Fructose Restores Susceptibility of Multidrug-Resistant Edwardsiella

View Sections. ACS2GO © 2018. ← → → ←. loading. To add this web app to the home screen open the browser option menu and tap on Add to hom...
3 downloads 0 Views 2MB Size
Subscriber access provided by Library, Special Collections and Museums, University of Aberdeen

Article

Fructose restores susceptibility of multidrugresistant Edwardsiella tarda to kanamycin Yubin Su, Bo Peng , Yi Han, Hui Li, and Xuanxian Peng J. Proteome Res., Just Accepted Manuscript • DOI: 10.1021/pr501285f • Publication Date (Web): 12 Feb 2015 Downloaded from http://pubs.acs.org on February 14, 2015

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 36

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

Fructose restores susceptibility of multidrug-resistant Edwardsiella tarda to kanamycin

Yu-bin Su§ a, Bo Peng§ b, Yi Han, Hui Li a, Xuan-xian Peng* a §

§

§

The first two authors contributed equally

a

Center for Proteomics and Metabolomics, State Key Laboratory of Biocontrol,

School of Life Sciences, MOE Key Lab Aquat Food Safety, School of Life Sciences, Sun Yat-sen University, Guangzhou 510275, People’s Republic of China b

Molecular Foundry, Lawrence Berkeley National Laboratory, Berkeley, California

94720-8197, USA. ________________________________________________ Running title: Kanamycin plus fructose kill E. tarda

*Corresponding author: Dr. Xuanxian Peng, State Key Laboratory of Biocontrol, School of Life Sciences, Sun Yat-sen University, University City, Guangzhou 510006, People’s Republic of China. Tel: +86-13580448832; Fax: +86-20-8403-6215; E-mail: [email protected]; [email protected]

1

ACS Paragon Plus Environment

Journal of Proteome Research

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 2 of 36

Abstract Edwardsiella tarda, the causative agent of Edwardsiellosis, imposes medical challenges both in the clinic and aquaculture. The emergence of multidrug resistant strains makes antibiotic treatment impractical. The identification of molecules, which facilitate or promote antibiotic efficacy, are in high demand. In the present study, we aimed to identify small molecules whose abundance is correlated with kanamycin resistance in E. tarda by gas chromatography-mass spectrometry. We found that the abundance of fructose was greatly suppressed in kanamycin-resistant strains. The incubation of kanamycin-resistant bacteria with exogenous fructose sensitized the bacteria to kanamycin. Moreover, the fructose also functioned in bacteria persisters and biofilm. The synergistic effects of fructose and kanamycin were validated in a mouse model. Furthermore, the mechanism relies on fructose in activating TCA cycle to produce NADH, which generates proton motive force to increase the uptake of the antibiotics. Therefore, we present a novel approach in fighting against multidrug resistant bacteria through exploration of antibiotic-suppressed molecules. Keywords: multidrug resistance; E. tarda; kanamycin; fructose; TCA cycle

2

ACS Paragon Plus Environment

.

Page 3 of 36

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. Introduction Edwardsiella tarda is a natural inhabitant of fresh and marine water and is typically found in the normal gut flora of fish and humans. It causes infectious diseases, known as Edwardsiellosis, in fish, birds, reptiles and mammals.1-3 Infection of E. tarda in humans causes gastroenteritis, diarrhea and extraintestinal diseases.4-7 And the infection in fish causes the death of fish both in freshwater and marine aquaculture, which resulted in extensive economic losses throughout the world.2,

3

Human

Edwardsiellosis may originated from fish.8 A variety of extraintestinal infections and many of the pathological changes induced by E. tarda in humans are consistently observed in infected fish.6 Edwardsiellosis is routinely treated with antibiotics due to the absence of applicable vaccines.9 However, the overuse of antibiotics leads to the incidence of antibiotic resistance both in human and fish.10 Accumulating evidence has shown that E. tarda strains harbor plasmids with multi-drug resistant genes against kanamycin, streptomycin, tetracycline, sulfonamide and chloramphenicol.11-13 This imposes the challenge of limited therapy to Edwardsiellosis. Thus, discovery of new antibiotics or improvement of the efficacy of the available antibiotic arsenal is in demand in nowadays.

