Unveiling the Crucial Roles of O2•– and ATP in Hepatic Ischemia–Reperfusion Injury Using Dual-Color/Reversible Fluorescence Imaging

Hepatic ischemia–reperfusion injury (HIRI) is mainly responsible for morbidity or death due to graft rejection after liver transplantation. During HIRI, superoxide anion (O2•–) and adenosine-5′-triphosphate (ATP) have been identified as pivotal biomarkers associated with oxidative stress and energy metabolism, respectively. However, how the temporal and spatial fluctuations of O2•– and ATP coordinate changes in HIRI and particularly how they synergistically regulate each other in the pathological mechanism of HIRI remains unclear. Herein, we rationally designed and successfully synthesized a dual-color and dual-reversible molecular fluorescent probe (UDP) for dynamic and simultaneous visualization of O2•– and ATP in real-time, and uncovered their interrelationship and synergy in HIRI. UDP featured excellent sensitivity, selectivity, and reversibility in response to O2•– and ATP, which rendered UDP suitable for detecting O2•– and ATP and generating independent responses in the blue and red fluorescence channels without spectral crosstalk. Notably, in situ imaging with UDP revealed for the first time synchronous O2•– bursts and ATP depletion in hepatocytes and mouse livers during the process of HIRI. Surprisingly, a slight increase in ATP was observed during reperfusion. More importantly, intracellular O2•–—succinate dehydrogenase (SDH)—mitochondrial (Mito) reduced nicotinamide adenine dinucleotide (NADH)—Mito ATP—intracellular ATP cascade signaling pathway in the HIRI process was unveiled which illustrated the correlation between O2•– and ATP for the first time. This research confirms the potential of UDP for the dynamic monitoring of HIRI and provides a clear illustration of HIRI pathogenesis.

Absorption spectra were recorded on a UV-Visible spectrophotometer (Evolution 220, Thermo Scientific).
Fluorescence spectra were obtained with a Hitachi F-4700 fluorescence spectrophotometer.CCK-8 assay was performed using a Triturus microplate reader.Confocal imaging was performed on Leica SP8 high-resolution fluorescence microscope.The mass spectra were obtained using the Bruker Maxis ultra-high-resolution-TOF MS system.Mice liver slices of 100 μm thickness were obtained with cryostat (Leica CM1950).Proteomic analysis was performed by Q Exactive (Thermo Fisher) through LC-MS/MS.Hematoxylin-eosin (H&E) staining images in Figure 5 were obtained using an optical microscope (Leica, DM2500).H&E staining images in Figure S17 were obtained using an optical microscope (Nikon, Eclipse Ci-L). 1 H NMR spectra were obtained at 400 MHz using Bruker NMR spectrometers, and 13 C NMR spectra were recorded at 100 MHz.HPLC analysis was carried out on a Shimadzu LC-16 system equipped with SPD-16 UV−vis detector.

ONOO −
0.6 M NaNO2, 0.6 M HC1 and 0.7 M H2O2 were added simultaneously to a 3 M NaOH solution at 0 ºC.The concentration of ONOO − was determined using extinction coefficient of 1670 M -1 cm -1 at 302 nm in 0.1 M NaOH (aq.).
• OH • OH (hydroxyl radical) was generated by the Fenton reaction of FeCl2 with H2O2 (1:6) in deionized water. 1 NO NO (Nitric oxide) was obtained from a stock solution prepared by sodium nitroprusside (SNP). 2 Briefly, NO was prepared in the following manner: 4 mL of 10 mM SNP was dissolved in PBS buffer solutions (pH = 7.4) under light irradiation for 30 min.Then, the concentrations of NO release were determined by the Griess method reported previously (NO concentration assay kit, Beyotime). 3,450 μL of Griess Reagent I and 50 μL of Griess Reagent II were added to 50 μL of SNP solutions, and then the mixture absorbance was measured at 540 nm.The concentrations of NO were calculated the standard curve of sodium nitrite standards.

O2
[7] Fluorescence spectra of UDP toward O2 •− in the presence of ATP A solution of ATP (22 mM) was added into UDP (25 μM), and the fluorescence spectrum was recorded after 25 min.Then various concentrations of O2 •− (0-65 μM) were added to the solution, and each line of the fluorescence spectra in O2 •− channel was recorded after 5 min.λex = 380 nm.

