Serum Metabolomics Study of the Acute Graft Rejection in Human

Mar 18, 2014 - The lower level of serum dehydroepiandrosterone sulfate was found in the acute graft rejection group before transplantation. The result...
0 downloads 0 Views 3MB Size
Download PDF
Article pubs.acs.org/jpr

Serum Metabolomics Study of the Acute Graft Rejection in Human Renal Transplantation Based on Liquid Chromatography−Mass Spectrometry Xinjie Zhao,†,§ Jihong Chen,‡,§ Lei Ye,‡ and Guowang Xu*,† †

Key Laboratory of Separation Science for Analytical Chemistry, Dalian Institute of Chemical Physics, Chinese Academy of Sciences, Dalian 116023, China ‡ Department of Nephrology, the first hospital affiliated of Xinjiang Medical University, 137 Liyushannan Road, Urumqi 830054, China S Supporting Information *

ABSTRACT: Acute graft rejection is one of the most common and serious postcomplications in renal transplantation. A noninvasive method is needed to specifically monitor acute graft rejection. We investigated metabolic alterations of acute graft rejection in human renal transplantation by applying a metabolomics approach. Sera from 11 acute graft rejection subjects and 16 nonacute graft rejection subjects were analyzed by a nontargeted liquid chromatography−mass spectrometry (LC-MS) metabolomics approach including both hydrophilic interaction chromatography and reversed-phase liquid chromatography separations. Discriminative metabolites of acute graft rejection after transplantation were detected, including creatinine, kynurenine, uric acid, polyunsaturated fatty acid, phosphatidylcholines, sphingomyelins, lysophosphatidylcholines, etc. The lower level of serum dehydroepiandrosterone sulfate was found in the acute graft rejection group before transplantation. The results revealed comprehensive metabolic abnormalities in acute graft rejection. The findings are valuable for the clinic noninvasive diagnosis or therapy of acute graft rejection. KEYWORDS: metabolomics, HILIC, RPLC, acute graft rejection, renal transplantation

■

monitoring liver and intestine tissue transplants,8−10 since metabolomics can reflect comprehensive changes of a variety of metabolites in tissues and biofluids.11,12 In renal transplantation metabolomics study,13−18 nuclear magnetic resonance (NMR) had more applications. Serkova et al. discussed the ischemia/ reperfusion injury,13 Kim et al. investigated the levels of serum metabolites in response to immunosuppression after kidney transplantation.14 Li et al. applied a NMR-based metabolomics method to monitor graft function after renal transplantation.15 Matrix-assisted laser desorption/ionization Fourier transform mass spectrometry was also used in renal transplantation study to predict acute cellular renal allograft rejection.16 In our previous work, a nontargeted metabolomics approach based on liquid chromatography−mass spectrometry (LC-MS) including both reversed-phase liquid chromatography (RPLC) and hydrophilic interaction chromatography (HILIC) separations was established to investigate acute graft rejection of renal transplantation in rat models.18 The combined metabolomics approach achieved a comprehensive metabolic fingerprint in rat

INTRODUCTION Renal transplantation has become a most common treatment choice for end-stage renal disease patients.1 A successful renal transplant provides patients with the greatest potential improvement in overall quality of life and prolonged life span.2,3 Acute graft rejection is one of the most common and serious postcomplications in renal transplantation, which is responsible for most death-censored graft loss after the first year.4 The early recognition of acute graft rejection and prompt treatment are important to prevent organ failure.4 At present, the diagnosis of acute graft rejection is based on a combination of clinical evaluation, ultrasound investigation, and biopsy analysis. Urine output, serum creatinine, and ultrasound are most commonly used to detect the acute graft rejections, but they are nonspecific indicators of kidney function. Repeated renal biopsy as an invasive procedure is posing a high risk to the transplant patients. As a consequence, there is a need to find noninvasive methods to specifically monitor acute graft rejection after kidney transplantation. Metabolite measurement is an important aspect for monitoring the function of transplanted organs, and most measurements have been performed using clinical biochemical methods.5−7 Recently, metabolomics has been applied in © 2014 American Chemical Society