Recently, the relationship between bacterial metabolic regulation and antibiotic efficacy has been proposed in several pathogens. Metabolites like indole, hydrogen sulfide, nitric oxide and gaseous ammonia modify the cellular metabolic environment and alter antibiotic susceptibility in Escherichia coli.14-17 Crc and CbrAB, two global 3

ACS Paragon Plus Environment

Journal of Proteome Research

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

metabolic regulators in Pseudomonas aeruginosa, modulate susceptibility of bacteria to a variety of clinically important antibiotics.18, 19 Down-regulation of NQR and Nar, two unique respiratory enzyme complexes, is essential for Vibro alginolyticus and E. coli in resistance to antibiotics, respectively.20, 21 On the contrary, exogenous addition of carbon sources may reduce aminoglycoside tolerance in E. coli and Staphylococcus aureus biofilms and persisters. The metabolite-enabled eradication of bacterial persisters is species-dependent. Mannitol, glucose and pyruvate showed strong potentiation against E. coli, but little effect on the potentiation against S. aureus.22 Whether such metabolite-dependent potentiation correlated with the infectious process remains elusive, where E. coli is an extracellular pathogen while S. aureus is an intracellular pathogen.23 Nonetheless, proton motive force (PMF) is required for the metabolite to take effects. However, the generation of metabolite-related PMF is not yet identified. 22

Here we present functional characterization of metabolites, whose expression is correlated with kanamycin resistance in E. tarda. We found the expression of fructose is highly suppressed in passage-generated kanamycin-resistant and clinical isolated multidrug-resistant E. tarda strains. We further demonstrated the exogenous fructose potentiates kanamycin to eliminate these E. tarda including biofilms and persisters. The mechanism of fructose-enabled bacteria killing is dependent on the promotion of TCA cycle, which in turn increases NADH, and thereby PMF generation. The increased PMF ultimately facilitates kanamycin uptake. The in vitro data were 4

ACS Paragon Plus Environment

Page 4 of 36

Page 5 of 36

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

reproducibly obtained in a mouse model. These findings showed that fructose restores sensitivity of multidrug–resistant E. tarda to kanamycin.

2. Materials and methods 2.1 Bacterial strains, cultivation and minimum inhibitory concentration. E. tarda strains (LTB4-S, EIB202, ATCC15947, WY28, WY37) in this study were obtained from professor Yuanxin Zhang, East China University of Science and Technology and professor Xiaohua Zhang, Ocean University of China. Out of them, LTB4-S, EIB202 and ATCC15947 were three reported E. trada strains

3, 24, 25

Complete genome sequence of EIB202 was published in 2009.13 Kanamycin-resistant E.

tarda

LTB4

strain

(LTB4-R)

was

derived

from

a

wide

type

of

kanamycin-susceptible E. tarda LTB4 strain (LTB4-S) as described previously.26 In brief, LTB4-S cells were grown in Tryptic Soy Broth (TSB) media containing kanamycin at 30 °C for twenty generations. MRSA-5 and Klebsiella pneumoniae 0210 were obtained from The Third Hospital, Sun Yat-sen University and Zhongshan Hospital, Xiamen University, respectively. V. parahaemolyticus was from the collection of our Lab. Strains were identified by sequencing 16S RNA. All E. tarda strains except ATCC15947 and V. parahaemolyticus were grown at 30 °C. ATCC15947, MRSA-5 and Klebsiella pneumoniae 0210 strains were cultured at 37 °C for 24 h in 50 mL LB broth in 250 mL flasks. Minimum inhibitory concentration (MIC) was determined according to the NCCLS procedure using the microdilution method as described previously.27 5

ACS Paragon Plus Environment

Journal of Proteome Research

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

2.2 Metabolomic profiling. Sample preparation procedure was as previously described.28, 29 Briefly, equivalent cells were quenched with 60% (v/v) cold of methanol (Sigma) and were collected by centrifugation at 8,000 rpm at 4 0C for 5 min. The metabolites were extracted by 1 mL cold of methanol. To normalize variations across samples, an internal standard (0.1 mg/mL ribitol) (Sigma) was used. The cells were lysed by sonication at power of 10W for 3 min and then were centrifuged at 12,000 rpm at 4 0C for 10 min. 500 µL supernatant was transferred into 1.5 mL microtube and was dried by vacuum centrifugation device (LABCONCO). Finally, the samples were performed on a GC/MS system. Each sample had five biological repeats with two technical replicas.

