1.5614 U876_17160 RNA transport 1.4601 U876_00445 Glycine, serine and threonine metabolism −1.5113 U876_10020 Purine metabolism|Drug metabolism - other enzymes −1.5726 U876_15390 Biotin metabolism −2.2592 U876_09705 Selenocompound metabolism|Aminoacyl-tRNA biosynthesis −3.4550 U876_00975 Biosynthesis of amino acids|Arginine biosynthesis −3.5694 U876_17185 Lysine degradation|Tropane, piperidine and pyridine alkaloid biosynthesis −3.8646 U876_15985 Fructose and mannose metabolism|Phosphotransferase system (PTS) −4.5035 U876_12875 Nitrogen metabolism −5.4990 U876_00965 Arginine biosynthesis −6.0546 Note: FC, Fold change, the ratio of different expression levels between the iron-limitation group and the normal TSB group transporting ATP synthase complex, catalytic core F(1)’ Enriched KEGG terms are listed under Additional file 4: (GO: 0045261), four as ‘proton-transporting ATP syn- Excel S4 and Additional file 5: Excel S5, as transcripto- thase complex, coupling factor F(o)’ (GO: 0045263), and mics and proteomics, respectively. When compared with seven as ‘bacterial-type flagellum hook’ (GO: 0009424). the whole genome, a total of 624 genes were present in Regarding proteomics, DEPs were mainly classified in the the 139 KEGG pathways as DEGs, and we selected the cell membrane and cytoplasm of GO terms, including ‘inte- 20 most critical KEGG pathways according to the enrich- gral component of membrane’ (GO: 0016021), ‘plasma mem- ment scores (Fig. 4a). The up-regulated KEGG pathways brane’ (GO: 0005886), ‘cell outer membrane’ (GO: 0009279), included 78 genes under the category of ‘ABC trans- ‘cytosol’ (GO: 0005829), and ‘cytoplasm’ (GO: 0005737). porters’ (ko02010), 20 genes under ‘TCA cycle’ (ko00020), In the molecular function category (Fig. 3a and b), 11 and 38 genes under ‘quorum sensing’ (ko02024). We in- genes were categorized as ‘receptor activity’ (GO: 0004872), ferred that transport, energy production and bacteria three as ‘energy transducer activity’ (GO: 0031992), three as interact with each other and may play important roles ‘cytochrome o ubiquinol oxidase activity’ (GO: 0008827), via stress responses that are regulated through several four as ‘siderophore uptake transmembrane transporter ac- pathways. The down-regulated KEGG pathways in- tivity’ (GO: 0015344), and three as ‘siderophore transmem- cluded 47 genes categorized as ‘Ribosome’ (ko03010), brane transporter activity’ (GO: 0015343). Regarding 71 as ‘Carbon metabolism’ (ko01200), 31 as ‘Pyruvate proteomics, DEPs were mainly related to protein activity metabolism’ (ko00620), and 35 genes as ‘Oxidative and binding capacity, including ‘siderophore transmem- phosphorylation’ (ko00190), which confirmed that bac- brane transporter activity’ (GO: 0015343), ‘receptor activity’ teria slowed down material synthesis and life activities. (GO: 0004872), ‘iron ion binding’ (GO: 0005506), ‘heme With respect to proteomics, a total of 41 proteins were binding’ (GO: 0020037), ‘metal ion binding’ (GO: 0046872), detected in the 34 KEGG pathways by DEP, while only and ‘porin activity’ (GO: 0015288). In summary, GO term eight pathways were found to be significantly enriched by enrichment analyses further explained that metabolism, filtration (Fig. 4b). The up-regulated KEGG pathways in- biosynthesis, transmembrane transport and redox homeo- cluded three that were labeled under ‘biosynthesis of sid- stasis should be tightly regulated. erophore group nonribosomal peptides’ (aha01053) and Teng et al. BMC Microbiology (2018) 18:52 Page 7 of 17 Fig. 2 Relationship patterns of all of the quantitative mRNA and protein. In the nine-quadrant diagram, the abscissa is the protein expression and the ordinate is the gene expression. Each color denotes a log mRNA ratio and a log protein ratio. Gray (filtered) represents genes and proteins 2 2 with no significant difference, red (Cor_up) indicates up-regulated genes and proteins, green (Cor_down) indicates down-regulated genes and proteins, purple (Opposite_Sig) indicates that DEGs and DEPs show opposite up- and down- regulation and blue (Single_Sig) indicates that one of the genes and proteins differ 10 that were labeled under ‘ABC transporters’ (aha02010), and generation of hemolysin (U876_04005, U876_15265, indicating clear changes in synthesis and transportation of U876_16300, and U876_16315). Heat map analyses siderophores. The down-regulated KEGG pathways in- (Fig. 5) were used to visualize genes and proteins, and cluded 11 proteins that were classified as ‘oxidative phos- the results indicated a comprehensive impact and clear phorytation’ (aha00190), six as ‘butanoate metabolism’ changes in the regulation of virulence factors. (aha00650), five proteins as ‘TCA cycle’ (aha00020), five as ‘pyruvate metabolism’ (aha00620), seven as ‘carbon Validation of selected DEGs/DEPs by qRT-PCR analyses metabolism’ (aha01200), and six as ‘two-component sys- To further evaluate the expression of genes in an iron- tem’ (aha02020), indicating the bacteria repress energy limited environment, 20 virulence genes (13 up-regulated metabolize to adaptive constraint environment. Conversely, and seven down-regulated genes) together with reference the total number of DEPs among them was far smaller genes (rpoB) were selected for investigation based on their than that of the DEGs, and most DEGs and DEPs were expressions, which were measured by real-time quantitative down-regulated. PCR (RT-qPCR) (Table 3) according to the results of the GO analyses. These selected genes were involved in Clustering of virulence genes and proteins in A. hydrophila virulence factors, hemolysis, secretion systems, lipases, in iron-limited medium phospholipids, serine-type peptidases, metallopeptidases, According to the