Fluorescence spectra of UDP toward ATP in the presence of O2 •−
A solution of O2 •− (65 μM) was added into UDP (25 μM), and the fluorescence spectrum was recorded after 5 min.Then various concentrations of ATP (0-22 mM) were added to the solution, and each line of the fluorescence spectra in ATP channel was recorded after 25 min.λex = 520 nm.

LOD calculation
The limit of detection (LOD) was calculated using the well-established method (LOD = 3σ/K), where K is slope of the calibration curve, σ represents is the standard deviation of the blank sample (11 times) of F588 (for ATP) or F470 (for O2 •− ) for UDP without addition of ATP or O2 •− . 8(Note F588 and F470 refer to the Fluorescence (F) emission wavelength peaks at 588 and 470 nm).

HPLC Analysis
For HPLC analysis, deionized water was used as eluent A and methanol as eluent B. HPLC conditions: 14% of A, 86% of B. The injection volume was 10 μL.The parameters of the HPLC-MS analytical column used were C18-WR, 5 μm, 4.6 mm × 150 mm (GL Sciences).The purity of UDP was calculated to be 98% as determined by HPLC analysis.Purity data was calculated based on the integration in the HPLC trace at 254 nm.The flow rate is 0.7 mL/min.

Hepatic ischemia−reperfusion injury models in cells
Hepatic ischemia−reperfusion injury (HIRI) cell models were established by oxygen-glucose-serum deprivation/reperfusion.For the phase of ischemia, HL-7702 cells were cultured in DMEM (without glucose and serum) and deoxygenated sodium dithionite (0.5 mM) for 20 min or 40 min ischemia.For the phase of subsequent reperfusion, these cells were incubated with high glucose and serum DMEM (standard DMEM) in a 5 % CO2 and 95 % O2 atmosphere for 20 min or 40 min reperfusion after 40 min ischemia.

Fluorescence imaging of O2 •− and ATP under 2-ME stimulation
HL-7702 cells were divided into three groups.The control group cells were stained with UDP (40 μM) for 20 min.The 2-ME stimulated group cells were incubated with 2-ME (3.0 μg/mL) for 1 h, then stained with UDP (40 μM) for 20 min.The Tiron + 2-ME stimulated group cells were pretreated with Tiron (10 μM) for 1 h, followed by adding 2-ME (3.0 μg/mL) for 1 h and then stained with UDP (40 μM) for 20 min.The cell culture medium of each group was removed, and all cells were washed with 1.0 mL of PBS three times before fluorescence imaging.One-photon confocal photographs were taken using a Leica SP8 high-resolution fluorescence microscope equipped with the Leica Application Suite X software package.

Fluorescence imaging of O2 •− and ATP under oligomycin A stimulation
The control group cells were stained with UDP (40 μM) for 20 min.The oligomycin A group cells were incubated with oligomycin A (50 μM) for 1 h, then stained with UDP (40 μM) for 20 min.The oligomycin A + ATP stimulated group cells were incubated with oligomycin A (50 μM) for 1 h prior to treating with ATP (10 mM) for 1 h.They were then stained with UDP (40 μM) for 20 min.The cell culture medium of each group was removed, and all cells were washed with 1.0 mL of PBS three times before fluorescence imaging.Onephoton confocal photographs were taken imaged by Leica SP8 high-resolution fluorescence microscope and Leica Application Suite X software.

Fluorescence imaging of O2 •− and ATP in hepatocytes during HIRI
In order to perform fluorescence imaging of O2 •− and ATP fluctuation during the process of HIRI, HL-7702 cells were divided into five groups.Control group hepatocytes were cultured with high glucose and serum DMEM in a 5 % CO2 and 95 % O2 atmosphere.The 20 min of ischemia group hepatocytes were cultured in DMEM (without glucose and serum) and deoxygenated sodium dithionite (0.5 mM) for 20 min.The 40 min of ischemia group hepatocytes were cultured in DMEM (without glucose and serum) and deoxygenated sodium dithionite (0.5 mM) for 40 min.For 40 min of ischemia followed by 20 min of reperfusion group, hepatocytes were cultured in DMEM (without glucose and serum) and deoxygenated sodium dithionite (0.5 mM) for 40 min.Afterwards, cells were incubated with high glucose and serum DMEM in a 5 % CO2 and 95 % O2 atmosphere for 20 min.For 40 min of ischemia followed by 40 min of reperfusion group, hepatocytes were cultured in DMEM (without glucose and serum) and deoxygenated sodium dithionite (0.5 mM) for 40 min.
Afterwards, cells were incubated with high glucose and serum DMEM in a 5 % CO2 and 95 % O2 atmosphere for 40 min.All the groups were treated with 40 μM UDP for 20 min before confocal imaging.

