Qualitative Alterations of Bacterial Metabolome after Exposure to Metal

Subscriber access provided by SUNY DOWNSTATE

Article

Qualitative Alterations of Bacterial Metabolome after Exposure to Metal Nanoparticles with Bactericidal Properties: A Comprehensive Workflow Based on H NMR, UHPLC-HRMS and Metabolic Databases 1

Theodoros G. Chatzimitakos, and Constantine D. Stalikas J. Proteome Res., Just Accepted Manuscript • DOI: 10.1021/acs.jproteome.6b00489 • Publication Date (Web): 19 Jul 2016 Downloaded from http://pubs.acs.org on July 20, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Journal of Proteome Research is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Proteome Research

Qualitative Alterations of Bacterial Metabolome after Exposure to Metal Nanoparticles with Bactericidal Properties: A Comprehensive Workflow Based on 1H NMR, UHPLC-HRMS and Metabolic Databases

Theodoros G. Chatzimitakos and Constantine D. Stalikas* Laboratory of Analytical Chemistry, Department of Chemistry, University of Ioannina, 45110 Ioannina, Greece

*Corresponding author. e-mail: [email protected], Fax: ××30 26510 08796

1 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ABSTRACT Metal nanoparticles (NPs) have proven to be more toxic than bulk analogues of the same chemical composition due to their unique physical properties. The NPs, lately, have drawn the attention of researchers because of their antibacterial and biocidal properties. In an effort to shed light on the mechanism, through which the bacteria elimination is achieved and the metabolic changes they undergo, an untargeted metabolomic fingerprint study was carried out on Gram-positive (Staphylococcus aureus) and Gram-negative (Escherichia coli) bacteria species. The 1H NMR spectroscopy, in conjunction with high resolution mass-spectrometry (HRMS) and an unsophisticated data processing workflow were implemented. The combined NMR / HRMS data, supported by an open-access metabolomic database, proved to be efficacious in the process of assigning a putative annotation to a wide range of metabolite signals and is a useful tool to appraise the metabolome alterations, as a consequence of bacterial response to NPs. Interestingly, not all the NPs diminished the intracellular metabolites; bacteria treated with iron NPs produced metabolites not present in the non-exposed bacteria sample, implying the activation of previously inactive metabolic pathways. In contrast, copper and iron-copper NPs reduced the annotated metabolites, alluding to the conclusion that the metabolic pathways (mainly: alanine, aspartate and glutamate metabolism, beta-alanine metabolism, glutathione metabolism, arginine and proline metabolism) were hindered by the interactions of NPs with the intracellular metabolites.

KEYWORDS Metal nanoparticles, bactericides, metabolomics, NMR, UHPLC-HRMS, metabolic databases 2 ACS Paragon Plus Environment

Page 2 of 31

Page 3 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


INTRODUCTION Antibiotics are of paramount importance for the treatment of infectious diseases caused by human pathogenic bacteria.1 However, their indiscriminate use promotes the development of antibiotic-resistant bacteria. Under various evolutionary processes, bacteria surmount the negative effects of antibiotics and form resistant strains, known as superbugs.2 Nanomaterials have emerged as alternative antibacterial agents, due to their small size and high surface-to-volume ratio to overcome this hindrance.3-5 Several nanomaterials comprised of silver, gold, metal oxides, graphene polymer-based materials, etc. have been examined for their detrimental effects on various bacterial species.2,

3, 6-10

Of those, special attention has been paid to the

metallic nanoparticles (NPs).11 Their satisfactory antimicrobial effects are accompanied by certain advantages, such as high potential for varied applications, low cost of synthesis and ease in controlling their structural characteristics. All these auspicious characteristics encouraged a new wave of research and accentuated a new class of antibiotic materials.6 Considering that some metals are self-sanitized, it is expected that metal NPs would transcend their bulk analogues and possess a substantially higher bactericidal effect.6, 12, 13

In a previous study, we have shed light on the antibacterial effect of three metal

NPs, viz. iron, copper and bimetallic iron-copper, against two common bacterial strains, Staphylococcus aureus (S. aureus) and Escherichia coli (E. coli).14 The three NPs exhibited noteworthy pernicious effect against the tested planktonic bacteria and ultimately, thwarted cell division, even at low quantities. Furthermore, the metal NPs were able to inhibit, to a certain degree, the growth of the notoriously resistant biofilms of the respective bacteria. Nevertheless, the effect of the NPs against the two 3 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 31

bacteria varied in respect to the growth arrangement of the bacteria (planktonic or biofilm). This result reinforces the concept that the way to eliminate bacteria is invariably dependent on the specific bacteria species.2, 15 According to recent data, the survivability of the microorganisms along with their virulence and their invasive capabilities are directly linked to metabolic processes.16-18 Therefore, obtaining holistic information about the metabolic status of the microorganisms and correlating it with exogenous stimuli are of high significance. 19, 20 By definition, metabolomics is the systematic study of the unique chemical fingerprints as a result of the dynamic multi-parametric response of a living system to pathophysiological stimuli.21, 22 For this reason, metabolomics is supposed to be the endpoint of the ‘omics cascade’ and the cornerstone of systems biology. Microbial metabolomics is one of the platforms for integrating biological information into systems microbiology, to facilitate the understanding of microbial interactions and cellular functions, since metabolomics can potentially provide a more accurate snap shot of the actual physiological state of the cell.23 Investigations in cellular metabolomics require the development of robust and reliable experimental protocols for all steps in the experimental procedure, ranging from biomass cultivation, quenching and extraction of the metabolome to putative annotation, identification and quantitation of metabolites.24 Bacterial metabolomics is a major part of cellular metabolomics, which uses cultivated bacteria samples for different kinds of researches,

such

as

bacterial

metabolic

profile,

cellular

responses

to

microenvironment changes (pH, temperature, exposure to drugs or toxic species) etc. 25

So far, there is no single technique able to carry out a comprehensive metabolome analysis due to the complexity and heterogeneity of the samples.26 Nuclear magnetic


Page 5 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


resonance (NMR) spectroscopy and MS, either in direct infusion mode or coupled with chromatography, are capable of evaluating a wide range of metabolites, simultaneously and they are well suited for targeted or untargeted metabolic fingerprinting.21, 23, 27-31 High resolution, superb reproducibility, universality and ease in determining the absolute concentrations of compounds are some of the advantages of the NMR spectroscopy.32,

33

In addition, the use of cryogenically cooled NMR

probe heads can alleviate the relatively low sensitivity of the technique.34 An alternative option, of much higher sensitivity, is the MS technique, which provides spectral information, such as the exact mass of the molecular ion and fragmentation patterns, assisting in the identification of the metabolites. Additionally, the potential of coupling it with chromatographic separation techniques, such as gas and liquid chromatography, facilitates the alleviation of errors arising from the complexity of the matrixes.35-38 The popularity of the two aforementioned techniques, which has been even increasing in the past years, has generated a cascade of innovative studies, in an effort to understand better the physiology of microbes and to elucidate the mode of action of antibacterial agents.