bioinformatics analyses, we found that flagella, polysaccharides, siderophore transporters, quorum there were 60 virulence factors in the differential genes, sensing, and outer membrane production. which mainly fell under the category of synthesis of iron The results of qPCR showed that the majority of the carriers (U876_01620, U876_18555, U876_21285, U876_ selected virulence factors (90%, 18/20) were consistent 21455, U876_23515, and U876_24445), motility of flagella with the transcriptome data. Notably, five virulence- (U876_20435, U876_07265, U876_07270, and U876_07305), related factors, U876_15265 (hemolysin, log FC = 3.80), 2 Teng et al. BMC Microbiology (2018) 18:52 Page 8 of 17 Fig. 3 GO enrichment analyses of DEGs and DEPs Control group vs Iron-Limitation group. GO term analyses of transcriptomics (a) and proteomics (b) that were catalogued as Biological Process, Cellular Component, and Molecular Function U876_15575 (secretin, log FC = 5.00), U876_18585 (hemin log FC = 3.33), and U876_09860 (2,3-dihydroxybenzoate- 2 2 ABC transporter substrate-binding protein, log FC = 4.71), AMP ligase, log FC = 6.20) were shown to be significantly 2 2 U876_20975 (transcriptional activator protein AhyR/AsaR, up-regulated (log FC > 3.00) under iron-limited conditions. 2 Teng et al. BMC Microbiology (2018) 18:52 Page 9 of 17 Fig. 4 KEGG enrichment analyses of DEGs and DEPs Control group vs Iron-Limitation group. KEGG enrichment analyses of transcriptomics (a) and proteomics (b) Teng et al. BMC Microbiology (2018) 18:52 Page 10 of 17 Fig. 5 Clustering of 60 mainly related virulence genes and proteins. Numbers are listed as the log value of difference multiples. Expression differences are shown in different colors; red indicates up-regulation, while green indicates down-regulation. A heatmap was used to visualize the genes and proteins that were related to virulence factor (hemolysis, secretion system, lipase, phospholipid, serine-type peptidase, metallopeptidase, flagellum, polysaccharides, siderophore transporter, quorum sensing, and outer membrane) Moreover, two selected genes, U876_07270 (flagellar hook concentration of 0.664 mg/100 g in the normal TSB protein FlgE) and U876_12225 (murein transglycosy- group strain cell was lower than 0.998 mg/100 g in lase A), showed appositive results to the RNA-seq the iron-limitation group strain cell. All of the results data, which might have been due to differences in the are shown in Table 4. analyses methods. Effect of iron-limitation on virulence factors production in Determination of iron concentration A. hydrophila Atomic absorption spectrophotometry revealed that the As shown in Table 5, the total protease activity in super- medium iron concentration of 0.44 mg/100 g in the normal natants from A. hydrophila NJ-35 growing without Bip TSB group was higher than 0.28 mg/100 g in the iron- was 0.105 (OD400 nm), whereas the presence of Bip re- limitation group, indicating that iron scavenger 2,2- bipyri- sulted in a significant increase in protease activity to 0.36 dine has a higher efficiency. After bacterial growth, the (OD400 nm) (Fig. 6a). When compared with the control medium iron content of the normal TSB group was higher group, the hemolytic activity of A. hydrophila NJ-35 was than that of the iron-limitation group. Surprisingly, the significantly enhanced under iron limitation, indicating Teng et al. BMC Microbiology (2018) 18:52 Page 11 of 17 Table 3 Primers and sequences used in this study for q-PCR Name Gene product Primer Sequence (5′➔3′) qRT-PCR Illumina FC FC Log Regulated Log Regulated 2 2 U876_04005 RTX toxin F GCCAAGAACCTGACCTAC 0.78 Up 1.06 Up R TAACTACCGTCCGACCAT U876_15265 hemolysin F TGCTCGTACTTGCTGTTG 3.85 Up 3.80 Up R GACTACCTGCTGCTGGAT U876_15575 secretin F CGATGCGTACCGATATGT 5.00 Up 5.33 Up R AGACTAACAACCAGGATGAG U876_16325 type I secretion system F GCTCATCGCCTCAATACC 1.42 Up 1.02 Up permease/ATPase R TAGCCAGTGTGAGTCAGG U876_02495 phosphatidylcholine-sterol F TTCGGTGTTCCAGCCATA 2.34 Up 1.87 Up acyltransferase R CCAAGTATCAGGTCATCAAC U876_00180 lysophospholipase L2 F AGCACATAATCGTCAAACTG 1.23 Up 1.51 Up R GCCATCCTCATCGTCAAC U876_12850 FAD/NAD(P)-binding F CGATTACCACAAGATTGACC 2.34 Down −5.25 Down oxidoreductase R TGATCCAGCAGCACTATG U876_06295 HPr family phosphocarrier F CGGAGACCACAGTGATCT −0.86 Down −2.07 Down protein R TGTACGAGAAGTCTGTTGTT U876_06300 cysteine synthase A F CAGAGCAATACCCGTGTT 1.69 Up 1.99 Up R TCAACCGTGTTACCAAGG U876_14760 peptidase T F CCGAGGATCAAACCCATTC −1.23 Down −6.23 Down R CTTGCCGTGGAAGTTGTG U876_07265 flagellar hook capping protein F CAATGTCGGTTACCTGGAA −0.86 Down −1.05 Down R GTCCTTGTCCTTGCCATC U876_07270 flagellar hook protein FlgE F TCAGCGACCTACAGCAAT 0.25 Up −1.25 Down R CACCAGACAGCAGAGACT U876_12225 murein transglycosylase A F CCAGACTGATGCCGTAAC 1.25 Up −1.16 Down R CAAGATGACTCGTCGCTAC U876_03850 PAP2 family protein F GATGGTGCCGTTGTTCTC 2.54 Up 2.07 Up R ACAGCAGTGGTAGACAGAG U876_17510 outer membrane protein F GGTGAGTGGAACGGTTAC −0.99 Down −2.14 Down R ATCGGAGTGCCAGTAGATA U876_18585 hemin ABC transporter F CGATCTGGTGCTGGTTAG 4.71 Up 7.32 Up substrate-binding protein R CTTGATCCACTTGGCGAT U876_21455 TonB-dependent siderophore F CGTCTCAGTCACCAGTCT 2.61 Up 6.07 Up receptor R ATCCAGGTTGTTGTTCTTGT U876_20975 transcriptional activator F TTGAACAGCACCACCTTG 3.33 Up 1.22 Up protein AhyR/AsaR R GCTTGAGTACCTCGAACAT U876_23540 LuxR family transcriptional F GAAGGAGTGCCTGTTCTG 1.14 Up 1.45 Up regulator R TATGATGCCGCTGGAGAT U876_09860 2,3-dihydroxybenzoate-AMP F TACAGGATGCCGATGGTTA 6.20 Up 9.23 Up ligase R ATCCGTGCTGACGATGAA U876_01300 DNA-directed RNA F GGATCACGGTGCCTACAT (rpoB) polymerase subunit beta R TAACGCTCGGAAGAGAAGA Teng et al. BMC Microbiology (2018) 18:52 Page 12 of 17 Table 4 