1H NMR-Based Global Metabolic Studies of Pseudomonas

Subscriber access provided by UNIV OF CALIFORNIA SAN DIEGO LIBRARIES

Article

1H-NMR-Based Global Metabolic Studies of Pseudomonas aeruginosa upon Exposure of the Quorum Sensing Inhibitor Resveratrol Tongtong Chen, Jiyang Sheng, Yonghong Fu, Ming-Hui Li, Jun-Song Wang, and Aiqun Jia J. Proteome Res., Just Accepted Manuscript • DOI: 10.1021/acs.jproteome.6b00800 • Publication Date (Web): 17 Jan 2017 Downloaded from http://pubs.acs.org on January 20, 2017

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 24

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

1

H-NMR-Based Global Metabolic Studies of Pseudomonas aeruginosa upon Exposure of the Quorum Sensing Inhibitor Resveratrol

Tongtong Chen a#, Jiyang Sheng a#, Yonghong Fu a, Minghui Li a, Junsong Wang a*, Ai-Qun Jia b,a*

a

School of Environmental and Biological Engineering, Nanjing University of Science

and Technology, Xiao Ling Wei No. 200, Nanjing 210094, China b

State Key Laboratory of Marine Resource Utilization in South China Sea, Hainan

University, Haikou 570228, China

*Corresponding authors: E-mail: [email protected] (Ai-Qun Jia); [email protected] (Junsong Wang). Telephone: +86 25 84303216. Fax: +86 25 84303216.

#These authors contributed equally to the work.

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

Quorum sensing (QS) is a process of bacterial communication that has been a novel target for drug discovery. Pyocyanin quantification assay confirmed that resveratrol was an effective quorum sensing inhibitor (QSI) against Pseudomonas aeruginosa PAO1. In this study, the global metabolite changes of P. aeruginosa PAO1 exposed to QSI resveratrol were investigated by 1H-NMR spectroscopy. A total of 40 metabolites containing amino acids, organic acid, organic amine and energy storage compounds were identified. The changed metabolic profile indicated that resveratrol influenced pathways including oxidative stress, protein synthesis and energy metabolism. Oxidative stress could upregulate the expression of genes related to QS in P. aeruginosa. It suggested that resveratrol could inhibit the QS systems in P. aeruginosa PAO1 by relieving oxidative stress due to its antioxidant activity. On the other hand, resveratrol could attenuate the pathogenicity of P. aeruginosa PAO1 by disturbing the TCA cycle so that anaerobic respiration could suppress the virulence since anaerobiosis could induce the loss of cytotoxicity regulated by QS in P. aeruginosa. These findings deepened our comprehending of the metabolic responses of P. aeruginosa PAO1 to resveratrol and pinpointed the possible underlying mechanism of resveratrol’s inhibition effect on QS in P. aeruginosa PAO1. KEYWORDS: resveratrol, Pseudomonas aeruginosa, quorum sensing, 1H-NMR, metabolomics

2


Page 2 of 24

Page 3 of 24

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

Quorum sensing (QS), one type of regulatory mechanism concerning the cell-to-cell communication, is widely adopted by various bacteria species1 in which bacteria coordinate their cell-density-dependent gene expression by producing diffusible signal molecules2. A great many of Gram-negative bacteria utilize LuxI/LuxR-type QS systems and employ N-Acyl homoserine lactones (AHLs) as signal molecules called autoinducers (AIs)3. When AHLs synthesized by LuxI synthases reach a threshold concentration in accordance with bacterial cell density, they form the transcriptional complex with LuxR receptors. The transcriptional complex then binds to DNA sequence and activates the expression of corresponding genes controlling biofilm formation, bioluminescence and virulence factors, etc. Additionally, the AHL-LuxR complex can also activate luxI expression to increase the production of AIs4. Pseudomonas aeruginosa, one important opportunistic human pathogen, is common in nosocomial infections that causes acute or chronic infections in immunocompromised patients with high morbidity and mortality5. P. aeruginosa uses QS to collectively produce many virulence factors such as elastase, alkaline protease, rhamnolipids and pyocyanin6. The formation of biofilm resulting in the resistance to antibiotics of P. aeruginosa is also relevant to QS7. There are two QS systems based on AHL in P. aeruginosa including the Las and Rhl systems. The Las system utilizes N-(3-oxododecanoyl)-L-homoserine lactone (3-oxo-C12-HSL) as signal molecule and the Rhl system uses N-butanoyl-L-homoserine lactone (C4-HSL) synthesized by LasI 3