39, 40

Therefore, the employment of metabolomics to study the

response of bacteria to potential antibacterial compounds may be more favorable than phenotypic screening. 40, 41 The aim of this study is to scrutinize the alterations in the metabolomes of E. coli and S. aureus induced by their exposure to metal NPs. The two bacterial strains were subjected to stress conditions, in such a way that alterations were caused in the metabolome but no total cell death was inflicted. A data processing workflow was implemented, where 1H NMR and liquid chromatography - high-resolution MS (LCHRMS) were mated to putatively annotate metabolites with the aid of E. coli



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Metabolome Database (ECMDB), an on-line open-access metabolomic database. Given the paucity of such data, a metabolic pathway analysis was carried out, by means of the pathway analysis tool of MetaboAnalyst 3.0, an on-line tool suite, in an effort to unravel the metabolic pathways involved. Using this approach, noticeable and diverse effects, expressed as remarkable bacterial alterations were seen, with respect to small molecules and functional entities, as a function of NPs composition.

EXPERIMENTAL SECTION Reagents All chemicals used were of analytical grade. Bacteriological peptone and yeast extract were from Biolife (Milano, Italy). Agar-Agar Danish was purchased from Carl Roth GmbH & Co. KG (Karlsruhe, Germany). Deuterated methanol (NMR quality) was purchased from Deutero (Kastellaun, Germany). Petri dishes 92 × 16 mm, without cams were purchased from Sarstedt AG & Co (Nümbrecht, Germany). Double distilled water (DDW) was used to prepare all solutions. Isotonic saline solution, made up of 0.90% (w/v) NaCl in DDW, was used throughout all the experiments. Lysogeny broth (LB) medium was made up of 10.0 g/L NaCl, 5.0 g/L yeast extract and 10.0 g/L bacteriological peptone. The pH was adjusted to 7.4 with 0.1 M NaOH. The plating medium was agar-LB containing 15.0 g/L agar. All solutions and glassware were sterilized by autoclaving at 121 oC for 15 min, at 15 psi.

Instrumentation 1

H NMR spectra were recorded on a Brüker AV-500 spectrometer equipped with a

TXI cryoprobe (Bruker BioSpin, Rheinstetten, Germany). The NMR system was controlled by the software TopSpin 2.1. (Copyright 2009, Bruker BioSpin). All


Page 6 of 31

Page 7 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


spectra were acquired with an acquisition time of 3.171 s, relaxation delay 5 s, 64 K data points, 90° pulse length and 256 scans. Phase and baseline were manually corrected after the Fourier transform. The chromatographic separation and the MS confirmatory analyses were carried out on a Hypersil GOLD 1.9 µm particle size (100 mm × 2.1 mm I.D) with an Ultimate 3000 HPLC (Dionex, Milan, Italy) ultra-high-performance liquid chromatographic (UHPLC) system. The oven temperature was kept constant at 30oC. The mobile phase consisted of water (A) and acetonitrile (B), both acidified with formic acid 0.1% (v/v). A gradient program was run, for optimum separation of the target analytes, as follows: 0-13.78 min, 20-90% B, 13.78-15.28 min, 90% B, 15.28-18.06 min, 90-20% B, followed by a 2-min re-equilibration time of the column. The flow rate of the mobile phase was set at 300 µL/min. A linear trap quadrupole (LTQ) Orbitrap mass spectrometer (Thermo Scientific, Bremen, Germany), equipped with an atmospheric pressure interface and an ESI ion source was used as detector. The effluents from the column were delivered to the ion source with nitrogen as sheath and auxiliary gas. Both positive and negative ionization modes were employed. The experimental conditions for the positive ion mode were: Injection volume: 2.5 µL, source voltage: 3.40 kV, tube lens 110 V while the heated capillary voltage was 40.00 V and the temperature was maintained at 320 °C. For negative ion mode the conditions were: injection volume: 10.0 µL, source voltage: 3.70 kV, tube lens 120 V while the heated capillary voltage was -30.00 V and the temperature was maintained at 320 °C. In both cases, two scan modes were used: a full scan mode, at a resolution of 60,000 and an m/z range of 50-1500 and a most-intense-ion scan (MS/MS fragmentation of the most abundant ion), with a resolution of 7,500. The system was controlled via the Thermo Xcalibur 2.1 software (Copyright 1998-2009, Thermo Fischer Scientific Inc.)



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Cultivation of bacteria and cell pellets formation The two bacterial strains, Staphylococcus aureus (NCTC 6571) and Escherichia coli (DH5 alpha) were stored in glycerol 20% (v/v), at −80oC. Bacterial pre-inoculums of each bacterium were prepared by adding a small amount of bacterial spores in 10 mL of fresh LB medium and incubating for 16 and 10 h for E. coli and S. aureus, respectively, at 37oC. Cultures were prepared by inoculating in 250 mL of LB with 2.5 mL of fresh pre-inoculums and incubating for 16 h, at 37oC, in an incubator shaker, at 250 rpm. In order to form bacteria pellets, the bacterial cultures were centrifuged at 4000 rpm at 4oC, for 5 min. The supernatant was decanted away and 30 mL of isotonic saline solution was added to the resulting pellets to wash away the growth medium residues. The tubes were vortexed for 1 min and centrifuged again for 5 min, under the same conditions. This cleaning step was repeated twice.