Determination of iron concentration under two culture bacterial multiplication was enhanced after injecting ex- conditions ogenous iron into experimentally infected animals, and the Group Medium before Medium after Strain cell/ virulence of pathogens including Vibrio cholerae, Pseudo- culture/(mg/100 g) culture/(mg/100 g) (mg/100 g) monas aeruginosa, Klebsiella pneumoniae,and Mycobacter- b a Normal TSB 0.44 ± 0.032 0.27 ± 0.029 0.664 ± 0.019 ium tuberculosis was established with sufficient iron [8, group 53]. A. hydrophila establishes virulence through many a b Iron-Limitation 0.28 ± 0.021 0.20 ± 0.030 0.998 ± 0.012 mechanisms [54], including iron-binding systems, secre- group tion systems, biofilm formation, flagella and pili adhe- Note: Means with different lowercase letters within the same column were sion, structural proteins, phospholipids, polysaccharides, significantly different (P < 0.05) hemolysis, collagenase, serine protease, metalloprotease, that NJ-35 produced 83.8% more hemolysin (Fig. 6b). To enolase, lipase, and nucleases [5, 6]. The pathogenesis of observe the hemolysis ability, sheep blood agar plates were diseases involves most virulence factors [1], beginning with used for rough detection. A. hydrophila NJ-35 under iron molecular changes at the micro level and progressing to limitation generated a large hemolytic zone on the blood phenotypic changes at the macro level [55]. Under iron- agar plates compared to the control group, but the lipase limited conditions, virulence genes and proteins were up- activity and swarming motility did not differ significantly regulated more than down-regulated (Fig. 5), suggesting (Table 5). Interestingly, the swimming ability of the bac- that virulence expression was enhanced in A. hydrophila to teria was strong under iron limiting conditions, which compensate for iron insufficiency, which was confirmed in could reflect attempts to move to areas with more suitable F. tularensis [56]. These virulence factors exerted synergis- conditions (Table 5). tic effects [57] and contributed to the production of toxins. The results of the infection assays further confirmed this Infection assays conclusion (Fig. 7). Theferric uptakeregulator (Fur) is a The isolated pathogenic bacteria were A. hydrophila after negative regulator in iron acquisition systems [58]that con- morphological, physiological and biochemical, molecular trols the expression of 90 virulence and metabolic genes [7, identification. Megalobrama amblycephala injected with 15, 59]. For example, the biosynthesis of rhizoferring, an A. hydrophila NJ-35 showed distinct mortality rates under iron siderophore in F. tularensis, is regulated by operon iron and non-iron limited conditions (Fig. 7). Although fslABCDEF [60]. In this study, the expression of the fur the difference was not significant, the survival rate in the gene (U876_15170) was up-regulated (log FC = 0.3187). group injected with A. hydrophila was substantially higher This phenomenon could be explained by the higher iron (by 19.77%) than that of the iron-limitation group at four concentration in bacterial cells of the iron-limited group. days post-challenge. At the sampling time-point, more iron was stored in the iron-limited group, after which fur was up-regulated to re- Discussion duce the iron absorption [61]. Comparative transcriptomic and proteomic analyses Iron homeostasis was coordinated by the absorption, The survival and proliferation of bacteria was sensitive transport, utilization, and storage of iron ions [62]. A. to environment factors. Many environmental stress fac- hydrophila utilized multiple iron sequestration systems tors, e.g., pH, temperature, oxygen, acidity and salinity to hijack host iron ions [63]. Under an iron deficient en- [49, 50] significantly affected the expression of virulence. vironment, A. hydrophila secreted large amounts of iron Iron limitation is an important external stimulus [51] transporters and iron-specific scavenger-siderophores. that has profound impacts on almost all bacteria. The The same results were confirmed by transcriptome ana- culturability and growth rate of A. hydrophila were lyses of Bacillus cereus ATCC 10987, which showed the reduced under iron-limited conditions [52]; however, upregulation of predicted iron transporters in the pres- ence of 2,2-Bipyridine [64]. As an important virulence characteristic of pathogens to both animals and plants Table 5 Effect of iron limitation on A. hydrophila extracellular [65], siderophores were formed and played a major role enzyme activity and motility in microbial iron acquisition. Siderophore-assisted iron Virulence NJ-35 uptake and reductive iron assimilation are both induced Control Iron-Limitation upon iron starvation [58]. In previous studies, A. hydro- Lipase (cm) 0.95 ± 0.16 1.06 ± 0.21 phila was found to secrete siderophores to compete with Blood-plate hemolysis (cm) 0.90 ± 0.07 1.07 ± 0.09 transferrin in vivo to meet the iron demand required for a b Swimming ability (cm) 1.02 ± 0.01 1.17 ± 0.01 growth and virulence [66]. Measurement of the iron b a concentration confirmed that the iron chelating ability of Swarming motility (cm) 0.89 ± 0.12 0.84 ± 0.07 bacterial siderophores was notable (Table 4), because E. coli Note: Means with different lowercase letters within the same row were significantly different (P < 0.05) [67]and A. hydrophila synthesize and secrete enterobactin Teng et al. BMC Microbiology (2018) 18:52 Page 13 of 17 Fig. 6 Effect of control and iron-limitation conditions on A. hydrophila NJ-35. a Total protease, and (b) hemolytic activity. The data represent the mean values of three independent experiments and are presented as the means ± SD siderophores [68] in response to iron