Fluorescence imaging of intervention effect of HIRI drug
To investigate injury remediation, HL-7702 cells were pretreated with 0.5 mM or 1 mM NAC for 1 h and then cultured with DMEM (without glucose and serum) and deoxygenated sodium dithionite (0.5 mM) for 40 min, followed by incubation with standard DMEM in a 5 % CO2 and 95 % O2 atmosphere for 40 min.All the groups were treated with 40 μM UDP for 20 min before confocal imaging.

Fluorescence imaging of intracellular ROS during HIRI
HL-7702 cells were divided into five groups.The control group cells were placed under normal culture conditions.The 20 min of ischemia group hepatocytes were cultured in DMEM (without glucose and serum) and deoxygenated sodium dithionite (0.5 mM) for 20 min.The 40 min of ischemia group hepatocytes were cultured in DMEM (without glucose and serum) and deoxygenated sodium dithionite (0.5 mM) for 40 min.
For 40 min of ischemia followed by 20 min of reperfusion group, hepatocytes were cultured in DMEM (without glucose and serum) and deoxygenated sodium dithionite (0.5 mM) for 40 min.Afterwards, cells were incubated with high glucose and serum DMEM in a 5 % CO2 and 95 % O2 atmosphere for 20 min.For 40 min of ischemia followed by 40 min of reperfusion group, hepatocytes were cultured in DMEM (without glucose and serum) and deoxygenated sodium dithionite (0.5 mM) for 40 min.Afterwards, cells were incubated with high glucose and serum DMEM in a 5 % CO2 and 95 % O2 atmosphere for 40 min.All the group cells were co-stained with DCFH-DA (10 μM) and Hoechst 33342 (1 μg/mL) for 20 min before confocal imaging.

Fluorescence imaging of O2 •− and ATP in mice during HIRI
Mice were randomly divided into five group.The mice in the control group were subjected to a laparotomy and the livers were exposed.The mice in 20 min of ischemia group underwent a laparotomy surgery and the portal vein and hepatic artery of the median and the left lateral lobes of the liver were clamped with a microvessel clip for 20 min.The mice in 40 min of ischemia group underwent a laparotomy surgery and the portal vein and hepatic artery of the median and the left lateral lobes of the liver were clamped with a microvessel clip for 40 min.For 40 min of ischemia followed by 20 min of reperfusion group, the portal vein and hepatic artery of the median and the left lateral lobes of the liver were clamped with a microvessel clip for 40 min.Subsequently, the vascular clamp was opened for 20 min reperfusion.For 40 min of ischemia followed by 40 min of reperfusion group, the portal vein and hepatic artery of the median and the left lateral lobes of the liver were clamped with a microvessel clip for 40 min.Subsequently, the vascular clamp was opened for 40 min reperfusion.Afterwards, UDP with a dose of 100 μM was intravenously injected into all groups of mice through the tail vein, after 20 min, the mice were sacrificed and the livers of all groups were sectioned using cryostat (Leica CM1950).Finally, the sections were imaged through Leica SP8 high-resolution fluorescence microscope.

In vivo toxicity of UDP
The C57 mice were divided into three groups consisting of control group, 100 μM of UDP-treated group and 1 mM UDP-treated group.The control group were intraperitoneally (i.p.) injected with saline (0.9 % NaCl) solution every day for two weeks.The UDP-treated group were intraperitoneally administrated UDP for two weeks with experimental concentration of 100 μM or 1 mM.The body weights of three groups of mice were recorded every day for two weeks.After intraperitoneal injection with saline or UDP for two weeks, hematoxylin and eosin (H&E) staining of major organ tissues (liver, spleen, lung, heart, and kidney) was conducted to identify the histological changes.The tissues samples were fixed with 4% paraformaldehyde, and dehydrated, embedded, sectioned and stained by hematoxylin and eosin.Finally, the sections were imaged using an optical microscope (Nikon, Eclipse Ci-L).