Received: January 29, 2014 Published: March 18, 2014 2659

dx.doi.org/10.1021/pr5001048 | J. Proteome Res. 2014, 13, 2659−2667

Journal of Proteome Research

Article

Coulter, Inc., USA). Serum creatinine was analyzed by a picric-acid rate method. Serum urea concentrations were measured using the enzymatic conductivity rate method. All histological specimens were routinely fixed in formalin and embedded in paraffin. Then the hematoxylin-eosin-stained sections were obtained and the major histologic features were assessed.

serum samples, which is a suitable tool to study the metabolic abnormality of the acute graft rejection. In the present work, we used the metabolomics approach to explore the changes of metabolites in the serum of renal transplant patients. Our aim is at investigating metabolic alterations of acute graft rejection to increase the understanding of the rejection mechanism and establishing a noninvasive method specifically for the early diagnosis of acute graft rejection after kidney transplantation.

■

Metabolomics Analysis

The metabolites were detected following our recently published method by using RPLC and HILIC, respectively, coupled with mass spectrometry (MS).18 Sera were thawed at 4 °C. A 200 μL serum was deproteinized with 2 volumes of acetonitrile and centrifuged at 13,000 g for 20 min, and the supernatant was dried in a vacuum centrifuge. Before analysis, the samples were redissolved in 200 μL of acetonitrile and water (8:2). The RPLC and HILIC separations were all performed using an ACQUITY-UPLC system (Waters Corp, Milford, USA). For RPLC separation, a 2.1 mm × 100 mm ACQUITYTM UPLC BEH 1.7 μm C8 column (Waters, Ireland) was used. The mobile phase contained (A) water with 0.1% formic acid and (B) acetonitrile. The gradient was 10% (B) for 0.5 min and was linearly increased to 100% in 24 min and kept for 4 min. Then it was reduced to 10% in 0.1 min and a 2 min of reequilibration period was employed. For HILIC separation, a 2.1 mm × 100 mm ACQUITYTM BEH 1.7 μm HILIC column (Waters, Ireland) packed with unmodified silica gel was used. The mobile phase contained 100 mM ammonium formate with 0.1% formic acid in water (C) and acetonitrile (B). The gradient was 5% (C) for 2.5 min and was linearly increased to 12% in 10 min and then was linearly increased to 50% in 18 min and kept for 3 min and then reduced to 5%, with a 3 min of re-equilibration period employed. The gradients were at a flow rate of 0.35 mL/min, and the column temperatures were kept constant at 35 °C. A 5 μL aliquot of each sample was injected. The UPLC system was coupled to a qTOF-MS (Micromass, Manchester, U.K.) equipped with an electrospray source operating in either the positive or negative ion mode (full scan mode m/z 80−1000). The source temperature was set to 120 °C, cone gas flow to 50 L/h, desolvation gas temperature to 300 °C, desolvation gas flow to 500 L/h, capillary voltage to 3100 V in the positive mode and 2500 V in the negative mode, cone voltage to 35 V, scan time to 0.4 s (using interscan delay of 0.1 s), and collision energy to 4 eV (collision gas: argon). Lock spray was used during data acquirement to ensure accuracy and reproducibility. Leucine enkephalin was used as the lock mass (lock spray frequency: 20 s). Potential biomarkers were identified by accuracy mass, fragmented according to our recently published strategy,22 and then verified by available standards. To achieve accuracy mass, a high resolution MS instrument, LTQ-Orbitrap XL (Thermo Fisher, San Jose, CA) was used at a resolution of 30,000. QC sampling was performed on an UPLC (Waters) coupled to a LTQ-Orbitrap XL (Thermo Fisher, San Jose, CA) with the same separation conditions, and the ESI parameters were set with sheath gas flow 35 arb, aux gas flow rate 5 arb, capillary temperature 325 °C, capillary voltage 21 V, and tube lens 70 V.