starvation. Enterobac- iron storage protein in A. hydrophila [72]. Thedatadem- tin synthase subunit E (entE), which is encoded by entA, onstrated that ferritin (U876_00270, log FC = − 0.4088) entB, and entC genes, is a key enzyme involved in the syn- and bacterioferritin (U876_02285, log FC = 13.4043) partic- thesis of isochorismate synthase. In both E. coli and A. ipated in iron ion transport and storage, which may benefit hydrophila, a 22 kB gene cluster including entD-fepA- the survival of bacteria. The up-regulation of this protein fes-entD-fepE-fepC-fepG-fepD-fepB-entC-entE-entB-entA- may be responsible for the increased intracellular iron ybdA genes encodes proteins responsible for the synthesis concentration in A. hydrophila. The expression levels and transport of enterobactin [69]. During this process, the of bacterioferritin in different isolates, including F. entE polypeptide is responsible for activating the DHBA tularensis, also varied [73, 74]. The TonB mechanism carboxylate group with ATP by forming the enzyme-bound is essential to the virulence of avian pathogenic E. coli 2,3-dihydroxybenzoyadenylate as an intermediary in the [75], indicating that a specific TonB-dependent outer biosynthetic pathway [70]. Genes with similar enterobactin membrane receptor might be involved in the transport of transport functions (iroN, fepC, cirA, fepC, and iroC) were iron from transferrin [76]. TonB-dependent outer mem- also found in Salmonella enterica [71]. After differential brane receptors TonB-2 (U876_00270, log FC = 5.9844), analyses of the genes and proteins, we found that the entE AHA_4249 (YP_858666.1, log FC = 6.1718), AHA_4250 expression level of gene U876_09860 (log FC = 9.39) and (YP_858667.1, log FC = 7.4891), and AHA_4251 (YP_ 2 2 protein YP_856992.1 (log FC = 15.46) had increased signifi- 858668.1, log FC = 10.7778) were found to be required for 2 2 cantly during the biosynthesis of the siderophore subunits the transfer of iron chelators and heme to the periplasm, (ko01053). Upon GO term analyses of the DEGs, the entE followed by transport to the cytoplasm by ATP-binding gene and protein expression levels were not increased sig- cassette (ABC)-type transporters. Inorganic iron in the nificantly, which may have been because of differences in periplasm is transported to the cytoplasm by membrane the analyses methods and software. Ferritin is the major transporters, such as Sfu ABC [77]. Iron influences a number of catalytic reactions involv- ing cell energy metabolism in vivo, including respiration and nucleic acid replication [78]. Overall, when iron de- mand is not met, some enzymes related to metabolism, the regulation of protein synthesis, and the ability of A. hydrophila to utilize nutrients, such as carbohydrates, decreased. It has been hypothesized that decreased viru- lence might be caused by the loss of metabolic activity and the lack of toxin production [79, 80]. According to bioinformatic analyses conducted in this study, the energy generation system and electron respiration chain appeared to be depressed under iron starvation, which is consistent with previous quantitative proteomic analyses of A. hydro- phila [24]. Iron scarcity reduces iron utilization in iron Fig. 7 Kaplan-Meier survival analyses of Megalobrama amblycephala nonessential pathways, and limited iron is used for the syn- challenged with A. hydrophila NJ-35 from normal and iron-limited media. thesis of iron-containing enzymes involved in the citric acid Data represent accumulative fish mortality in three replicates cycle and the electron transport chain [81]. For example, Teng et al. BMC Microbiology (2018) 18:52 Page 14 of 17 the expression of NADP-dependent glyceraldehyde-3- isolated from protease deficient strains of A. hydrophila phosphate significantly altered the antioxidant activity were found to lead to the death of catfish [88]. Blood-plate of bacteria, and NADPH is involved in the transform- hemolysis results showed that the hemolytic ability of A. 3+ 2+ ation of Fe into Fe in some of the identified hydrophila under iron deficiency was stronger than that bacteria [16]. Similar to S. pneumoniae in manganese under normal conditions, and it caused greater toxicity limited environments [82], the metabolic activity of and damage to the host. The results also showed that iron bacteria will become inert, so bacteria can survive in these exerted an inhibitory effect on extracellular hemolysin and environments for a long time [83]. Based on the high- protease activity. Notability, the presence of hemolysin throughput data analyses, it is apparent that 969 genes de- alone does not cause disease [89]. creased, 905 genes increased, 146 proteins decreased, 90 The invasion of pathogenic bacteria was found to be proteins increased, the gene and protein ratio was down- significantly correlated with the level of corresponding regulated, and the regulation of bacteria itself was also enzyme production, and protease activity [90], which is used to interpret iron starvation. consistent with the results of this trial. Not only can pro- teases degrade a variety of proteins to provide amino Virulence evaluation of A. hydrophila under iron-limited acids for bacterial survival and growth, but they can also environment directly cause tissue injury, resulting in the spread Many studies have shown that the virulence of A. hydro- through the defense mechanism and evasion of the im- phila increased in response to iron deficiency [52]. Two mune system of the host [91]. In addition, the A. hydro- aspects may contribute to the establishment of bacterial phila family of extracellular proteases can cooperate pathogenicity: invasiveness and toxin production [84]. with other virulence factors [92] to