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

and RhlI synthases respectively. Besides these two AHL-based QS systems, there is also a quinolone signal (PQS) system in P. aeruginosa. These three systems are hierarchically arranged, with the Rhl and PQS systems regulated by the Las system8. Owing to the abuse of antibacterial drugs, antibiotic-resistance of P. aeruginosa is increasing, an in-depth study of the development of novel and potent drugs known as quorum sensing inhibitors (QSIs) which focus on inhibition of QS systems other than killing bacteria thus possesses a profound significance9. Different from ordinary antibacterial drugs, QSIs target the pathogenicity itself instead of inhibiting the growth of bacteria, successfully avoid selective pressure which would finally result in drug resistance10. Although a lot of efforts have been made to explore the effects and mechanisms of QSIs of P. aeruginosa such as halogenated furanones11, garlic extracts12 and N-decanoyl-L-homoserine benzyl ester (C2)13, little is known about their influence on metabolic profiles and the underlying inhibition mechanism. As a new rising omics technology, metabolomics has been extensively applied to many fields such as toxicology, disease diagnosis and nutrition. Among various analytical methods in metabolomics applied, including liquid chromatography-mass spectrometry (LC-MS), gas chromatography-mass spectrometry (GC-MS) and nuclear magnetic resonance (NMR), NMR is the most widely used technique because of its associated advantages: easy sample preparation, fast, unbiased and non-destructive analysis, and rich in structural information14. On the basis of 1H-NMR technique, metabolomics has been introduced to explore the global metabolisms of bacteria15,16. 4


Page 4 of 24

Page 5 of 24

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


Resveratrol, a naturally occurring phytoalexin existing in some medicinal and edible plants, has been found to potentially exhibit antioxidant, anti-inflammatory and anticancer activities17. In our previous research, resveratrol has been proved to be an effective QSI against P. aeruginosa PAO118, while nothing was known about the potential mechanism concerning the QSI-inhibition effect of resveratrol to the best of our knowledge. Hereby, a 1H-NMR-based metabolomics method was employed firstly to study the QSI-inhibition effect of resveratrol on the metabolic changes in P. aeruginosa PAO1 to assess the underlying mechanism. Our results indicated that oxidative stress, disturbances of protein synthesis and energy metabolism were involved in the QSI activity of resveratrol.

MATERIALS AND METHODS Chemicals Resveratrol was isolated from a traditional Chinese medicine plant, Smilax china L by our group, and was dissolved in DMSO at the appropriate concentration. All other chemical materials applied in this research were supplied by Sigma Chemical Co. (St. Louis, MO, USA). Bacterial Strains and Culture Conditions Pseudomonas aeruginosa PAO1 was kindly provided by Prof. Q. H. Gong of Ocean University of China in Qingdao. P. aeruginosa PAO1 were cultivated in Nutrient Broth (NB) at 37 oC. Pyocyanin Quantification Assay According to our previous report18, 5 mL overnight bacteria culture supplied with 5



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

400 µM resveratrol from each flask was extracted with 3 mL chloroform and then the organic layer was re-extracted with 1 mL 0.2 M HCl. The absorbance of the supernatant was measured using a microplate reader at 520 nm. Metabolite Extraction Procedures Overnight P. aeruginosa PAO1 cultures were diluted (1:1000) into 30 mL of fresh NB, resveratrol was added at 400 µM as treated group, while the same amount of DMSO was used in the control group. 13 biological replicates were used for treated group and control group respectively. After incubation with shaking at 180 rpm for 16-18 h, the cell cultures were chilled by a brief incubation on ice, and then the cell pellet was obtained by an 8-min centrifugation (10,000×g, 4 oC). At the same time, the cultures were used for pyocyanin quantification assay. After washing three times with PBS (phosphate buffered saline), the cell pellets were transferred into a microtube (10 mL) and mixed with 3.8 mL pre-cooled methanol/water (1/0.9; v/v, 4 oC). Mixtures were then extracted by intermittent sonication (i.e., 2 s sonication with 2 s break) for 5 min under the ice bath, after which 4 mL chloroform was added. After vigorously vortexing, the mixtures were centrifuged (10,000×g, 4 oC) for 8 min. To remove methanol, the supernatants were transferred into new centrifuge tubes and treated under vacuum with a Speed-Vac Concentrator (Thermo SAVANT, SC110A-230). The obtained supernatants were stored at -80 oC overnight and then lyophilized in a freeze-drier. All groups were performed under the same extraction process. For 1