Cell viability assay Bacteria cell pellets were resuspended in suspensions of the three nanomaterials, at different concentrations (5, 50, 100, 200 µg/mL) in isotonic saline solution, until the optical density at 540 nm (OD540) was 0.1 (approximately 8×107 CFU/mL) with respect only to the bacteria cells. The spectrophotometer was zeroed with the respective nanomaterial suspension. Bacteria were incubated at 37oC, under stirring at 250 rpm, for 4 h. Also, a control sample was prepared in the absence of any of the nanomaterials tested. The colony counting method was used to assess the loss of bacteria viability. Therefore, samples were taken from the incubated bacteria, at different time periods and successive 10-fold dilutions were carried out. Finally, 100 µL of the last dilution were spread on LB plates. The plates were left overnight to


Page 8 of 31

Page 9 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


grow, at 37oC. The colonies in the plates were counted and compared with those of the control sample. The experiments were carried out in triplicate and the mean values were calculated. The loss of viability was expressed as a % percentage of the control sample.9

Metabolomic assay Α bacteria cell pellet was resuspended in a suspension of NPs (100 µg/mL) in isotonic saline solution, so that the OD540 of the whole suspension was 0.1 (refers only to bacteria cells, as the spectrophotometer was zeroed with the respective nanomaterial suspension). Then, the suspension was incubated, at 37oC, under shaking (at 250 rpm), for 30 min. The whole suspension was centrifuged at 4000 rpm at 4oC, for 5 min. The supernatant was discarded and the resulting pellet was cleaned twice with isotonic saline solution. Subsequently, 0.5 mL of cold methanol (−80oC) was added to the new pellet. Then, the pellet was subjected to three freeze-thaw cycles using liquid nitrogen in order to quench the metabolic processes and assist the extraction of the intracellular metabolites.42 The mixture was centrifuged at 4000 rpm at −4oC, for 10 min. The supernatant was retracted and transferred to an amber glass vial. This procedure was repeated once more and the supernatants were pooled. Half portion of the metabolites extract was transferred to an NMR glass tube and both portions were evaporated to dryness, under a gentle nitrogen steam. For 1H-NMR measurements, the residue was resuspended in 500 µL of deuterated methanol while for LC-HRMS it was resuspended in 20 µL of acetonitrile. The reproducibility of the process was ensured by assessing the 1H-NMR spectra, after repeating each experiment sample pretreatment three times.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 31

Data processing After acquiring the 1H-NMR spectra, the signal positions, in ppm, were input to the 1D NMR search engine of ECMDB (http://ecmdb.ca) so that the tentative identification of the components contained in the metabolic extracts, could be carried out.43, 44 The search engine returned a list of tentative metabolites for each sample. The

Jaccard

index

(similarity

coefficient)

was

adopted

for

comparing

the similarity of data sets. The lower the Jaccard index, the weaker the probability of the presence of the corresponding metabolite in the sample. Further evidence on metabolites structure was given using HRMS (LTQ Orbitrap), after separation on a UHPLC system. The obtained chromatograms were scanned, manually, for the exact mass values of the metabolites (accurate to 4 decimal places) with mass tolerance of ± 5 ppm. Mass accuracy of the peak was calculated using the elemental composition calculator provided by Xcalibur 2.1. After verification of the metabolites, a metabolic pathways analysis was conducted to appraise, ultimately, alterations as a consequence of the exposure of bacteria to NPs. The pathway analysis was accomplished using the MetaboAnalyst 3.0, a comprehensive

tool

suite

for

metabolomic

data

analysis

(http://www.metaboanalyst.ca).45-47 The verified metabolites were input to the pathway analysis tool and the analysis was carried out by selecting the appropriate library for E. coli or S. aureus. The significance of pathways was defined by the calculated p values. Figure 1 illustrates the design and the data flow of this workflow, which results in a comprehensive untargeted metabolite fingerprint for E. coli and S. aureus.

RESULTS 10 ACS Paragon Plus Environment

Page 11 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Cell viability Τhe three metallic NPs were found to cause remarkable cell mortality, as shown in Figs S1 and S2 of Supporting Information. In brief, a concentration of 100 µg/mL of the metal NPs, after 4h of incubation, was able to kill all the bacteria cells of the two species, while 200 µg/mL of NPs achieved the same effect, in 30 min. Between the two bacteria species, E. coli proved to be more susceptible than S. aureus, at all concentrations tested. The minimum bactericidal concentration (MBC90, 90% of bacteria killed) for the three metal NPs was 50 µg/mL, for both bacteria species, after 4h of incubation. Comprehending the susceptibility differences between the two bacteria strains can be a challenging task, due to their structural and genetic properties.

48

Taking into

consideration that the metabolic processes are directly linked to the survivability of the microorganisms, a comparison of the metabolic profiles, between bacteria exposed and non-exposed to NPs, can provide insights into the biochemical changes occurring in the bacteria metabolome, which ultimately, could be responsible for their death.

Metabolomic assay In the study below, all the experiments were carried out in saline instead of LB. In the former, the differences in the metabolome can reasonably be attributed only to the exposure of the bacteria to the metallic NPs whereas in the latter, metabolic changes may arise from the multiple interactions with the growth medium. In addition, the bacteria used for the metabolomic study were harvested at the end of the exponential growth phase (see Supporting Information, Fig. S3), when they are metabolically more active than at any other growth phases.

49, 50

To obtain comparable and

reproducible results, all the experimental conditions including bacteria culture growth



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

time, incubation periods and sample processing time and conditions, were meticulously and strictly executed. This way, even though experiments were repeated on different batches of bacteria, the cultures were grown in a reproducible manner and hence the results were comparable. The metabolome alterations were evaluated after exposing the bacteria to high stress conditions.