GC-MS analysis was carried out on the two-stage technique as described previously.30 Before analysis, samples were derivatized. First, carbonyl functions were protected by methoximation through a 90 min 37 0C reaction with 40 µL of 20 mg/mL methoxyamine hydrochloride (Sigma-Aldrich) in pyridine. Then, acidic protons were exchanged against trimethylsilyl group by 37

0

C reaction with 80 µL

N-methyl-N-trimethylsilyltrifluoroacetamide (MSTFA, Sigma-Aldrich) for 30 min. The derivatized sample of 1 µL aliquot was injected into a DBS (Dodecyl Benzene Sulfonic Acid) column (30 m length × 250 µm i.d. × 0.25 µm thickness, Thermo Fisher Scientifc) using splitless model. The temperature programmed started at 85 0C for 5 min and then increased to a final temperature of 330 0C and held constant for 5 min, followed by at a rate of 15 0C min-1. Electron impact ionization (EI) mode was 6

ACS Paragon Plus Environment

Page 6 of 36

Page 7 of 36

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

selected and ionization was energy of 70 eV. Helium was used as the carrier gas with the flow rate of 1 mL min-1. The range of mass full scan mode was 50-600 m/z.

2.3 Data processing and statistical analysis. The mass fragmentation spectrum was analyzed from the TIC with XCalibur software (Thermo Fisher Scientific, version 2.1) to identify the compounds by matching the data with the National Institute of Standards and Technology (NIST) library and NIST MS search 2.0 program. Rank and score were used to evaluate the hits. The first hit rank with a score larger than a predefined threshold (600) was considered. Peak areas of all identified metabolites were normalized by ribitol as internal standard. All metabolomics profile data were analyzed by independent component analysis (ICA) (http://metagenealyse.mpimp-golm.mpg.de/).25

2.4 Antibiotic bactericidal assays. Bacterial cells were collected by centrifugation at 8,000 rpm for 5 min. The samples were then washed with sterile saline three times and suspended in M9 minimal media containing 10 mM acetate, 1 mM MgSO4 and 100 µM CaCl2, diluted to OD600 of 0.2. Fructose or/and antibiotic were added, and incubated at 30 0C or 37 0C (dependent on bacterial strains tested) for 6 h, 200 rpm. In all experiments using CCCP, samples were pretreated for 5min with the proton ionophor before antibiotic addition. To determine bacterial counts, 100 µL of cultures were removed, and then serially diluted.

7

ACS Paragon Plus Environment

Journal of Proteome Research

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

Aliquot 10 µL of each dilution was plated in TSB agar plates and incubated at 30 0C or 37 0C for 18 – 22 h. The colonies were counted and CFU/mL was calculated.

2.5 Anaerobic culture. E. tarda EIB202 cells were grown in 100 mL TSB at 30 0C for 24 h. The cells were collected, washed and then re-suspended in M9 minimal medium to an OD600 of 0.2. Fructose or/and antibiotics were added and incubated in an anaerobic jar with an Anoxomat Mark II system (MART Microbiology, Lichtenvoorde, The Netherlands) at 30 0C for 6 h. The cultures were spotted onto TSB agar plates to determine CFU/mL as described above.

2.6 Measurement of kanamycin concentration We used the kanamycin ELISA rapid diagnostic kit (Beijing Clover Technology Group Inc., Beijing, China) to assess kanamycin concentration in bacterial cells. EIB202 cells were re-suspended to OD 0.2 and incubated with fructose or/and kanamycin at 30 0C for 6 h. After centrifugation and washing three times with sterile saline, the cells were re-suspended with the same buffer and adjusted to OD600 to 1.0. Aliquot 1 mL was sonicated for 3 min. The resulting supernatant was collected for detection of kanamycin. The luminescence intensity was measured in a path length (10 mm) quartz cell with excitation and emission spectra, which were recorded as 287 and 450 nm, respectively.