activate other patho- The invasive ability of A. hydrophila is closely related to genic factors. In this study, A. hydrophila NJ-35 under low- their motility, as well as the secretion of toxins, including iron growth conditions were detected with higher protease aerolysin, hemolysin, and enterotoxin, and extracellular activity than the control, demonstrating that iron scarcity protease. To evaluate the virulence of A. hydrophila more can promote NJ-35 virulence factor expression. comprehensively, we conducted an encompassing study of A. hydrophila hemolytic and enzymatic activity in vitro Conclusion and lethality rate in vivo. In this paper, we simulated the iron restriction environ- A. hydrophila pilus is an important coagulation factor ment in the fish host, coalition analyzed the transcriptome and a major colonization factor that enables bacteria to and proteomics data of A. hydrophila, and identified the adhere to host digestive epithelial cells during the invasion changes of enzyme activity, comprehensively revealed the process. In terms of virulence establishment, pili-assisted pathogenicity of A. hydrophila increased. This study also adhesion bacteria were 10–200 times more effective than provide a profound theoretical basis for the effect of ex- bacteria that do not express pili [85]. Flagella-mediated ogenous iron preparation on the toxicity of bacteria. motility also promotes the initial stages of adhesion [86]. In this study, although the swimming ability of the control Additional files group was significantly stronger than that of the iron re- striction group, swimming ability was enhanced under Additional file 1: Excel S1. The results of differentially expressed transcripts iron-limited conditions, indicating that A. hydrophila can andproteinsto findthe correspondinggenes andproteins. (XLSX87kb) overcome unfavorable conditions by accelerating their Additional file 2: Excel S2. The enriched DEGs GO terms between control and iron-limitation groups using bioinformatics methods. (XLSX 42 kb) swimming and adhesion abilities, thereby enhancing Additional file 3: Excel S3. The enriched DEPs GO terms between control their resilience to environmental restraints. Alterna- and iron-limitation groups using bioinformatics methods. (XLSX 42 kb) tively, these findings demonstrate the complexity of Additional file 4: Excel S4. The enriched DEGs KEGG terms between Aeromonas sp. virulence. control and iron-limitation groups using bioinformatics methods. (XLSX 25 kb) Lethal pathogenic extracellular products (ECPs) of A. Additional file 5: Excel S5. The enriched DEPs KEGG terms between hydrophila are produced to compete with rivals for lim- control and iron-limitation groups using bioinformatics methods. (XLSX 13 kb) ited iron resources [87]. After removal of ECPs by re- peated washing with normal saline, the invasion and Abbreviations pathogenicity of the pathogenic bacteria to the host cells ABC: ATPbinding cassette; Bip: 2,2’-Bipyridyl; BLAST: Basic Local Alignment Search Tool; CA: Citric acid; CID: Collision-induced dissociation; was reduced or even completely lost. As a typical ECP, DEGs: Differentially expressed genes; DEPs: Differentially expressed proteins; hemolysin that is synthesized and secreted into the organ- ECP: Extracellular products; FDR: False discovery rate; GO: Gene Ontology; ism’s environment can dissolve various sources of iron by HCD: High-energy collision dissociation; IDA: Information dependent acquisition; iTRAQ: Isobaric tags for relative and absolute quantification; destroying intracellular red blood cells or hydrolyzing KEGG: Kyoto Encyclopedia of Genes and Genomes; LC MS/MS: Liquid hemoglobin. Hemolytic activity was detected both in vivo chromatography tandem mass spectrometry; LD : 50% lethal dose; and in vitro in septic animals, and beta hemolysins MS: Mass spectrometry; NCBI: National Centre for Biotechnology Information; Teng et al. BMC Microbiology (2018) 18:52 Page 15 of 17 PBS: Phosphate buffer solution; PRIDE: PRoteomics IDEntifications; QC: Quality 9. Halliwell B, Gutteridge JM. Oxygen toxicity, oxygen radicals, transition control; qPCR: Real-time Quantitative polymerase chain reaction; RIN: RNA metals and disease. Biochem J. 1984;219:1–14. integrity value; RNA-seq: RNA Sequencing; SRA: Sequence Read Archive; 10. Braun V. Avoidance of iron toxicity through regulation of bacterial iron TCA: Tricarboxylic acid; TSB: Tryptic soy broth transport. Biol Chem. 1997;378:779–86. 11. Miller RA, Britigan BE. Role of oxidants in microbial pathophysiology. Clin Microbiol Rev. 1997;10:1–18. Funding This study was supported by The earmarked fund for China Agriculture Research 12. Zhang YC, Shen YY, Yan XH, Wang FD. Molecular mechanisms of System (CARS-45), Natural Science Foundation of China (31572662), Jiangsu Natural mammalian iron homeostasis. Chin. J Cell Biol. 2011;33:1179–90. Science Foundation (BK20171152), and Postgraduate Research & Practice Innovation 13. Teng T, Xi BW, Xie J, Chen K, Pao X, Pan LK. Molecular cloning and Program of Jiangsu Province (KYLX16_1082). The funding agencies have not been expression analysis of Megalobrama amblycephala transferrin gene and effects of involved in the design of research and collection, analysis and interpretation of data exposure to iron and infection with Aeromonas hydrophila. Fish Physiol Biochem. and in writing of manuscripts. We also thank OEbiotech.co.ltd for technical support 2017;43:987–97. in our transcriptome and proteomics analyses. 14. Telford JR, Raymond KN. Amonabactin: a family of novel siderophores from a pathogenic bacterium. J Biol Inorg Chem. 1997;2:750–61. 