Reversible fluorescence imaging of O2 •− fluctuations in mice
The C57 mice were divided into five groups.The control group mice were intraperitoneally injected with UDP (100 μM) for 20 min.The 2-ME group of mice were intraperitoneally injected with UDP (100 μM) for 20 min, followed by the injection of 2-ME (15 μg/mL) for 20 min.The GSH group of mice were intraperitoneally injected with UDP (100 μM) for 20 min and 2-ME (15 μg/mL) for the next 20 min, followed by the injection of GSH (5 mM) for 20 min.Another 2-ME group of mice were intraperitoneally injected with UDP (100 μM) for 20 min and 2-ME (15 μg/mL) for next 20 min, followed by the injection of GSH (5 mM) for 20 min and the injection of 2-ME (15 μg/mL) for 20 min.Another GSH group of mice were intraperitoneally injected with UDP (100 μM) for 20 min and 2-ME (15 μg/mL) for next 20 min, followed by the injection of GSH (5 mM) for 20 min, the injection of 2-ME (15 μg/mL) for 20 min and the final injection of GSH (5 mM) for 20 min.
All groups of mice were sacrificed and the livers of all groups were sectioned using cryostat (Leica CM1950).
Finally, the liver sections were imaged through Leica SP8 high-resolution fluorescence microscope.

Reversible fluorescence imaging of ATP fluctuations in mice
The C57 mice were divided into five groups.The control group mice were intraperitoneally injected with UDP (100 μM) for 20 min.The ATP group of mice were intraperitoneally injected with UDP (100 μM) for 20 min, followed by the injection of ATP (30 mM) for 30 min.The apyrase group of mice were intraperitoneally injected with UDP (100 μM) for 20 min and ATP (30 mM) for next 30 min, followed by the injection of apyrase (3 U/N) for 30 min.Another ATP group of mice were intraperitoneally injected with UDP (100 μM) for 20 min and ATP (30 mM) for the next 30 min, followed by the injection of apyrase (3 U/N) for 30 min and the injection of ATP (30 mM) for 30 min.Another apyrase group of mice were intraperitoneally injected with UDP (100 μM) for 20 min and ATP (30 mM) for the next 30 min, followed by the injection of apyrase (3 U/N) for 30 min, the injection of ATP (30 mM) for 30 min and the final injection of apyrase (3 U/N) for 30 min.All groups of mice were sacrificed and the livers of all groups were sectioned using cryostat (Leica CM1950).Finally, the liver sections were imaged through Leica SP8 high-resolution fluorescence microscope.

H&E staining of major organs in control group mice and HIRI group mice
The C57 mice were divided into two groups consisting of control group and HIRI group.Both control group and HIRI group were intraperitoneally (i.p.) injected with 100 μM of UDP.After intraperitoneal injection with UDP, H&E staining of major organ tissues (liver, spleen, lung, heart, and kidney) was conducted to identify the histological changes.The tissues samples were fixed with 4% paraformaldehyde, and dehydrated, embedded, sectioned and stained by hematoxylin and eosin.Finally, the sections were imaged using an optical microscope (Leica, DM2500).

BCA protein assays
The BCA protein assays were carried out by BCA Protein Assay Kit (GenStar).The principle of BCA determination method is that Cu 2+ can be reduced to Cu + by protein under alkaline conditions, and Cu + binds with BCA reagent to form purple complex.By measuring the absorbance of samples at 562 nm and comparing with the standard curve of reference samples, the protein concentration of the sample to be measured can be calculated.The hepatocytes samples under various treatment were collected and placed in ice-water bath for homogenate crushing.20 μL of reference sample and hepatocytes samples were added in microplate, respectively.200 μL of BCA working solution was added in each well and vibrate to mix sufficiently.The microplate was covered and incubated for 30 min at 37 °C.After cooling to room temperature, the absorbance at 562 nm was measured using a Triturus microplate reader.The protein concentrations of samples were determined according to the standard curve of reference sample and the dilution ratio of samples.

ATP contents assays
The ATP contents in hepatocytes per unit protein weight were calculated combined with BCA Protein Assay kit and ATP Contents Assay Kit (Nanjing Jiancheng Bioengineering Institute).The principle of ATP determination method is that creatine kinase catalyzes ATP and creatine to produce creatine phosphate.For ATP measurement, the phosphomolybdic acid colorimetric method was used to detect the generation of creatine phosphate.The hepatocytes samples under various treatment were collected and placed in ice-water bath for homogenate crushing.The cell suspensions were heated in a boiling water bath for 10 min, vortex mixed for 1 min and centrifuged at 3,500 rpm for 10 min.To measure ATP concentrations, 30 µL supernatant, 100 µL substrate I, 200 µL substrate II and 30 µL accelerator were combined and incubated at 37 °C for 30 min.Then, 50 µL precipitant was added, and the mixture was centrifuged at 4,000 rpm for 5 min.A volume of 500 µL chromogenic fluids were combined with 300 µL supernatant and incubated at room temperature for 2 min.Finally, 500 µL termination fluids were added and incubated at room temperature for 5 min.A 200 µL volume of the mixture was added to a 96-well plate, and the absorbance values were measured at 636 nm on a Triturus microplate reader.The ATP concentrations were calculated as follows: CATP= (ASample − AControl) / (AReference − ABlank) × CReference × N / Cpr in which CATP is the concentration of ATP (nmol/mg protein), ASample is the absorbance of the sample, AControl was the absorbance of the control, AReference is the absorbance of the reference, ABlank is the absorbance of the blank, CReference is the concentration of the reference (1000 µmol/L), N is the sample dilution ratio before the determination, Cpr is the concentration of homogenate protein.