8

ACS Paragon Plus Environment

Page 8 of 36

Page 9 of 36

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

2.7 Measurement of membrane potential. BacLight bacterial membrane potential kit (Invitrogen) was used for measurement of bacterial membrane potential. Membrane potential induced by fructose was diluted to 106 CFU/mL and stained with 10 µl of 3 mM DiOC2 (3), followed by incubation for 30 min. Samples were analyzed using a FACSCalibur flow cytometer (Becton Dickinson, San Jose, CA, USA). The green/ red fluorescence was detected through a 488- to 530/610-nm bandwidth band-pass filter, respectively. The membrane potential was determined and normalized as the intensity ratio of the red fluorescence (a membrane potential-dependent signal) and the green fluorescence (a membrane potential-independent signal). Relative PMF was determined in test samples compared to control samples without fructose.

2.8 Measurement of NADH. NADH was measured using the EnzyChrom™ NAD/NADH Assay Kit (BioAssay Systems, USA). In brief, EIB202 cells were re-suspended to OD 0.2 and incubated with fructose or/and kanamycin at 30 0C for 6 h. Cells were harvested by centrifugation at 8000 rpm for 5 min at 4 0C. Bacterial pellets were washed with sterile saline three times, re-suspended in NADH extraction buffer and incubated in 60 0C water for 5min. Assay buffer and NAD exaction buffer were added. After vortex briefly and centrifugation at 14,000 rpm for 5 min, the resulting supernatant was used for NADH measurement by the EnzyChrom™ NAD/NADH Assay Kit.

9

ACS Paragon Plus Environment

Journal of Proteome Research

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

2.9 Western blot. Bacterial proteins were separated using SDS/PAGE gels and transferred to nitrocellulose (NC) membranes using constant voltage of 70 V for 1 h in transfer buffer (48 mM Tris, 39 mM glycine, and 20% methanol) at 4 °C. The membranes were blocked overnight with 5% non-fat milk in Tris-buffered saline buffer containing 0.05% Tween-20 (TTBS) at 4 °C. After rinsing three times for 10 min with TTBS buffer, rabbit antisera to SucC or GltA were used as the primary antibodies and horseradish peroxidase (HRP)-conjugated goat anti-rabbit antibody was used as the secondary antibody (Guangzhou Chengxue Biotech. Corp. China). The membranes were washed and developed with dimethylaminoazobenzene (DAB) substrate system until maximum color appearance.

2.10 Measurement of enzyme activity. Activity of the three key enzymes, citrate synthase, isocitrate dehydrogenase and α-oxoglutarate dehydrogenase, of the TCA cycle was measured using commercial assay kits (Genmed Scientifics Inc, USA). In brief, EIB202 cells were grown at 30 0C for 24 h. Cells were diluted to an OD600 of 0.2 and incubated in 30 0C with exogenous fructose for 6 h. Cells were collected by centrifugation at 8, 000 rpm for 5 min at 4 0C. The cell pellets were washed with sterile saline (0.85% NaCl) three times, resuspended in lysate (from the assay kits) and disrupted by intermittent sonic oscillation for a total of 9 min intervals of 5 s on ice. Following centrifugation at 1, 3000 rpm for 15 min at 4 0C, supernatant was transferred to new tube and used 10

ACS Paragon Plus Environment

Page 10 of 36

Page 11 of 36

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

Bradford Assay to detect protein concentrations.

2.11 Biofilm and mouse infection assay. Biofilm-formation culture was determined as described previously.22,

32

In brief,

stationary phase bacteria were diluted 1:100 into fresh TSB and then 6 mm PE50 catheters (0.58mm×0.96mm) were added. The catheters were incubated at 30 0C or 37 0