15. Litwin CM, Calderwood SB. Role of iron in regulation of virulence genes. Availability of data and materials Clin Microbiol Rev. 1993;6:137–49. The RNA-seq data and analyses discussed in this publication were deposited 16. Ratledge C, Dover LG. Iron metabolism in pathogenic bacteria. Annu Rev in the NCBI Sequence Read Archive (SRA) database under accession number Microbiol. 2003;54:881–941. SRR5894319. The mass spectrometry proteomics data have been deposited 17. Miethke M, Marahiel M. Siderophore-based iron acquisition and pathogen to the ProteomeXchange Consortium via the PRIDE partner repository with control. Microbiol Mol Biol Rev. 2007;71:413–51. the dataset identifier PXD007641. 18. Ramanan N, Wang Y. A high-affinity iron permease essential for candida albicans virulence. Science. 2000;288:1062–4. Authors’ contributions 19. Horsburgh MJ, Clements MO, Crossley H, Ingham E, Foster SJ. Perr controls BX, JX and PX conceived and designed the experiments; BX guided the oxidative stress resistance and iron storage proteins and is required for experiments, TT performed the experiments, analyzed the data; TT and BX wrote virulence in staphylococcus aureus. Infect Immun. 2001;69:3744. the paper, revised the paper, they contributed equally to this work; LP, KC 20. Velayudhan J, Hughes NJ, Mccolm AA, Bagshaw J, Clayton CL, Andrews SC, participated in the collection of samples, planning and coordination of the study, Kelly DJ. Iron acquisition and virulence in helicobacter pylori :a major role for provided general supervision. All authors read and approved the final manuscript. feob, a high-affinity ferrous iron transporter. Mol Microbiol. 2000;37:274–86. 21. Deng K, Blick RJ, Liu W, Hansen EJ. Identification of Francisella tularensis Ethics approval and consent to participate genes affected by iron limitation. Infect Immun. 2006;74:4224–36. The study protocol was granted by the Research Ethics Committee, Wuxi 22. Lenco J, Hubálek M, Larsson P, Fucíková A, Brychta M, Macela A, Stulík J. Fisheries College of Nanjing Agriculture University (Permit No. NJYY20160929–1), Proteomics analysis of the Francisella tularensis LVS response to iron and all methods were performed in accordance with the approved guidelines restriction: induction of the F. tularensis pathogenicity island proteins and regulations. IglABC. FEMS Microbiol Lett. 2007;269:11–21. 23. Folsom JP, Parker AE, Carlson RP. Physiological and proteomic analysis of Competing interests Escherichia coli iron-limited chemostat growth. J Bacteriol. 2014;196:2748–61. The authors declare that they have no competing interests. 24. Yao Z, Wang Z, Sun L, Li W, Shi Y, Lin L, Lin WX, Lin XM. Quantitative proteomic analysis of cell envelope preparations under iron starvation stress in Aeromonas hydrophila. BMC Microbiol. 2016;16:161. Publisher’sNote 25. Pang MD, Jiang JW, Xie X, Wu YF, Dong YH, Kwok AH, Zhang W, Yao HC, Lu Springer Nature remains neutral with regard to jurisdictional claims in published CP, Leung FC, Liu YJ. Novel insights into the pathogenicity of epidemic maps and institutional affiliations. Aeromonas hydrophila ST251 clones from comparative genomics. Sci Rep. 2015;5:9833. Author details 26. Caliaperumal J, Wowk S, Jones S, Ma YL, Colbourne F. Bipyridine, an iron chelator, Wuxi Fisheries College, Nanjing Agricultural University, Wuxi 214081, China. does not lessen intracerebral iron-induced damage or improve outcome after Key Laboratory of Freshwater Fisheries and Germplasm Resources intracerebral hemorrhagic stroke in rats. Transl Stroke Res. 2013;4:719–28. Utilization, Ministry of Agriculture, Freshwater Fisheries Research Center, 27. Alencar TD, Wilmart-Gonçalves TC, Vidal LS, Fortunato RS, Leitão AC, Lage C. Chinese Academy of Fishery Sciences, Wuxi 214081, China. Bipyridine (2,2′-dipyridyl) potentiates Escherichia coli lethality induced by nitrogen mustard mechlorethamine. Mutat Res. 2014;765:40. Received: 28 September 2017 Accepted: 5 April 2018 28. Lee P, Tan KS. Effects of epigallocatechin gallate against Enterococcus faecalis biofilm and virulence. Arch Oral Biol. 2015;60:393. 29. GB/T 5009.90–2003. Determination of iron, magnesium and manganese in foods. References Beijing: Standardization Administration of the People’s republic of China; 2003. 1. Janda JM, Abbott SL. The genus Aeromonas: taxonomy, pathogenicity, and 30. Wang XK, Yang RQ, Zhou YL, Gu ZX. A comparative transcriptome infection. Clin Microbiol Rev. 2010;23:35–73. and proteomics analysis reveals the positive effect of supplementary 2. Feelders RA, Vreugdenhil G, Eggermont AM, Kuiper-Kramer PA, van Eijk HG, 2+ Ca on soybean sprout yield and nutritional qualities. J Proteome. Swaak AJ. Regulation of iron metabolism in the acute-phase response: 2016;143:161. interferon gamma and tumour necrosis factor alpha induce hypoferraemia, 31. Qin N, Tan X, Jiao Y, Liu L, Zhao W, Yang S, Jia AQ. RNA-Seq-based ferritin production and a decrease in circulating transferrin receptors in transcriptome analysis of methicillin-resistant Staphylococcus aureus biofilm cancer patients. Eur J Clin Investig. 1998;28:520–7. inhibition by ursolic acid and resveratrol. Sci Rep. 2014;4:5467. 3. Reines HD, Cook FV. Pneumonia and bacteremia due to Aeromonas 32. Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ. Basic local alignment hydrophila. Chest. 1981;80:264–7. search tool. J Mol Biol. 1990;215:403–10. 4. Brenden RA, Huizinga HW. Pathophysiology of experimental Aeromonas hydrophila infection in mice. J Med Microbiol. 1986;21:311–7. 33. Robinson MD, Mccarthy DJ, Smyth GK. Edger: a Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics. 5. Rasmussen-Ivey CR, Figueras MJ, Mcgarey D, Liles MR. Virulence factors of 2010;26:139–40. Aeromonas hydrophila: in the wake of reclassification. Front Microbiol. 2016;7:1337. 6. Toma'S JM. The main Aeromonas pathogenic factors. ISRN Microbiol. 