H-NMR analysis, the lyophilized extracts were reconstituted with PBS (pH 7.4), with

99.8% D2O and 0.05% (w/v) sodium 3-(trimethylsilyl) propionate-2,2,3,3-d4 (TSP) for referencing purposes. The mixture solutions were vortexed and then centrifugated to discard undesired sediments. The afforded supernatants were shifted into clean NMR tubes for the following 1H-NMR analysis tests. 6


Page 6 of 24

Page 7 of 24

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 Measurements Following M.H. Li et al, 1H-NMR spectra were recorded on a Bruker-AV500 MHz spectrometer at 298 K, adopting the “CPMG” (Call-Purcell-Meiboom-Gill) pulse sequence (RD-90 (τ-180-τ) n-ACQ). D2O was the solvent intended for field frequency locking and TSP was internal standard served as a chemical shift reference. 1

H-NMR spectra were measured with 128 scans into 32 K data points by a spectral

width of 10,330 Hz, with 3.27 s acquisition time and 3.0 s delay time relation. All spectral were adjusted and accepted the line broadening of 0.5 Hz before Fourier transformation19. Data Pre-processing and Peak Assignments Before analysis, operations including baseline adjustment and phasing control to the initial 1H-NMR spectra were carefully carried out after zeroing adjustment with the peak of TSP as 0.0 ppm by means of the software of Bruker Topspin 3.0 (version 2.1). Having finished the previous preparations above, we exported the processed spectra to ASCII format files using MestReNova (Version 8.0.1, Mestrelab Research SL),

after

that

the

generated

data

were

read

into

‘‘R’’

software

(http://cran.r-project.org/) for the following multivariate analysis. The spectra between 0.2 to 10 ppm were segmented with an average binning of 0.005 ppm. The residual resonance of water as well as its influenced regions between 4.5 and 5.0 ppm were cut off. All spectra were then normalized and mean-centered before multivariate statistical analysis. The metabolite assignment was made by referring publicly accessible metabolomics

databases

such

as

Human

Metabolome

Database

(HMDB,

http://www.hmdb.ca) and Madison-Qingdao Metabolomics Consortium Database (MMCD, http://mmcd.nmrfam.wisc.edu/), associated with Chenomx NMR suite 7.5 7



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

(Chenomx Inc., Edmonton, Canada) and the statistical total correlation spectroscopy (STOCSY) technique14. The obtained results were initiatory and still need to be affirmed, so to further confirm the primary metabolite identification result, TOCSY (total correlation spectroscopy) and HSQC (heteronuclear single quantum correlation) were applied. Multivariate Data Analysis Multivariate statistical analysis, including unsupervised principal component analysis (PCA) and supervised orthogonal signal correction partial least-squares discriminant analysis (OSC-PLS-DA) method were applied to NMR data performing20. Unsupervised PCA was firstly adopted to reduce the dimensionality of imported NMR data by which new latent variables namely principal components were achieved, such components were smaller than variables before transforming. OSC-PLS-DA, one supervised method, was then introduced to filter out irrelevant effects and maximize the discrimination of intergroup differences. The orthogonal signal correction (OSC) was a technique that applied prior to PLS-DA to filter out unrelated variables that were not concerning the class discrimination so as to minimize the influence of unrelated signals. Repeated two-fold cross-validation and permutation test (n=2,000) were conducted to obtain two values named R2 and Q2 which explained the total variation and model predictability, respectively, thus were deemed as crucial makers to verify the validity of constructed models. P value, another key parameter obtained from permutation testing, reflected the significance of the established model, particularly, when p value was less than 0.05, we could confirm the model at a 95% confidence level. Variables that contributed to the group separation were assigned and visualized by S-plot and color-coded loadings plots. In addition, receiver operating characteristic (ROC) curves were applied to validate 8


Page 8 of 24

Page 9 of 24

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


classifier performance of the established OSC-PLS-DA models after 200 times repeated 2-fold cross-validation. Univariate Analysis Univariate analysis, including parametric Student’s t-test and non-parametric Mann-Whitney test14 were performed to verify the alteration of crucial metabolites. The fold change values of the identified metabolites as well as their p values between groups were calculated. The obtained p values were then adjusted by the Benjamini & Hochberg method with the purpose of controlling the false discovery rate when proceeding with multiple comparisons. RESULTS Pyocyanin Quantification Assay The production of virulence factor pyocyanin which is regulated by QS was used to assess the anti-QS performance of resveratrol. At 400 µM, the production of pyocyanin was inhibited by resveratrol with a significant reduction (P < 0.001) (Figure 1).