51

Therefore, the exposure conditions were chosen in accordance with the

results of the cell viability assessment (Supporting information, Fig S2). Moderate bactericidal activity (~40% loss of viability, high stress conditions) was achieved by two combinations of time and concentration, namely, 50 µg/mL NPs - 1 h exposure or 100 µg/mL NPs - 30 min exposure. Higher concentrations or longer incubation time periods were unsuitable due to the higher bacterial mortality, while shorter incubation time or lower concentrations were not enough to monitor noticeable alterations among the metabolic fingerprints. Between the two aforementioned combinations, that of 100 µg/mL NPs - 30 min exposure was selected for both bacteria, as more differences were seen in NMR spectra, under these conditions, as compared to the 50 µg/mL NPs - 1 h exposure, probably, due to fewer interactions between the bacteria and NPs. After exposing the bacteria to the NPs and following a simple workup procedure, the metabolites were extracted. Subsequently, the proposed approach, which combines 1

H-NMR and UHPLC-HRMS with the online process of the metabolomic data, via

the open-access metabolic data processing websites, was organized in six steps, as described in detail below.

Metabolomic fingerprinting by NMR The one-dimensional (1D) 1H NMR spectra of multi-component samples (1st step) are rather complex but provide an unbiased picture of the metabolites. Figures 2 and 3


Page 12 of 31

Page 13 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


delineate the numerous changes in 1H-NMR spectra in the metabolic profiles of E. coli and S. aureus bacteria, respectively, after their exposure to NPs. More specifically, a close look at Fig. 2 demonstrates a few differences in the spectra of the control and Fe-treated E. coli, in the short region near 1.0 and 7.2 ppm. The other two spectra of Cu – and Cu-Fe – treated E. coli samples, shown in the same Figure, manifest numerous differences from the control spectrum. It can be seen that many signals of the most deshielded protons such as aromatic, carboxylic and aldehydic are absent, as compared to the control spectrum, bespeaking the absence of pertinent compounds. Furthermore, most resonances between 1.5 and 3.0 ppm are diminished, implying that metabolites like alcohols, ethers or esters are limited in the metabolite pool of the treated bacteria. The above observations are indicative of the changes caused to the metabolome of E.coli, as a response to the exposure to the metallic NPs. As regards S. aureus, it can be deduced from the spectra of Fig 3, that two of the three NPs, i.e. Cu and Fe-Cu, have significant impact on its metabolome, considering the differences that the respective spectra exhibit, in various spectral regions, as compared with the spectrum of control sample. The number of resonances from metabolites like aldehydes and carboxylic acids seem to be restricted in NMR spectra of S. aureus, exposed to Cu and Fe-Cu NPs (there are only a very few signals in the high region of ppm) whereas differences worthy of attention appear in the vicinity of 3 ppm, where protons of alkyl halogenides, alcohols, carbohydrates and esters resonate. Again, inspection of the NMR spectrum of Fe-exposed bacteria shows that there are no noticeable alterations in comparison to control sample. Finally, NMR spectra of the control samples of both bacterial strains exhibit more similarities than differences. The similarities can be justified by the heterotrophic character of both



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

bacteria; since they were grown in the same medium, pathways common to the bacteria species are expected to be active. To unravel the NMR spectra information and obtain an in-depth view of the metabolic profiles, all peaks were introduced into the 1D-NMR search tool of ECMDB (2nd step of the workflow), for further processing. More than 99% of the spectral peaks were fitted using the ECMDB spectral analysis software and were assigned to 312 tentative metabolites (see Supporting Information, Table S1). Apparently, each bacterium profile contains a substantial number of hydrophilic molecules dominated by organic acids, amino acids, organic phosphates, alcohols, various derivatives of sugars as well as fatty acids. These findings are in accordance with the metabolite classification list of ECMDB.

Metabolomic fingerprinting by UHPLC-HRMS (Orbitrap) and data verification To increase the degree of confidence in the assignment of metabolites signal resulting from the NMR spectra analysis with the ECMDB, the next steps (3rd and 4th of the workflow) were carried out by means of UHPLC-HRMS. Mass accuracy values lower than 4 ppm were acceptable for the annotation of the metabolites. Bacterial extracts were analyzed in both positive and negative ionization mode, as changes in polarity can often alter competitive ionization and suppression effects revealing otherwise suppressed metabolite signals. Metabolite assignments were made by comparing the HRMS data (accurate mass, isotopic distribution and fragmentation pattern) of the compounds detected with those reported in the ECMDB, for the respective metabolites. Following this procedure, the positive-ionization MS spectra revealed more observable peaks, with a total of 282 mass signals extracted using the Xcalibur software, than the negative ionization mode where 74 mass signals found, most likely


Page 14 of 31

Page 15 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


due to higher efficiency of protonation relative to deprotonation.52 The obtained MS/MS spectrum of the most abundant ion further confirmed the findings of HRMS data. The identities, retention times, theoretical and experimental molecular ions of the individual components as well as MS/MS data and basic NMR spectral information are presented in Table S2. The combination of accurate mass data for the large collection of metabolites, theoretical isotope abundance data and knowledge of the different ion types detected provided a great number of mass spectrometric signals, which were assigned to metabolites in the samples studied, with greater confidence. Specifically, of the 312 tentative metabolites resulting from the NMR spectra deconvolution of 1D NMR search engine of ECMDB, thirty metabolites with low Jaccard index were weakly annotated in the samples, following the above-mentioned MS data process. These metabolites, whose presence in the samples could not be confirmed by the followed procedure, were not included in the pathway analysis detailed below. Detailed information about the classes of metabolites and the sample(s) in which they were found can be seen in Table S1 of Supporting Information. Once again, it can be seen that the detected metabolites belong to various classes of compounds. Among others, the presence of amino acids, carboxylic acids, fatty acids, organic phosphates, saccharides, nucleosides and sugar alcohols was corroborated, suggesting that the proposed workflow is suitable for the putative annotation of a wide range of metabolites.