2012; 34. Yan MX, Dai WJ, Cai EP, Deng YZ, Chang CQ, Jiang ZD, Zhang LH. Transcriptome 2012:256261. analysis of Sporisorium scitamineum reveals critical environmental signals for 7. Mekalanos JJ. Environmental signals controlling expression of virulence fungal sexual mating and filamentous growth. BMC Genomics. 2016;17:354. determinants in bacteria. J Bacteriol. 1992;174:1–7. 35. Kanehisa M, Araki M, Goto S, Hattori M, Hirakawa M, Itoh M, Katayama T, 8. Sritharan M. Iron as a candidate in virulence and pathogenesis in mycobacteria Kawashima S, Okuda S, Tokimatsu T, Yamanishi Y. KEGG for linking genomes and other microorganisms. World J Microbiol Biotechnol. 2000;16:769–80. to life and the environment. Nucleic Acids Res. 2008;36:480–4. Teng et al. BMC Microbiology (2018) 18:52 Page 16 of 17 36. Isaacson T, Damasceno CM, Saravanan RS, He Y, Catalá C, Saladié M, Rose 60. Girija R. Iron and virulence in Francisella tularensis. Front Cell Infect JKC. Sample extraction techniques for enhanced proteomic analysis of plant Microbiol. 2017;7:107. tissues. Nat Protoc. 2006;1:769–74. 61. Skaar EP. The battle for Iron between bacterial pathogens and their 37. Smith PK, Krohn RIG, Hermanson G, Mallia AKFD, Gartner FJH, Provenzano vertebrate hosts. PLoS Pathog. 2010;6:e1000949. MD, Fujimoto EK, Goeke NM, Olson BJ, Klenk DC. Measurement of protein 62. Wrighting DM, Andrews NC. Iron homeostasis and erythropoiesis. Curr Top using Bicinchoninic acid. Anal Biochem. 1985;150:76–85. Dev Biol. 2008;82:141–67. 38. Candiano G, Bruschi M, Musante L, Santucci L, Ghiggeri GM, Carnemolla B, 63. Maltz M, Levarge BL, Graf J. Identification of iron and heme utilization genes Orecchia P, Zardi L, Righetti PG. Blue silver: a very sensitive colloidal in Aeromonas and their role in the colonization of the leech digestive tract. Coomassie G-250 staining for proteome analysis. Electrophoresis. 2004; Front Microbiol. 2015;6:763. 25:1327–33. 64. Hayrapetyan H, Siezen R, Abee T, Groot MN. Comparative genomics of Iron- 39. Wisniewski J, Zougman A, Nagaraj N, Mann M. Universal sample preparation transporting systems in Bacillus cereus strains and impact of Iron sources on method for proteome analysis. Nat Methods. 2009;6:359–62. growth and biofilm formation. Front Microbiol. 2016;7:842. 40. You C, Lin C. He H, et al. iTRAQ-based proteome profile analysis of superior 65. Neilands JB. Siderophores: structure and function of microbial iron transport and inferior Spikelets at early grain filling stage in japonica Rice. BMC Plant compounds. J Biol Chem. 1995;270:26723–6. Biol. 2017;17(1):100. 66. Long H, Zeng Y. Studies on resistance property of fish serum Transferrins 41. Shilov IV, Seymour SL, Patel AA, Loboda A, Tang WH, Keating SP, Hunter CL, against Aeromonas hydrophila. Journal of Hubei Agricultural College. Nuwaysir LM, Schaeffer DA. The paragon algorithm, a next generation 2004;24:119–23. search engine that uses sequence temperature values and feature 67. Gehring AM, Mori I, Walsh CT. Reconstitution and characterization of the probabilities to identify peptides from tandem mass spectra. Mol Cell Escherichia coli Enterobactin Synthetase from EntB, EntE, and EntF. Proteomics. 2007;6:1638–55. Biochemistry. 1998;37:2648–59. 42. Zhang MC, Cao YN, Yao B, Bai DQ, Zhou ZG. Characteristics of quenching 68. Neilands JB. Molecular aspects of regulation of high affinity iron absorption enzyme AiiO-AIO6 and its effect on Aeromonas hydrophila virulence factors in microorganisms. Adv Inorg Biochem. 1990;8:63–90. expression. J Fish China. 2011;35:1720–8. 69. Crosa JH, Walsh CT. Genetics and assembly line enzymology of Siderophore 43. Swift S, Lynch MJ, Fish L, Kirke DF, Tomás JM, Stewart GSAB, Williams P. biosynthesis in Bacteria. Microbiol Mol Biol Rev. 2002;66:223–49. Quorum sensing-dependent regulation and blockade of exoprotease 70. Franza T, Enard C, Van GF, Expert D. Genetic analysis of the Erwinia chrysanthemi production in Aeromonas hydrophila. Infect Immun. 1999;67:5192–9. 3937 chrysobactin iron-transport system: characterization of a gene cluster 44. Chu W, Zhou S, Zhu W, Zhuang X. Quorum quenching bacteria Bacillus sp. involved in uptake and biosynthetic pathways. Mol Microbiol. 1991;5:1319–29. QSI-1 protect zebrafish (Danio rerio) from Aeromonas hydrophila infection. 71. Bearson BL, Bearson SM, Uthe JJ, Dowd SE, Houghton JO, Lee I, Toscano MJ, Sci Rep. 2014;4:5446. Lay Jr DC. Iron regulated genes of Salmonella enterica serovar typhimurium 45. Gang L, Huang L, Su Y, Qin Y, Xu X, Zhao L, Yan Q. Flra, flrb and flrc regulate in response to norepinephrine and the requirement of fepDGC for adhesion by controlling the expression of critical virulence genes in Vibrio norepinephrine-enhanced growth. Microbes Infect. 2008;10:807–16. alginolyticus. Emerging Microbes Infect. 2016;5:e85. 72. Bou-Abdallah F. The iron redox and hydrolysis chemistry of the ferritins. 46. Tsou AM, Zhu J. Quorum sensing negatively regulates hemolysin transcriptionally Biochim Biophys Acta. 2010;1800:719–31. and posttranslationally in Vibrio cholerae. Infect Immun. 2010;78:461–7. 73. Hubálek M, Hernychová L, Havlasová J, Kasalová I, Neubauerová V, Stulík J, 47. Teng T, Liang LG, Chen K, Xi BW, Xie J, Xu P. Isolation, identification and Macela A, Lundqvist M, Larsson P. Towards proteome database of Francisella phenotypic and molecular characterization of pathogenic Vibrio vulnificus tularensis. J Chromatogr B. 2003;787:149–77. isolated from Litopenaeus vannamei. PLoS One. 2017;12:e0186135. 74. Hubálek M, Hernychová L, Brychta M, Lenco J, Zechovská J, Stulík J. 48. Saganuwan SA. A modified arithmetical method of reed and Muench for Comparative proteome analysis of cellular proteins extracted from highly determination of a relatively ideal median lethal dose (LD 50). Afr J Pharm virulent Francisella tularensis ssp. tularensis and less virulent F. tularensis ssp. Pharmacol. 