9



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 1. Effects of 400 µM resveratrol on the production of pyocyanin in P. aeruginosa PAO1. 1

H-NMR Spectra of the P. aeruginosa PAO1 Extracts Representative 1H-NMR spectra of P. aeruginosa PAO1 extracts in resveratrol

treated group and control group were shown in Figure 2 with metabolites labeled: a total of 40 metabolites were identified with the assistance of HSQC (Figure S-1A) and TOCSY (Figure S-1B), including amino acids, organic acid, organic amine and energy storage compounds. The details concerning the identified metabolites, including the assignments and fold change calculation were listed in Table 1. Metabolic Changes due to Resveratrol The PCA score plot (Figure 3) for P. aeruginosa PAO1 exhibited a clear discrimination between resveratrol treated group and control group which suggested that these two may have distinct metabolomic processes. To overall examine the variations between resveratrol treated group and control group, supervised OSC-PLS-DA was performed and the classifier performance of the established OSC-PLS-DA models were verified by receiver operating characteristic (ROC) curves (Figure S-2). The score plots exhibited well separation between groups with a high goodness of fit (R2 = 0.90, Q2 = 0.76) (Figure S-3A) and statistical significance (p = 0.029) (Figure S-3B). In the S-plot (Figure 4B), differential metabolites were shown in points with different shapes and colors and contributions of those metabolites to the grouping were related to their distance to the center: variables that were further from

10


Page 10 of 24

Page 11 of 24

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


the center, gaining more significant contributions to the group separation. On the basis of correlation coefficients, the loadings plots (Figure 4C and D) were coded with cool color and warm color tone, from blue to red, the relativity gradually enhanced. Significant increase of isoleucine, leucine, valine, acetate, acetamide, succinate, aspartate, dimethylamine, sarcosine, trimrthylamine, ethanolamine, methanol, serine and formate, and marked decrease of 2-hydroxyisobutyrate, N-acetylglutamine, glutamine, 2-oxoglutarate, betaine, fumarate and oxypurinol were found in the resveratrol dosed group. The assigned metabolites and their fold change values as well as p values were displayed in Table 1.

11



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 2. Typical 500 MHz CPMG 1H-NMR spectra of P. aeruginosa PAO1 extracts from resveratrol treated group (black line) and control group (red line). Labeled metabolites: 1 Cholate, 2 Isoleucine, 3 Leucine, 4 Valine, 5 Lactate, 6 2-hydroxyisobutyrate, 7 Alanine, 8 Lysine, 9 Putrescine, 10 Acetate, 11 Ornithine, 12 Acetamide, 13 N-acetylglutamine, 14 Glutamate, 15 N-acetylcysteine, 16 Methionine, 12


Page 12 of 24

Page 13 of 24

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


17 Succinate, 18 2-Oxoglutarate, 19 Methylamine, 20 Aspartate, 21 Dimethylamine, 22 Sarcosine, 23 Trumethylamine, 24 Ethanolamine, 25 Arginine, 26 Choline, 27 Betaine, 28 Methanol, 29 Glycine, 30 Serine, 31 UDP-galactose, 32 UDP-glucose, 33 Maleate, 34 NADP+, 35 Fumarate, 36 Tyrosine, 37 Histidine, 38 UDP-glucuronate, 39 AMP, 40 Formate. Table 1. Important metabolites assignments in P. aeruginosa PAO1, their fold change values and associated p-values. Metabolites

Assignments

Chemical shifta(ppm)

Foldb

Pc

Cholate

CH3

0.73(s)

0.93

*

Isoleucine

δ-CH3, γ-CH3, α-CH

0.91(t), 1.00(d), 3.66(d)

1.16

**

Leucine

δ-CH3, δ-CH3, α-CH

0.94(t), 0.95(t),

1.34

*

Valine

γ-CH3, γ-CH3, α-CH

0.97(d), 1.05(d), 3.60(d)

1.24

*

Lactate

CH3

1.33(d)

1.08

2-Hydroxyisobutyrate

CH3

1.36(s)

0.75

Alanine

CH3, CH

1.48(d), 3.77(q)