Metabolic pathway analysis A pathway analysis (5th step) was conducted to transform the obtained results into biological information using the appropriate tool of MetaboAnalyst 3.0 tool suite.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

This way, the metabolic network in which the metabolites are involved, are easily unveiled and a case-by-case presentation of the numbers of relevant metabolic pathways, is given in Fig. 4. They are also viewed in detail, in Table S3 of Supporting Information, where the top metabolic pathways (evaluated by the respective p values) are boldfaced. In the case of E. coli, the metabolites deriving from the non-exposed bacteria were assigned to sixty seven different metabolic pathways (Fig. 4A). In all studied cases of bacterial exposure, the following fourteen metabolic pathways were not affected by the presence of NPs: alanine, aspartate and glutamate metabolism, beta-alanine metabolism, biosynthesis of unsaturated fatty acids, citrate cycle, fatty acid metabolism, galactose metabolism, glycerolipid metabolism, glycine, serine and threonine metabolism, glycolysis or gluconeogenesis, glyoxylate and dicarboxylate metabolism, methane metabolism, pantothenate and CoA biosynthesis, pyrimidine metabolism and pyruvate metabolism. However, a certain number of other metabolic pathways were blocked or even activated although being inactive in the control bacteria. This signifies that an alternative option is ‘switched on’ for cells to assist their survivability or to adapt themselves to the malign environment.53 More specifically, when Fe particles were used against E. coli, seventy one pathways were confirmed, i.e. sixty seven unaltered pathways of the non-exposed bacteria along with four new ones: the alpha-linolenic acid metabolism, the lipoic acid metabolism, the metabolism of xenobiotics by cytochrome P450 and the phenylalanine, tyrosine and tryptophan biosynthesis. Contrarily, for Fe-Cu – exposed E. coli, the metabolic pathways were limited only to the aforementioned fourteen pathways, which remained unaffected. The rest of the pathways with the respective metabolites were not visible in the samples. In the case of Cu, the fourteen pathways mentioned above, manifest themselves as active along with five more, i.e.: the amino sugar and


Page 16 of 31

Page 17 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


nucleotide sugar metabolism, the fructose and mannose metabolism, the pentose and glucuronate interconversions, the pentose phosphate pathway and the propanoate metabolism. In the case of S. aureus, fifty seven active metabolic pathways were found in the control sample (Fig. 4B) and the following ten remained unchanged during the exposure to three NPs: arginine and proline metabolism, butanoate metabolism, citrate cycle, galactose metabolism, glycerolipid metabolism, glycolysis or gluconeogenesis, glyoxylate and dicarboxylate metabolism, methane metabolism, pentose phosphate pathway, pyruvate metabolism. When Fe was used, fifty eight pathways were observed; the extra pathway, in this case, was the terpenoid backbone biosynthesis. On the contrary, Cu NPs limited the number of pathways down to forty six, blocking the following eleven pathways: C5-branched dibasic acid metabolism, d-alanine metabolism, inositol phosphate metabolism, lipoic acid metabolism, phenylalanine metabolism, lysine degradation, nicotinate and nicotinamide metabolism, novobiocin biosynthesis, phenylalanine, tyrosine and tryptophan biosynthesis, tryptophan metabolism and tyrosine metabolism. Bimetallic Fe-Cu NPs were able to diminish most of the metabolites and reduce the active metabolic pathways down to ten, as mentioned above.

DISCUSSION To the authors’ knowledge, this is the first study to embark upon unravelling the bacteria metabolome alterations, stemming from their exposure to NPs. Global metabolomic fingerprint of bacteria pose a challenge due to the diversity of the metabolites, a serious impediment to the progress in the study of metabolomics. By employing the workflow proposed herein, we were able to obtain global



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

profiling results regarding the metabolic status of the bacteria prior to and after their exposure to the studied NPs. From the above results useful points deserve further discussion. Of particular note, the presence of Fe NPs scarcely alters the metabolome of both E. coli and S. aureus, since no inactivation of any metabolic pathway was observed. Apparently, both bacteria were compelled to activate alternative pathways, in response to the Fe NPs, qualified as exogenous stimuli. Therefore, the bactericidal effect of Fe NPs cannot be attributed to damage of the metabolic network, implying an alternative way of action, which seemingly cannot affect the bacterial metabolism. Quite the opposite effect was observed on the metabolites pool for the two other metallic NPs. Both Cu and Fe-Cu NPs were found to distinctly alter the metabolic network of the bacteria by limiting the number of metabolites. Certain distinctions between bacteria were noticed when Cu NPs were employed: forty eight out of sixty seven and eleven out of fifty seven identified metabolic pathways were blocked, in E.coli and S. aureus, respectively. This fact, hints at a different way of action of the Cu NPs on the studied bacteria, which is dependent, in some ways, on the different composition of the metabolome of the two bacteria species. Bimetallic Fe-Cu NPs seem to have the highest bearing on the cellular metabolome, constraining significantly the number of metabolites by almost 85%, in both target bacterial strains. Therefore, it is conceivable that Fe-Cu NPs, compared to the two other NPs, are able to interact with intracellular metabolites and block the normal metabolism of the cells, resulting eventually in the bacterial cell death. Overall, the action of NPs is not assumed to discriminate between the two bacterial strains (Gram positive and negative), despite their structural differences, if we


Page 18 of 31

Page 19 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


consider their similar tendency in the direction of increase or decrease in the metabolic pathways, under the same conditions of exposure to NPs. The data gathered in the present study can be compared to those already published and concern metals in soluble form. Booth et al. showed that the exposure of Pseudomonas pseudoalcaligenes KF707 to aluminum and copper affects, in a different manner, the metabolic pathways of the microorganism. Copper caused more metabolic changes, focused mostly on certain amino acid metabolic pathways like cysteine/methionine metabolism, valine, leucine, and isoleucine biosynthesis and arginine/proline metabolism. Pathways like beta-alanine metabolism and alanine, aspartate, and glutamate metabolism remained unchanged.54 In another study, the metabolic changes of Pseudomonas fluorescens species caused by copper were evidenced. Again, the major differences of the metabolome are identified in the amino acid metabolic pathways. These differences on the metabolome of the aforementioned bacteria due to copper resemble, to a certain degree, those of Cu NPs on E. coli, especially in terms of amino acid metabolism pathways. This, however, is not the case with S. aureus, exposed to Cu NPs, most likely due to the fact that this is a Grampositive species, in contrast to the rest of bacteria already mentioned.55 Finally, it is evident that the data obtained, simply, from 1-D 1H NMR are informative and necessary but not sufficient for the identification of the metabolites. The HRMS data collected to refine the metabolic signature, confirm the 1H NMR data, to a large but definitely not to a full extent (~90%). This unequivocally supports the notion that 1