2011;5:1543–6. holarctica and F. tularensis ssp. mediaasiatica. Proteomics. 2004;4:3048–60. 49. Liu W, Dong H, Li J, Ou Q, Lv Y, Wang X, Xiang Z, He Y, Wu Q. RNA-seq 75. Holden KM, Browning GF, Noormohammadi AH, Markham PF, Marenda MS. reveals the critical role of OtpR in regulating Brucella melitensis metabolism TonB is essential for virulence in avian pathogenic Escherichia coli.Comparative and virulence under acidic stress. Sci Rep. 2015;5:10864. Immunology Microbiology and Infectious Diseases. 2012;35:129–38. 50. Nagar V, Bandekar JR, Shashidhar R. Expression of virulence and stress 76. Dong Y, Liu J, Pang M, Du H, Wang N, Awan F, Lu C, Liu Y. Catecholamine- response genes in Aeromonas hydrophila under various stress conditions. J stimulated growth of Aeromonas hydrophila requires the TonB2 energy Basic Microbiol. 2016;56:1132–7. transduction system but is independent of the Amonabactin Siderophore. 51. Teixeiragomes AP, Cloeckaert A, Zygmunt MS. Characterization of heat, Front Cell Infect Microbiol. 2016;6:183. oxidative, and acid stress responses in Brucella melitensis. Infect Immun. 77. Angerer A, Gaisser S, Braun V. Nucleotide sequences of the sfuA, sfuB, and 2000;68:2954–61. sfuC genes of Serratia marcescens suggest a periplasmic-binding-protein- 52. Casabianca A, Orlandi C, Barbieri F, Sabatini L, Cesare AD, Sisti D, Pasquaroli S, dependent iron transport mechanism. J Bacteriol. 1990;172:572–8. Magnani M, Citterio B. Effect of starvation on survival and virulence expression of 78. Andrews NC. Iron homeostasis: insights from genetics and animal models. Aeromonas hydrophila from different sources. Arch Microbiol. 2015;197:431–8. Nat Rev Genet. 2000;1:208–17. 53. Payne SM, Lawlor KM. Chapter 11 – molecular studies on iron acquisition by 79. Rahman MH, Suzuki S, Kawai K. Formation of viable but non-culturable state non- Escherichia coli, species. Bacteria. 1990:225–48. (VBNC) of Aeromonas hydrophila, and its virulence in goldfish, Carassius auratus. Microbiol Res. 2001;156:103–6. 54. Allan BJ, Stevenson RM. Extracellular virulence factors of Aeromonas hydrophila in fish infections. Can J Microbiol. 1981;27:1114–22. 80. Maalej S, Gdoura R, Dukan S, Hammami A, Bouain A. Maintenance of 55. Lv J, Yu GC, Sun ZH, Wang NJ, Zhu Y, Wang HC, Sun XS. Proteomic analysis pathogenicity during entry into and resuscitation from viable but of effects of iron depletion on Streptococcus pyogenes MGAS5005. nonculturable state in Aeromonas hydrophila, exposed to natural seawater Microbiology China. 2012;39:515–25. at low temperature. J Appl Microbiol. 2004;97:557–65. 56. Bhatnagar, Elkins, Fortier. Heat stress alters the virulence of a rifampin-resistant 81. Mchugh JP, Rodríguezquinoñes F, Abdultehrani H, Svistunenko DA, Poole mutant of Francisella tularensis LVS. I nfect Immun 1995; 63:154–159. RK, Cooper CE, Andrews SC. Global iron-dependent gene regulation in 57. Ellis AE, Burrows AS, Stapleton KJ. Lack of relationship between virulence of Escherichia coli. A new mechanism for iron homeostasis J Biol Chem. 2003; Aeromonas salmonicida and the putative virulence factors: A-layer, extracellular 278:29478–86. proteases and extracellular haemolysins. J Fish Dis. 1988;11:309–23. 82. He X. Proteomic analysis of biological affects on Streptococcus pneumoniae induced by manganese depression. Jinan University 2011. 58. Brickman TJ, Cummings CA, Liew SY, Relman DA, Armstrong SK. 83. Cunningham AB, Sharp RR, Jr FC, Gerlach R. Effects of starvation on bacterial Transcriptional profiling of the iron starvation response in Bordetella pertussis transport through porous media. Adv Water Resour. 2007;30:1583–92. provides new insights into siderophore utilization and virulence gene expression. J Bacteriol. 2011;193:4798–812. 84. Elgaml A, Miyoshi SI. Regulation systems of protease and hemolysin 59. Sheikh MA, Taylor GL. Crystal structure of the Vibrio cholerae, ferric uptake production in Vibrio vulnificus. Microbiol Immunol. 2017;61:1–11. regulator (Fur) reveals insights into metal co-ordination. Mol Microbiol. 85. Tang TS, Lu CP. An Acinetobacter baumannii strain isolated from mandarin 2009;72:1208–20. fish possesses type 4 pili. J Nanjing Agric Univ. 1997;20:114–6. Teng et al. BMC Microbiology (2018) 18:52 Page 17 of 17 86. Chen XQ, Cai HY, Zhang W, Yan MY, Gao H, Duan GX, Yan ZY. The Role of Fur Involved in Bioflim Formation of Vibrio Cholerae. Prog Mod Biomed. 2013;13:2841–5. 87. Vasil ML, Ochsner UA. The response of Pseudomonas aeruginosa, to iron: genetics, biochemistry and virulence. Mol Microbiol. 1999;34:399–413. 88. Thune RL, Johnson MC, Graham TE, Amborski RL. Aeromonas hydrophila B- haemolysin: purification and examination of its role in virulence in 0-group channel catfish, Ictalurus punctatus (Rafinesque). J Fish Dis. 1986;9:55–61. 89. Zhu DL, Li AH, Wang JG, Li M, Cai TZ, Hu G. The correlation between the distribution pattern of virulence genes and the virulence of Aeromonas hydrophila strains. Acta Sci Nat Univ Sunyatseni. 2006;45:82–5. 90. Chu WH. Invasion mechanism and extracellular protease of Aeromonas hydrophila. Nanjing Agric Univ. 2002; 91. Cao Q, Zhang XY, Pang MD, Wang NN. Identification and molecular typing of the epidemic Aeromonas hydrophila strains in one farm of Nanjing. J Fish China. 2017;41:134–41. 92. Cascón A, Yugueros J, Temprano A, Sánchez M, Hernanz C, Luengo JM, Naharro GN. A major secreted elastase is essential for pathogenicity of Aeromonas hydrophila. Infect Immun. 2000;68:3233–41.

Journal

BMC MicrobiologySpringer Journals

Published: Jun 4, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off