1.19

Lysine

δ-CH2, β-CH2, ε-CH2, α-CH

1.72(m), 1.91(m), 3.00(t), 3.76(t)

0.93

Putrescine

CH3, CH2N2

1.78(m), 3.06(m)

0.98

Acetate

CH3

1.92(s)

1.29

Ornithine

CH2

1.96(m)

1.3

Acetamide

CH3

2.0(s)

1.15

N-Acetylglutamine

CH3, CH2

2.03(s), 2.30(t), 2.32(t)

0.86

***

Glutamate

β-CH2, γ-CH2, α-CH

2.06(m), 2.10(m), 2.33(m), 2.37(m), 3.77(m)

0.85

***

N-Acetylcysteine

CH3, CH2

2.08(s), 2.90(m), 2.94(d)

0.99

Methionine

CH3, CH2

2.14(s), 2.63(t)

0.93

***

**

Acetoacetate

CH3, CH2

2.29(s), 3.43(s)

1.02

Succinate

CH2

2.41(s)

1.45

2-Oxoglutarate

CH2,CH2

2.45(t), 3.00(t)

0.89

Methylamine

CH3

2.59(s)

0.92

Aspartate

CH2

2.68(dd), 2.82(dd)

1.23

***

Dimethylamine

CH3

2.72(s)

1.1

***

Sarcosine

CH3

2.73(s)

1.16

***

Trimethylamine

CH3

2.89(s)

1.26

** ***

Ethanolamine

CH2, CH2

3.15(t), 3.82(s)

1.11

Arginine

CH3

3.22(t)

1.08

Choline

CH3

3.20(s)

1.03

Betaine

CH2, CH3

3.27(s), 3.89(s)

0.56

13


***

***


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

Methanol

Page 14 of 24

CH3

3.36(s)

1.25

Glycine

CH2

3.57(s)

1.18

Serine

CH, CH2

3.84(dd), 3.94(m), 3.98(m)

1.17

Ribose-5-phosphate

CH

5.57(m)

1.07

Cytosine

CH

5.95(d)

1.28

UDP-glucose

CH, CH, CH

5.97(d), 5.98(d), 7.96(d)

1.07

Maleate

CH=CH

6.00(s)

1.05

Fumarate

CH

6.52(s)

0.91

Oxypurinol

CH

8.22(s)

1.34

AMP

CH, CH, CH

6.15(d), 8.27(s), 8.62(s)

0.92

Formate

CH

8.46(s)

1.21

a

Multiplicity: (s) singlet, (d) doublet, (t) triplet, (q) quartets, (m) multiplets.

b

Color coded according to the log2(fold), red and blue represented the increased and

decreased metabolites respectively in resveratrol treated group. c

P-values were calculated based on a parametric Student’s t-test or a nonparametric

Mann-Whitney test and were corrected by BH (Benjamini Hochberg) methods. Numbers of symbol * denoted extent of significance: * P < 0.05, ** P < 0.01, *** P < 0.001.

Figure 3. PCA Score plot of 1H-NMR data of P. aeruginosa PAO1. Two PCs 14


***

***

***

*

Page 15 of 24

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


explained 28.14% and 14.2% of total variances in P. aeruginosa PAO1. The ellipses represent the 95% confidence interval for each group.

Figure 4. OSC-PLS-DA analysis of NMR data from P. aeruginosa PAO1. (A) Score plot. Component 1 and component 2 explained 39.6% of total variances in the extracts of P. aeruginosa PAO1 samples. (B and C) Color-coded loadings plots. The color bar was applied in which red and blue color represented metabolites that statistically significantly or indistinctively contributed to the separation of groups, respectively. Peaks in positive and negative status revealed decreased and increased metabolites relative to score plot in the resveratrol treated group. (D) S-plot.