Succinate dehydrogenase activity assays
SDH activity assays were performed using a SDH Activity Assay Kit (Nanjing Jiancheng Bioengineering Institute).The SDH activity per unit protein weight (U/mg protein) were calculated combined with BCA Protein Assay kit and SDH Activity Assay Kit.Flavin adenine dinucleotide (FAD) is the auxiliary group of the SDH catalyzed reaction, and FAD is reduced to FADH, coupled with the reduction of 2,6-dichlorophenol indigo (2, 6-DPIP).The reduction rate of 2, 6-DPIP is determined to calculate the activity of SDH.The working solution was placed at 37 °C for 10 min.2.6 mL of working solution were added to 100 µL of the samples and mixed.At the moment, the time point was defined as 0 min.5 seconds later, the absorbance values at 600 nm were measured on a Triturus microplate reader.The absorbance values at 600 nm were measured again after 1 min and the difference between two absorbances was calculated.One unit of SDH activity was defined as a decrease of 0.01 in absorbance per minute at 600 nm in the reaction solutions.The SDH activity was calculated as follows: ASDH= [∆A / (0.01 × T)] / (VSample × Cpr) in which ASDH is the activity of SDH (U/mg protein), ∆A is the difference between absorbance at 5 s and absorbance at 1 min 5 s, T was reaction time (1 min), VSample is the volume of homogenate, Cpr is the concentration of homogenate protein.

Mitochondria extraction from hepatocytes
The mitochondria extraction was performed using a Mitochondria Extraction Kit (CS0201, Bjbalb, China).
Hepatocytes under various treatment were trypsinized and centrifuged for 10 min at 1000 rpm.The cells were resuspended in PBS buffer and centrifuged for 10 min at 1000 rpm.The collected cells were added in 1 mL of reagent A and 10 µL of reagent D. Dounce homogenizer was used to homogenate many times in ice-bath.The homogenate was centrifuged for 5 min at 600 × g at 4 °C and liquid supernatant was collected.Then the liquid supernatant was centrifuged for 10 min at 11000 × g at 4 °C and sediment was collected.The resulted sediment was intact mitochondria.

Mitochondrial NADH determination
Experiments to determine the mitochondrial NADH changes in hepatocytes under various treatments were performed using the NAD + /NADH assay kit with WST-8 (Beyotime) and BCA Protein Assay kit.Ethanol is oxidized to acetaldehyde by alcohol dehydrogenase, in which NAD + is reduced to NADH, the resulting NADH reduces 2-(2-methoxy-4-nitrophenyl)-3-(4-nitrophenyl)-5-(2,4-disulfophenyl)-2H-tetrazolium sodium salt (WST-8) to formazan with the action of 1-Methoxy-5-methylphenazinium Methyl Sulfate (1-mPMS).After heating at 60 °C for 0.5 h, the NAD + in the samples will decompose and only the NADH will remain.NADH reduces WST-8 to formazan, and the concentrations of formazan produced by the reaction is determined by colorimetry, and the concentrations of NADH in the samples can ultimately be determined.For NAD + /NADH extraction, 200 μL of NAD + /NADH extraction buffer was added in the above mitochondria samples and blow gently, followed by spinning at 12000×g (5 min, 4 °C) to get the supernatant as a test sample.A series of various concentration of NADH standards were prepared.100 μL of samples to be tested were placed in centrifuge tubes and heated at 60 °C for 0.5 h in a water bath.Alcohol dehydrogenase working solutions were mixed with NADH standards and test samples, respectively.After incubating for 10 min at 37 °C in dark place, 10 μL of chromogenic solution was added and incubated for 30 min at 37 °C.Finally, absorbances at 450 nm were recorded.The concentrations of mitochondrial NADH were calculated according to standard curve of NADH standards.