H NMR and LC-HRMS are highly complementary. This is even more exemplified

by the positive data for the bacterial intracellular metabolites in the bacteria metabolic pools.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

CONCLUSIONS Our experimental results demonstrated that the exposure of E. coli and S. aureus strains to the three metal nanomaterials is accompanied by pronounced and diverse metabolic differences. It is interesting that in both cases, Fe NPs had little impact on the intracellular metabolites while in contrast, Cu and Fe-Cu NPs caused the most modifications, in comparison to the control sample. This similar trend implies a similar way of action of the NPs on the studied bacteria. The study can provide a useful background in order to develop new, target-specific bactericidal materials. Our experimental data should also serve as a useful benchmark to assess next methodological improvements in bacterial studies. The procedure, which uses open-source software packages, seems to be applicable to figure out which metabolites might serve as markers of biological effects caused by the exposure to NPs. From a microbiological standpoint, we think that the information obtained could provide microbiologists and chemists with a convenient, centralized source, from which learning more about bacterial metabolomics and unique biochemical functions is rendered easier. Other unidentified variables (signals) may also be discriminant when E. coli and S. aureus are exposed to NPs, such as autoinducers and signaling molecules, which are extracellular and at considerably low concentration levels. Further research in this direction is required to capitalize on the findings, extending the study to quantitation as well as to other bacterial species in response to external variables.


Page 20 of 31

Page 21 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Supporting Information Figure S1. Images of plates for bactericidal properties of the NPs on E. coli and S. aureus strains. Figure S2. Bactericidal effect of iron, copper and iron-copper NPs on E. coli and S. aureus as a function of their concentration, at different exposure times. Figure S3. Growth curves of E. coli and S. aureus cell cultures. Table S1. Putatively annotated metabolites by the NMR and search tool of ECMDB and HRMS. Table S2. Identities, retention times, theoretical and experimental molecular ions, MS/MS data and NMR spectral information of the individual components. Table S3. Relevant metabolic pathways of exposed and non-exposed bacteria and respective p values.

Notes The authors declare no competing financial interest.

Acknowledgements Special thanks are due to NMR and Mass Spectrometry Centers of University of Ioannina, Greece, for providing their facilities.

REFERENCES (1)

(2)

Kalhapure, R. S.; Suleman, N.; Mocktar, C.; Seedat, N.; Govender, T., Nanoengineered drug delivery systems for enhancing antibiotic therapy. J. Pharm. Sci. 2015, 104, 872-905. Hajipour, M. J.; Fromm, K. M.; Akbar Ashkarran, A.; Jimenez de Aberasturi, D.; Larramendi, I. R. d.; Rojo, T.; Serpooshan, V.; Parak, W. J.; Mahmoudi, M., Antibacterial properties of nanoparticles. Trends Biotechnol. 2012, 30, 499-511.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(3)

(4)

(5)

(6)

(7)

(8)

(9)

(10)

(11)

(12)

(13)

(14)

(15) (16)

(17)

Beyth, N.; Houri-Haddad, Y.; Domb, A.; Khan, W.; Hazan, R., Alternative antimicrobial approach: Nano-antimicrobial materials. J. Evidence-Based Complementary Altern. Med. 2015, Article ID 246012, 2015. Singh, R.; Shedbalkar, U. U.; Wadhwani, S. A.; Chopade, B. A., Bacteriagenic silver nanoparticles: synthesis, mechanism, and applications. Appl. Microbiol. Biotechnol. 2015, 99, 4579-4593. Ravishankar, R.; Jamuna, B. Nanoparticles and their potential application as antimicrobials. In Science against microbial pathogens: communicating current research and technological advances, Mendez-Vilas, A. Eds. Formatex Research Center; 2011; pp 197-209 Moritz, M.; Geszke-Moritz, M., The newest achievements in synthesis, immobilization and practical applications of antibacterial nanoparticles. Chem. Eng. J. 2013, 228, 596-613. Kim, J. S.; Kuk, E.; Yu, K. N.; Kim, J.-H.; Park, S. J.; Lee, H. J.; Kim, S. H.; Park, Y. K.; Park, Y. H.; Hwang, C.-Y.; Kim, Y.-K.; Lee, Y.-S.; Jeong, D. H.; Cho, M.-H., Antimicrobial effects of silver nanoparticles. Nanomedicine. 2007, 3, 95-101. Zhou, Y.; Kong, Y.; Kundu, S.; Cirillo, J. D.; Liang, H., Antibacterial activities of gold and silver nanoparticles against Escherichia coli and bacillus CalmetteGuérin. J. Nanobiotechnol. 2012, 10:19 Liu, S.; Zeng, T. H.; Hofmann, M.; Burcombe, E.; Wei, J.; Jiang, R.; Kong, J.; Chen, Y., Antibacterial activity of graphite, graphite oxide, graphene oxide, and reduced graphene oxide: Membrane and oxidative stress. ACS Nano. 2011, 5, 6971-6980. Krishnamoorthy, K.; Veerapandian, M.; Zhang, L. H.; Yun, K.; Kim, S. J., Antibacterial efficiency of graphene nanosheets against pathogenic bacteria via lipid peroxidation. J. Phys. Chem. C. 2012, 116, 17280-17287. Shanmugasundaram, T.; Radhakrishnan, M.; Gopikrishnan, V.; Pazhanimurugan, R.; Balagurunathan, R., A study of the bactericidal, antibiofouling, cytotoxic and antioxidant properties of actinobacterially synthesised silver nanoparticles. Colloids Surf., B. 2013, 111, 680-687. Bondarenko, O.; Ivask, A.; Käkinen, A.; Kahru, A., Sub-toxic effects of CuO nanoparticles on bacteria: Kinetics, role of Cu ions and possible mechanisms of action. Environ. Pollut. 2012, 169, 81-89. Marambio-Jones, C.; Hoek, E. M. V., A review of the antibacterial effects of silver nanomaterials and potential implications for human health and the environment. J. Nanopart. Res. 2010, 12, 1531-1551. Chatzimitakos, T.; Kallimanis, A.; Avgeropoulos, A.; Stalikas, C. D., Antibacterial, anti-biofouling and antioxidant prospects of metal based nanomaterials. CLEAN – Soil, Air, Water. 2016, DOI: 10.1002/clen.201500366. Kohanski, M. A.; Dwyer, D. J.; Collins, J. J., How antibiotics kill bacteria: from targets to networks. Nat. Rev. Microbiol. 2010, 8, 423-435. Liebeke, M.; Lalk, M., Staphylococcus aureus metabolic response to changing environmental conditions – A metabolomics perspective. Int. J. Med. Microbiol. 2014, 304, 222-229. Somerville, G. A.; Proctor, R. A., At the crossroads of bacterial metabolism and virulence factor synthesis in staphylococci. Microbiol. Mol. Biol. Rev. 2009, 73, 233-248.