MATERIALS AND METHODS

Patients

Twenty-seven primary renal transplant recipients were included in the study. The protocol was approved by the ethical committee of the hospital, and all patients signed an informed consent. The patient information including etiology of primary renal disease was shown in Table 1. All kidneys were harvested Table 1. Patient Information

patient information

primary renal disease

a

age gender duration of dialysis (months) HLA mismatch warm ischemia times (hr) chronic renal failureunknown etiology glomerulonephritis hypertension pyelonephntis diabetes Henoch−Schonlein purpura interstitial nephritis

nonacute graft rejection group (n = 16)

acute graft rejection group (n = 11)

35.5 ± 8.8 male 13, female 3 20.6 ± 9.0

40.6 ± 9.8 male 8, female 3 14.5 ± 10.2

3.7 ± 0.7 2.9 ± 0.6a

3.6 ± 1.0 3.6 ± 1.4

4

4

4 1 2 3 1

3 3 1

1

p < 0.05 compared with acute graft rejection.

from cadaveric nonrelated donors. After transplantation, eleven patients were diagnosed as acute graft rejection, and 16 patients were nonacute graft rejection according to the Banff classification of renal allograft pathology.19 After transplantation, all patients were given immunosuppressants consisting of a combination of 30 mg/day prednisolone, 5 mg/kg/day oral cyclosporin A solution, 1.5−3 g/day mycophenolic acid, and 0.1 mg/kg/day FK506. The cyclosporin A dosage was adjusted to maintain whole blood levels of 250−400 ng/mL during the induction phase and then by reduction to a level of 100−200 ng/mL in the maintenance phase (measured by monoclonal specific RIA20). In this study no patients showed cyclosporin A nephrotoxicity. All rejection patients were treated with intravenous methylprednisolone. For metabolomics analysis, fasting blood samples were collected at pretransplant and seven days post-transplant. Sera were collected and stored following our recently published preanalytical strategy,21 except the preparation of serum was carried out at room temperature.

Data Collection and Data Analysis

Clinical Biochemical and Histological Analysis

All the ion features were extracted and aligned by using the Micromass MarkerLynx Applications Manager version 4.0 (Waters, Manchester, U.K.). The total strength of metabolite

Clinical biochemical analysis of blood samples was performed with the LX20 automated chemistry analyzer (Beckman 2660

dx.doi.org/10.1021/pr5001048 | J. Proteome Res. 2014, 13, 2659−2667

Journal of Proteome Research

Article

ions from one analysis on RPLC-MS or HILIC-MS in the positive or negative mode was respectively normalized. Only the variables having more than 80% of the nonzero measurement values in at least one group were kept (modified 80% rule).23 Then the normalized strengths from four analyses of a sample were integrated into a data set for further data handling to achieve more metabolite information. For the multivariate statistical analysis, the SIMCA-P software was used (version 11.0; Umetrics, Umea, Sweden). After the Pareto scaling, principal component analysis (PCA) and partial least-squares-discriminant analysis (PLS-DA) were performed. The S-plot was used to find discriminative serum metabolites. The 7-fold cross-validation and response permutation testing were used to evaluate the predictive ability of the model. The Wilcoxon Mann−Whitney test was performed on metabolomics data and clinical chemical data, false discovery rate (FDR) was used for multiple comparisons, and p < 0.05 was considered as significant.

■

Figure 2. Typical LC-qTOF-MS base peak intensity chromatograms (m/z 80−1000) from a patient serum sample. (A) RPLC-MS in the positive ion mode; (B) RPLC-MS in the negative ion mode; (C) HILIC-MS in the positive ion mode; (D) HILIC-MS in the negative ion mode.

RESULTS AND DISCUSSION

Clinical Biochemical Data and Histopathological Finding

The diagnosis of acute graft rejection depends upon the pathological findings according to the Banff classification of

renal allograft pathology.19 Under light-microscopic examination, histological sections of the normal glomerulus (Figure 1A) and normal renal tubules (Figure 1B) appeared in the nonacute graft rejection group, whereas the typical morphological alterations of acute graft rejection injury were easily identified in the acute graft rejection group, such as a striking hypertrophy of the glomeruli with a marked increase in the cellularity of the mesangium (Figure 1C), severe tubular dilatation of frank necrotic tubular epithelial cells accompanied by loss into the tubular lumen, and interstitial and perivascular infiltration by mononuclear cells (Figure 1D). The patient information is summarized in Table 1: there were no significant differences in age, gender, duration of dialysis (months), and human leukocyte antigen (HLA) mismatch between the nonacute graft rejection group and the acute graft rejection group. The warm ischemia times of donor kidneys (3.6 ± 1.4 h) in the acute graft rejection group were significantly longer than those in the nonacute graft rejection group (2.9 ± 0.6 h, p < 0.01), which indicated that prolongation of warm ischemia times of donor kidneys was one of the reasons leading to acute graft rejection in the renal transplant recipients. The clinical biochemical data are summarized in Table 2. The two most important indicators of renal function, creatinine and urea, showed a decrease in the nonacute graft rejection group after transplantation, while the levels had no significant differences in the acute graft rejection group between before and after transplantation. Combining the results of renal histopathology, the renal function of the