C (dependent on bacterial strains tested) for 24 h. The medium was changed per day

with total of 3 days. The resulting catheters were washed with sterile saline to remove loosely attached cells. In vitro experiment, the catheters were dislodged by sonication in water bath for 30 min at 40 kHz to release bacterial cells for counting. In vivo studies, 4-week-old female Balb/c (body weight, 20-25g) received surgical implantation in the biofilm-coated catheters. At 48 h after surgery, mice were divided into four groups, 8 mice each group. Intravenous treatment with saline (saline control), 250 mg kg-1 fructose (fructose control), 3 mg kg-1 kanamycin (kanamycin control), 250 mg kg-1 fructose plus kg-1 kanamycin 3 mg (test group) twice-daily for 3 days. After the last treatment for 24 h, the catheters and mouse kidneys were removed to determine bacterial load using plate counting. Female BALB/c mice were obtained from the animal center of Sun Yat-sen University, Guangzhou, China and animal protocols were approved by the Committee for the Use and Care of Laboratory Animals, Sun Yat-sen University, China.

3. Results 11

ACS Paragon Plus Environment

Journal of Proteome Research

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

3.1 Decreased fructose represents a characteristic feature in kanamycin-resistant and multidrug-resistant E. tarda. We put forward the hypothesis that abundance of fructose might be changed in aminoglycoside antibiotic-resistant intracellular pathogen E. tarda; this change affected the bacterial metabolic environment and subsequently modulated antibiotic susceptibility. To demonstrate this idea, we used three kanamycin-resistant E. tarda strains including a lab-generated kanamycin-resistant strain, LTB4-R, from kanamycin-sensitive wild-type bacterium (LTB4-S) through sequential propagations of LTB4-S in medium with kanamycin, a human-derived strain ATCC15947 and a fish-derived strain EIB202, where EIB202 is a multidrug-resistant strain including kanamycin resistance, and ATCC15947 has doubled MIC to kanamycin to LTB4-S (Fig. 1A). Then gas chromatography/mass spectrometry (GC/MS) was used to quantify fructose levels in these strains. Significantly decreased abundance of fructose was detected in LTB4-R than that in LTB4-S (Fig. 1B). Similarly, lower fructose was detected in EIB202 and ATCC15947 than in LTB4-S (Fig. 1B). These results indicate that fructose may be related to kanamycin resistance in E. tarda.

3.2 Fructose restores bacterial susceptibility to kanamycin. Viabilities of LTB4-R were reduced by 2.5 – 155.9 fold with various doses of exogenous fructose plus 500 µg/mL kanamycin as indicated in Fig. 2A and by 1.7 3013.6 fold with increasing doses of kanamycin plus 2.5 mM fructose (Fig. 2B). The killing was also time-dependent with the largest effect at 6-8 h (Fig. 2C). Additionally, 12

ACS Paragon Plus Environment

Page 12 of 36

Page 13 of 36

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

five other wild strains ATCC15947, EIB202, LTB4-S, WY28 and WY37 were used for

this

investigation.

Similarly,

the

killing

was

fructose-dose-

or/and

incubation-period- dependent (Figs. 2D-G; Supplementary Figs. 1 and 2). Higher bactericide folds were found in ATCC15947 and EIB202 than in LTB4-R in the presence of 2.5 mM fructose even if lower kanamycin was used in ATCC15947 and EIB2020 than LTB4-R, which is consistent with the result that lower MIC of kanamycin was detected in ATCC15947 or EIB2020 than LTB4-R (Fig. 1A). Similar results were obtained when replacement of kanamycin with gentamicin was used (Fig. 2H). This potentiation by fructose was demonstrated in other pathogens MRSA-5, K. peneumoniae and V. parahaemolyticus (Supplementary Fig. 3). These results indicate that fructose may effectively potentiate kanamycin to eliminate multidrug-resistant E. tarda and other pathogens.

3.3 Exogenous fructose promotes the TCA cycle. We reasoned that exogenous fructose played a role through pathway activation and metabolic regulation, which promoted generation of NADH and in turn PMF. To demonstrate this, GC/MS was used to analyze metabolomes of LTB4-R cells in the presence or absence of fructose. Five biological and two technical replicates yielded 20 data. Total 63 metabolites were detected in each sample. These metabolites were ranked and visualized in Supplementary Fig. 4A. Z value based on LTB4-S was calculated for comparing purpose (Supplementary Fig. 4B).

Thirty-six (57.1%) out

of the 63 metabolites showed significant difference (P