DISCUSSION

The virulence of P. aeruginosa PAO1 is ascribed to its ability to produce many kinds of virulence factors such as rhamnolipids, pyocyanin, LasB elastase, exotoxin A, hydrogen cyanide and cytotoxic lectins, which can degrade the host tissues21. These virulence factors are under the regulation of the QS systems22. Previously, we reported 15



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

that resveratrol could inhibit the QS system of P. aeruginosa PAO118. Whether resveratrol, as a QSI, has influence on metabolic system in P. aeruginosa PAO1 has never been reported. In this study, pyocyanin quantification assay further confirmed that resveratrol was an effective QSI against P. aeruginosa PAO1. The effects of resveratrol on metabolic system in P. aeruginosa PAO1 were deeply investigated using

1

H-NMR-based metabolomics approach. Results showed that resveratrol

evidently disturbed the metabolic profiles of P. aeruginosa PAO1. Multivariate statistical analysis revealed that some altered metabolites were related with oxidative stress as well as disturbances of protein synthesis and energy metabolism. To assess resveratrol as one potential QSI, we focused on its capacity for disturbing QS system, and at the same time, we also paid attention to its effect on bacterial growth. Compared with the control group, the content of putrescine was almost not changed after resveratrol treatment. As reported before, level of putrescine in vivo was an indicator of the bacterial growth status, and in E. coli cells, putrescine played an important role in cell division23. Our previous reports found that, at 400 µM, resveratrol had nearly no growth-hindering effect on P. aeruginosa PAO1 by growth curve analysis18. The metabolomics result was in consistent with our previous work, providing another evidence that resveratrol had no inhibition on P. aeruginosa PAO1 growth, which assured resveratrol could be a promising QSI: effective inhibitor of QS with little influence on bacterial growth. As an effective antioxidant in vivo, betaine plays a crucial role in counteracting against oxidative damage and maintaining the cell membrane with integrated structure 16


Page 16 of 24

Page 17 of 24

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


and function. The significant decreased level of betaine could be ascribed to its excessive consumption to counteract free radicals and repair the damaged membrane caused by reactive oxygen species (ROS). In normal cell, ROS are generated physiologically, hence high cell density would inevitably result in the excessive production and its resultant accumulation of ROS. Such oxidative stress was also supported by the increased level of ethanolamine. As crucial component of cellular membranes which is susceptible to oxidative damage24, the increase of ethanolamine level effectively indicated the impaired integrity of cell membrane under the attack of free radicals and ROS. It is generally believed that QS is dependent on cell density, while there are also other evidences suggesting that environmental factors should also be other key elements that influence the QS systems except for cell density. Recently, Juhas et al. have found that oxidative stress could upregulate the expression of genes in P. aeruginosa controlled by QS25. In consideration of antioxidant activity of resveratrol, we could propose that resveratrol relieves oxidative stress, and then inhibits the QS system in P. aeruginosa. Compared with the control group, branched chain amino acids (BCAAs) valine, leucine and isoleucine were dramatically increased in P. aeruginosa PAO1 treated with resveratrol. BCAAs, essential amino acids in vivo, act as vital substrates and regulators of protein synthesis, their increases suggested a breakdown of normal protein synthesis owing to the affected QS in P. aeruginosa PAO1 after dosing resveratrol. Considering that many virulence factors controlled by QS are proteins, the significant increased level of BCAAs would indicate the attenuation of pathogenicity 17



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

in P. aeruginosa PAO1. Resveratrol also introduced disturbance of energy metabolism in P. aeruginosa PAO1, which was evidenced by the significant increase of succinate and decrease of fumarate. Both of succinate and fumarate are intermediates of the tricarboxylic acid (TCA) cycle, under normal physiological conditions, succinate can be converted to fumarate catalyzed by succinate dehydrogenase. The increase of succinate and decrease of fumarate thus reflected an altered activity of succinate dehydrogenase and concomitantly blocked TCA cycle. Futhermore, marked decreases of 2-oxoglutarate and glutamate were found in P. aeruginosa. As 2-oxoglutarate, an intermediate of TCA cycle, serves as a substrate for glutamate synthesis under the catalysis of glutamate synthase. Decreases of the two TCA-related compounds further confirmed the disturbed TCA cycle. As the most important metabolic pathway producing energy for biological organisms, any disturbances in TCA cycle would contribute to the disturbances of energy metabolism, leading to energy deficit and dysfunction in bacteria which would finally influence their pathogenic virulence. Since the TCA cycle was inhibited, other means such as anaerobic respiration had to be accelerated for compensation, which could be concluded by the increased level of acetate according to previous report by Ringo et al.26: anaerobic growth leaded to the accumulation of acetate. It has been reported that anaerobiosis could induce the loss of cytotoxicity in P. aeruginosa, such effect could be attributed to the restrained secretion of one virulence factor namely elastase which is regulated by QS27. Therefore, in summary, the QSI resveratrol may inhibit the pathogenicity of P. 18