Aspartate aminotransferase activity assays
Aspartate aminotransferase (AST) activity assays were performed using an AST Activity Assay Kit (Elabscience).AST catalyze the reaction between α-ketone glutaric acid and aspartic acid, generating glutamic acid and oxaloacetic acid.Oxaloacetic acid decarboxylates to pyruvic acid, and pyruvic acid and 2,4dinitrobenzene hydrazine generate 2,4-dinitrobenzene hydrazone, which appears as reddish brown in alkaline solution.The AST activity per unit protein weight (IU/g protein) were calculated combined with BCA Protein Assay kit and AST Activity Assay Kit.20 μL of substrate solution was added, followed by the addition of 5 μL of samples.The mixture of substrate solution and samples were mixed and placed at 37 °C for 30 min.
Afterwards, 20 μL of chromogenic agent was added in the mixture and incubated at 37 °C for 20 min.200 μL of Alkali reagent was added and mixed at room temperature for 15 min.The absorbance at 510 nm was measured.The levels of AST in samples were calculated according to standard curve.

Alanine aminotransferase activity assays
Alanine aminotransferase activity assays (ALT) activity assays were performed using an ALT Activity Assay Kit (Elabscience).ALT catalyzed the reaction between alanine and α-ketone glutaric acid, producing glutamic acid and pyruvic acid.2,4-dinitrobenzene hydrazine was added to stop this reaction, generating pyruvate

Figure S11 .
Figure S11.The measurement of hepatocyte injury markers in hepatocytes and mice.(A, B) ALT and AST activity assays in control hepatocytes and HIRI hepatocytes.(C) TNF-α concentrations in the cell supernatant of control hepatocytes and HIRI hepatocytes.(D) LDH release in the cell supernatant of control hepatocytes and HIRI hepatocytes.(E) ALT and AST activity assays in serum of control group mice and HIRI group mice.(F) TNF-α concentrations in the serum of control group mice and HIRI group mice.

Figure S12 .
Figure S12.In situ visualization of O2 •− and ATP dynamics in hepatocytes during the whole process of HIRI.(A) Fluorescence imaging of O2 •− (blue channel, λex = 405 nm, λem = 420-490 nm) and ATP (red channel, λex = 514 nm, λem = 525-668 nm) by UDP (40 μM) in hepatocytes undergoing 0 min, 20 min or 40 min of ischemia and 8 min, 16 min, 24 min, 32 min or 40 min of reperfusion after 40 min of ischemia.(B, C) Relative blue and red fluorescence intensity output of (A).The blue fluorescence intensity of control group was defined as 1.The data are expressed as the mean ± SD. ***P < 0.001.Concordant results were obtained from five independent experiments.

Figure S13 .
Figure S13.Confocal fluorescence imaging of ROS levels in hepatocytes during the whole process of HIRI.(A) Fluorescence imaging of ROS (green channel, λex = 488 nm, λem = 505-540 nm) and nucleus (blue channel, λex = 405 nm, λem = 420-500 nm) by DCFH-DA (10 μM) and Hoechst 33342 (1 μg/mL) in hepatocytes undergoing 0 min, 20 min or 40 min of ischemia and 20 min or 40 min of reperfusion after 40 min of ischemia.(B) Relative green fluorescence intensity output of (A).Note: The green fluorescence intensity of control group was defined as 1.0.The data are expressed as the mean ± SD.Similar results were obtained in five independent experiments.

Figure S17 .Figure S18 .
Figure S17.Hematoxylin and eosin (H&E) staining of major organ tissues (liver, spleen, lung, heart, and kidney) in control group, 100 μM UDP-treated group and 1 mM UDP-treated group.The control group were intraperitoneally injected with saline (0.9 % NaCl) solution every day for two weeks.The UDP-treated group were intraperitoneally administrated UDP for two weeks with experimental concentration of 100 μM or 1 mM.

Figure S21 .
Figure S21.HPLC trace of UDP in 254 nm UV.HPLC analysis was carried out on a Shimadzu LC-16 system equipped with SPD-16 UV−vis detector.Deionized water was used as eluent A and methanol as eluent B. HPLC conditions: 14% of A, 86% of B. The injection volume was 10 μL.The parameters of the HPLC-MS analytical column used were C18-WR, 5 μm, 4.6 mm × 150 mm (GL Sciences).The flow rate is 0.7 mL/min.