Page 22 of 31

Page 23 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(18) Eisenreich, W.; Dandekar, T.; Heesemann, J.; Goebel, W., Carbon metabolism of intracellular bacterial pathogens and possible links to virulence. Nat. Rev. Microbiol. 2010, 8, 401-412. (19) Aros-Calt, S.; Muller, B. H.; Boudah, S.; Ducruix, C.; Gervasi, G.; Junot, C.; Fenaille, F., Annotation of the Staphylococcus aureus metabolome using liquid chromatography coupled to high-resolution mass spectrometry and application to the study of methicillin resistance. J. Proteome Res. 2015, 14, 4863-4875. (20) Mousavi, F.; Bojko, B.; Bessonneau, V.; Pawliszyn, J., Cinnamaldehyde Characterization as an Antibacterial Agent toward E. coli Metabolic Profile Using 96-Blade Solid-Phase Microextraction Coupled to Liquid Chromatography-Mass Spectrometry. J. Proteome Res. 2016, 15, 963-975. (21) Goodacre, R.; Vaidyanathan, S.; Dunn, W. B.; Harrigan, G. G.; Kell, D. B., Metabolomics by numbers: Acquiring and understanding global metabolite data. Trends Biotechnol. 2004, 22, 245-252. (22) Fiehn, O., Combining genomics, metabolome analysis, and biochemical modelling to understand metabolic networks. Comp. Funct. Genomics. 2001, 2, 155-168. (23) Tang, J., Microbial Metabolomics. Curr. Genomics. 2011, 12, 391-403. (24) Winder, C. L.; Dunn, W. B.; Schuler, S.; Broadhurst, D.; Jarvis, R.; Stephens, G. M.; Goodacre, R., Global Metabolic Profiling of Escherichia coli Cultures: an Evaluation of Methods for Quenching and Extraction of Intracellular Metabolites. Anal. Chem. 2008, 80, 2939-2948. (25) Bertini, I.; Hu, X.; Luchinat, C., Global metabolomics characterization of bacteria: pre-analytical treatments and profiling. Metabolomics 2014, 10, 241249. (26) Werf, M. J. v. d.; Overkamp, K. M.; Muilwijk, B.; Coulier, L.; Hankemeier, T., Microbial metabolomics: Toward a platform with full metabolome coverage. Anal. Biochem. 2007, 370, 17-25. (27) Djukovic, D.; Nagana Gowda, G. A.; Raftery, D., Chapter 18 - Mass Spectrometry and NMR Spectroscopy–Based Quantitative Metabolomics. In Proteomic and Metabolomic Approaches to Biomarker Discovery, Veenstra, H. J. I. D., Ed. Academic Press: Boston, 2013; pp 279-297. (28) Robertson, D. G.; Reily, M. D.; Baker, J. D., Metabonomics in pharmaceutical discovery and development. J. Proteome Res. 2007, 6, 526-539. (29) Xu, Y.-J.; Wang, C.; Ho, W. E.; Ong, C. N., Recent developments and applications of metabolomics in microbiological investigations. TrAC, Trends Anal. 2014, 56, 37-48. (30) Nagana Gowda, G. A.; Raftery, D., Can NMR solve some significant challenges in metabolomics? J. Magn. Reson. 2015, 260, 144-160. (31) Misra, B. B.; van der Hooft, J. J. J., Updates in metabolomics tools and resources: 2014-2015. Electrophoresis. 2016, 37, 86-110. (32) Sotelo, J.; Slupsky, C. M., 2 - Metabolomics using nuclear magnetic resonance (NMR). In Metabolomics in Food and Nutrition, Weimer, B. C.; Slupsky, C., Eds. Woodhead Publishing; 2013; pp 29-43. (33) Ye, Y.-F.; Zhang, L.-M.; An, Y.-P.; Hao, F.-H.; Tang, H.-R., Nuclear Magnetic Resonance for Analysis of Metabolite Composition of Escherichia Coli. Chinese J. Anal. Chem. 2011, 39, 1186-1194. (34) Chatzimitakos, T.; Exarchou, V.; Ordoudi, S. A.; Fiamegos, Y.; Stalikas, C., Ion-pair assisted extraction followed by 1H NMR determination of biogenic amines in food and biological matrices. Food Chem. 2016, 202, 445-450 23 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(35) Xiao, J. F.; Zhou, B.; Ressom, H. W., Metabolite identification and quantitation in LC-MS/MS-based metabolomics. TrAC, Trends Anal. Chem. 2012, 32, 1-14. (36) Watson, D. G., A rough guide to metabolite identification using high resolution liquid chromatography mass spectrometry in metabolic profiling of metazoans. Comput. Struct. Biotechnol. J. 2013, 4, 1-10. (37) Theodoridis, G.; Gika, H. G.; Wilson, I. D., LC-MS-based methodology for global metabolite profiling in metabonomics/metabolomics. TrAC, Trends Anal. Chem. 2008, 27, 251-260. (38) Karamani, A. A.; Fiamegos, Y. C.; Vartholomatos, G.; Stalikas, C. D., Fluoroacetylation/fluoroethylesterification as a derivatization approach for gas chromatography–mass spectrometry in metabolomics: Preliminary study of lymphohyperplastic diseases. J. Chromatogr. A. 2013, 1302, 125-132. (39) Vincent, I. M.; Ehmann, D. E.; Mills, S. D.; Perros, M.; Barrett, M. P., Untargeted metabolomics to ascertain antibiotic modes of action. Antimicrob. Agents Chemother. 2016, 60, 2281-2291. (40) Dörries, K.; Schlueter, R.; Lalk, M., Impact of antibiotics with various target sites on the metabolome of Staphylococcus aureus. Antimicrob. Agents Chemother. 2014, 58, 7151-7163. (41) Kumar, B.; Prakash, A.; Ruhela, R. K.; Medhi, B., Potential of metabolomics in preclinical and clinical drug development. Pharmacol. Rep. 2014, 66, 956-963. (42) Marcinowska, R.; Trygg, J.; Wolf-Watz, H.; Mortiz, T.; Surowiec, I., Optimization of a sample preparation method for the metabolomic analysis of clinically relevant bacteria. J. Microbiol. Methods. 2011, 87, 24-31. (43) Guo, A. C.; Jewison, T.; Wilson, M.; Liu, Y.; Knox, C.; Djoumbou, Y.; Lo, P.; Mandal, R.; Krishnamurthy, R.; Wishart, D. S., ECMDB: The E. coli Metabolome Database. Nucleic Acids Res. 2013, 41, D625-D630. (44) Sajed, T.; Marcu, A.; Ramirez, M.; Pon, A.; Guo, A. C.; Knox, C.; Wilson, M.; Grant, J. R.; Djoumbou, Y.; Wishart, D. S., ECMDB 2.0: A richer resource for understanding the biochemistry of E. coli. Nucleic Acids Res. 2016, 44, D495D501. (45) Xia, J.; Sinelnikov, I. V.; Han, B.; Wishart, D. S., MetaboAnalyst 3.0—making metabolomics more meaningful. Nucleic Acids Res. 2015, 43, W251-W257. (46) Xia, J.; Mandal, R.; Sinelnikov, I. V.; Broadhurst, D.; Wishart, D. S., MetaboAnalyst 2.0—a comprehensive server for metabolomic data analysis. Nucleic Acids Res. 2012, 40, W127-W133. (47) Xia, J.; Psychogios, N.; Young, N.; Wishart, D. S., MetaboAnalyst: a web server for metabolomic data analysis and interpretation. Nucleic Acids Res. 2009, 37, W652-W660. (48) Russell, A. D., Bacterial resistance to disinfectants: Present knowledge and future problems. J. Hosp. Infection. 1999, 43, S57-S68. (49) Shimizu, K., Metabolic regulation and coordination of the metabolism in bacteria in response to a variety of growth conditions. In Adv. Biochem. Eng./Biotechnol. 2016, 155, 1-54. (50) Merlo, M.; Jankevics, A.; Takano, E.; Breitling, R., Exploring the metabolic state of microorganisms using metabolomics. Bioanalysis. 2011, 3, 2443-2458. (51) Tremaroli, V.; Workentine, M. L.; Weljie, A. M.; Vogel, H. J.; Ceri, H.; Viti, C.; Tatti, E.; Zhang, P.; Hynes, A. P.; Turner, R. J.; Zannoni, D., Metabolomic investigation of the bacterial response to a metal challenge. Appl. Environ. Microbiol. 2009, 75, 719-728.