Page 18 of 24

Page 19 of 24

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


aeruginosa by disturbing the TCA cycle so that anaerobic respiration would inactivate the QS of P. aeruginosa and this dysfunction could attenuate the pathogenicity of P. aeruginosa. As QS system is utilized by most Gram-negative bacteria for their group behavior coordination and various bacterial functions regulation3, disturbing QS system may inevitably bring influences on normal metabolic system of bacteria. The influence on oxidative stress as well as protein synthesis and energy metabolism in P. aeruginosa PAO1 speculated that QS system was damaged, illustrating a valid QSI activity of resveratrol. CONCLUSION In this study, 1H-NMR-based metabolomics approach was applied for the first time to assess the effects of QSI resveratrol and its induced metabolic alterations in P. aeruginosa PAO1. Multivariate statistical analysis highlighted altered metabolites that contributed to oxidative stress, disturbances of protein synthesis and energy metabolism, thus elucidated a valid QSI activity of resveratrol. By characterizing the global metabolic status dynamically and holistically, metabolomics techniques provided a potent and feasible tool to probe the underlying mechanism of QSI comprehensively. ASSOCIATED CONTENT Supporting Information Figure S-1. Overlay of the HSQC (A) and TOCSY (B) contour plots with assignment examples of some metabolites. 19



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 S-2. Receiver operating characteristic (ROC) curves of classifier performance of OSC-PLS-DA models on 1H-NMR data of resveratrol treated group and control group. Figure S-3. OSC-PLS-DA model permutation testing results (C) and parameters (A and B) for P. aeruginosa PAO1 extracts from resveratrol treated group and control group. ACKNOWLEDGMENTS This work was supported by the grants from the National High Technology Research and Development Program of China (863 Program) (2014AA022208), the National Natural Science Foundation of China (31170131), Six talent peaks project in Jiangsu Province and the Fundamental Research Funds for the Central Universities (30916011307). REFERENCES (1) Schauder, S.; Bassler, B. L. The languages of bacteria. Gene Dev. 2001, 15, 1468-80. (2) Fuqua, C.; Greenberg, E. P. Listening in on bacteria: acyl-homoserine lactone signalling. Nat Rev Mol Cell Biol. 2002, 3, 685-95. (3) Whitehead, N. A.; Barnard, A. M. L.; Slater, H.; Simpson, N. J. L.; Salmond, G. P. C. Quorum-sensing in Gram-negative bacteria. Fems Microbiol Rev. 2001, 25, 365-404. (4) Rutherford, S. T.; Bassler, B. L. Bacterial Quorum Sensing: Its Role in Virulence and Possibilities for Its Control. CSH Perspect Med. 2012, 2, 705-709. 20


Page 20 of 24

Page 21 of 24

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


(5) Bjarnsholt, T.; Givskov, M. The role of quorum sensing in the pathogenicity of the cunning aggressor Pseudomonas aeruginosa. Anal Bioanal Chem. 2007, 387, 409-14. (6) Jimenez, P. N.; Koch, G.; Thompson, J. A.; Xavier, K. B.; Cool, R. H.; Quax, W. J. The multiple signaling systems regulating virulence in Pseudomonas aeruginosa. Microbiol Mol Biol Rev. 2012, 76, 46-65. (7) Parsek, M. R.; Greenberg, E. P. Sociomicrobiology: the connections between quorum sensing and biofilms. Trends Microbiol. 2005, 13, 27-33. (8) Smith, R. P. aeruginosa quorum-sensing systems and virulence. Curr Opin Microbiol. 2003, 6, 56-60. (9) Rasmussen, T. B.; Givskov, M. Quorum-sensing inhibitors as anti-pathogenic drugs. Int J Med Microbiol. 2006, 296, 149-161. (10) Kalia, V. C. Quorum sensing inhibitors: an overview. Biotechnol Adv. 2013, 31, 224-45. (11) Hentzer, M.; Wu, H.; Andersen, J. B.; Riedel, K.; Rasmussen, T. B.; Bagge, N.; Kumar, N.; Schembri, M. A.; Song, Z.; Kristoffersen, P. Attenuation of Pseudomonas aeruginosa virulence by quorum sensing inhibitors. Embo J. 2003, 22, 3803–3815. (12) Jakobsen, T. H.; van Gennip, M.; Phipps, R. K.; Shanmugham, M. S.; Christensen, L. D.; Alhede, M.; Skindersoe, M. E.; Rasmussen, T. B.; Friedrich, K.; Uthe, F.; Jensen, P. O.; Moser, C.; Nielsen, K. F.; Eberl, L.; Larsen, T. O.; Tanner, D.; Hoiby, N.; Bjarnsholt, T.; Givskov, M. Ajoene, a sulfur-rich molecule from garlic, inhibits genes controlled by quorum sensing. Antimicrob Agents Ch. 2012, 56, 2314-25. 21



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

(13) Yang, Y. X.; Xu, Z. H.; Zhang, Y. Q.; Tian, J.; Weng, L. X.; Wang, L. H. A new quorum-sensing inhibitor attenuates virulence and decreases antibiotic resistance in Pseudomonas aeruginosa. J Microbiol. 2012, 50, 987-93. (14) Liu, Y.; Chen, T.; Li, M.-H.; Xu, H.-D.; Jia, A.-Q.; Zhang, J.-F.; Wang, J.-S. 1H NMR based metabolomics approach to study the toxic effects of dichlorvos on goldfish (Carassius auratus). Chemosphere. 2015, 138, 537-545. (15) Chen, F.; Zhang, J.; Song, X.; Yang, J.; Li, H.; Tang, H.; Liao, Y. C. Combined metabonomic and quantitative real-time PCR analyses reveal systems metabolic changes of Fusarium graminearum induced by Tri5 gene deletion. J Proteome Res. 2011, 10, 2273-85. (16) Ye, Y.; Zhang, L.; Hao, F.; Zhang, J.; Wang, Y.; Tang, H. Global Metabolomic Responses of Escherichia coli to Heat Stress. J Proteome Res. 2012, 11, 2559-2566. (17) Frémont L. Biological effects of resveratrol. Life sciences. 2000, 66, 663-673. (18) Sheng, J. Y.; Chen, T. T.; Tan, X. J.; Chen, T.; Jia, A. Q. The quorum-sensing inhibiting effects of stilbenoids and their potential structure-activity relationship. Bioorg Med Chem Lett. 2015, 25, 5217-20. (19) Li, M.; Wang, J.; Lu, Z.; Wei, D.; Yang, M.; Kong, L. NMR-based metabolomics approach to study the toxicity of lambda-cyhalothrin to goldfish (Carassius auratus). Aquat Toxicol. 2014, 146, 82-92. (20) Li, P.; Liao, S.; Wang, J.; Xu, D.; Zhang, Q.; Yang, M.; Kong, L. NMR metabolic profiling of lipopolysaccharide-induced mice sepsis and the treatment effects of berberine. RSC Advances. 2016, 6, 47474-47485. 22


Page 22 of 24

Page 23 of 24

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


(21) Nicas, T. I.; Iglewski, B. H. The contribution of exoproducts to virulence of Pseudomonas aeruginosa. Can J Microbiol. 1985, 31, 387-92. (22) Delden, C. V.; Iglewski, B. H. Cell-to-cell signaling and Pseudomonas aeruginosa infections. Emerg Infect Dis. 1998, 4, 551-560. (23) Tkachenko, A. G.; Nesterova, L. Y.; Pshenichnov, M. P. Role of Putrescine in the Regulation of the Expression of the Oxidative Stress Defense Genes of Escherichia coli. Microbiol. 2001, 70, 133-137. (24) Schmidley, J. W. Free radicals in central nervous system ischemia. Stroke. 1990, 21, 1086-90. (25) Juhas, M.; Wiehlmann, L.; Huber, B.; Jordan, D.; Lauber, J.; Salunkhe, P.; Limpert, A. S.; von Gotz, F.; Steinmetz, I.; Eberl, L.; Tummler, B. Global regulation of quorum sensing and virulence by VqsR in Pseudomonas aeruginosa. Microbiol. 2004, 150, 831-41. (26) Ringø, E.; Stenberg, E.; Strøm, A. R. Amino acid and lactate catabolism in trimethylamine oxide respiration of Alteromonas putrefaciens NCMB 1735. Appl Environ Microb. 1984, 47, 1084-9. (27) Lee, K. M.; Yoon, M. Y.; Park, Y.; Lee, J. H.; Yoon, S. S. Anaerobiosis-induced loss of cytotoxicity is due to inactivation of quorum sensing in Pseudomonas aeruginosa. Infect Immun. 2011, 79, 2792-800.

23



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)

24


Page 24 of 24

1H NMR-Based Global Metabolic Studies of Pseudomonas

Recommend Documents