Page 24 of 31

Page 25 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(52) Yanes, O.; Tautenhahn, R.; Patti, G. J.; Siuzdak, G., Expanding Coverage of the Metabolome for Global Metabolite Profiling. Anal. Chem. 2011, 83, 2152-2161. (53) Fong, S. S.; Nanchen, A.; Palsson, B. O.; Sauer, U., Latent pathway activation and increased pathway capacity enable Escherichia coli adaptation to loss of key metabolic enzymes. J. Biol. Chem. 2006, 281, 8024-8033. (54) Booth, S. C.; Weljie, A. M.; Turner, R. J., Metabolomics reveals differences of metal toxicity in cultures of Pseudomonas pseudoalcaligenes KF707 grown on different carbon sources. Front. Microbiol. 2015, 6, 827. (55) Booth, S. C.; Workntine, M. L.; Shaykhutdinov, R.; Vogel, H. J.; Ceri, H.; Turner, R. J.; Weljie, A. M., Differences in metabolism between the biofilm and planktonic response to metal stress. J. Proteome Res. 2011, 10, 2190-3199.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

FIGURE LEGENDS Figure 1: Workflow to obtain untargeted metabolomic fingerprint data for E. coli and S. aureus. Figure 2: 1H NMR spectra of the extracted metabolites from E. coli. Black-colored spectra represent the control samples. Spectra received after exposure to Cu, Fe and Fe-Cu NPs are depicted in purple, green and red, respectively. Figure 3: 1H NMR spectra of the extracted metabolites from S. aureus. Black-colored spectra represent the control samples. Spectra received after exposure to Cu, Fe and Fe-Cu NPs are depicted in purple, green and red, respectively. Figure 4: Venn diagrams for E. coli (A) and S. aureus (B) metabolic pathways.


Page 26 of 31

Page 27 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


NMR

ECMDB

• 1Η NMR spectra of the samples are acquired

• NMR signal positions, in ppm, are input to the 1D NMR search engine of the ECMDB 2.0 and processed in order to obtain qualitative data

UHPLC-MS

• MS spectra of the samples are acquired

Verification

• MS spectra are scanned for the exact masses (4 decimals) of the metabolites resulting from the ECMDB NMR search engine and their presence in the sample is verified

MetaboAnalyst

• A list of the verified metabolites is uploaded to the Pathway analysis tool of MetaboAnalyst 3.0 and processed

Metabolic Pathways

• A list of the metabolic pathways is obtained

Figure 1



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48

7.5 – 10 ppm

5 – 7.5 ppm

3 – 5 ppm

Figure 2

28

ACS Paragon Plus Environment

Page 28 of 31

0 – 3 ppm

Page 29 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48


7– 10 ppm

5 – 7 ppm

3 – 5 ppm

Figure 3

29


0 – 3 ppm


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 30 of 31

A

B

Figure 4


Page 31 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


For TOC only 489x125mm (300 x 300 DPI)


Qualitative Alterations of Bacterial Metabolome after Exposure to Metal

Recommend Documents