Figure 1. Histology of graft samples after transplantation. (A) Representative light micrographs of normal glomerulus in the nonacute graft rejection group. (B) Representative light micrographs of normal renal tubules in the nonacute graft rejection group. (C) Representative light micrographs of acute graft rejection glomerulus injury in the acute graft rejection group. (D) Representative light micrographs of acute graft rejection tubules injury in the acute graft rejection group.

Table 2. Clinical Biochemical Data nonacute graft rejection group (n = 16) before transplantation creatinine (μmol/L) urea (mmol/L) AST ALT a

809.3 18.7 26.8 29.4

± ± ± ±

162.0 3.1 4.6 5.3

after transplantation 188.6 15.2 42.1 40.6

± ± ± ±

acute graft rejection group (n = 11) before transplantation

74.8ba 3.8ba 6.5ba 9.4ba

748.6 19.6 28.7 30.9

± ± ± ±

153.0 4.9 4.9 4.4

after transplantation 621.3 18.0 74.9 74.4

± ± ± ±

346.5 5.6 12.7a 11.3a

p < 0.05 compared with before transplantation. bp < 0.05 compared with acute graft rejection. 2661

dx.doi.org/10.1021/pr5001048 | J. Proteome Res. 2014, 13, 2659−2667

Journal of Proteome Research

Article

Figure 3. (A) Scores plots of a PCA model separating pretransplant from post-transplant and separating acute graft rejection from nonacute graft rejection; R2X = 0.595. (B) Scores plots of a PLS-DA model separating acute graft rejection from nonacute graft rejection after transplantation; R2Y = 0.964, Q2 = 0.723; no overflitting was found according to the permutation validation in the PLS-DA model. (C) Scores plots of a OCS-PLS-DA model separating acute graft rejection from nonacute graft rejection before transplantation; R2Y = 0.964, Q2 = 0.723; no overfitting was found according to the permutation validation in the OCS-PLS-DA model. (D) S-plot of the OCS-PLS-DA model separating acute graft rejection from nonacute graft rejection before transplantation. (▲) QC samples, (○) nonacute graft rejection subjects in pretransplant, (*) acute graft rejection subjects in pretransplant, (□) nonacute graft rejection subjects in post-transplant, (■) acute graft rejection subjects in post-transplant.

4149 metabolite features were detected, 781 in RPLC positive mode, 640 in RPLC negative mode, 1634 in HILIC positive mode, and 1049 in HILIC negative mode, including both hydrophilic and hydrophobic metabolites. All 4149 metabolite features were used for the following multivariate statistical analysis after the Pareto scaling. The pooled quality control (QC) samples from all sera were analyzed after each 6 serum samples. In the PCA score plot (Figure 3A), the QC samples are tightly located in the middle, which means the reproducibility of analysis was good for metabolomics study.24

nonacute graft rejection group after transplantation was improved significantly. Besides, the serum levels of aspartate aminotransferase (AST) and alanine aminotransferase (ALT) were measured to monitor the liver function of patients, which was found increased after transplantation, especially in the acute graft rejection group. Analysis of the Metabolite Fingerprint in Serum by UPLC-qTOF-MS

Serum samples were analyzed using our recently published metabolomics approach including HILIC and RPLC separations coupled with qTOF-MS;18 the typical base peak intensity chromatograms of serum were shown in Figure 2. The combined metabolomics approach has been shown to achieve a comprehensive metabolic fingerprint with good analytical characteristics in rat serum samples, and it was a suitable tool to investigate the metabolic abnormality in the acute graft rejection in rat renal transplantation. In this study, patient sera were analyzed with the combined metabolomics approach. After being handled with the modified 80% rule,23 a total of

Identification of Significantly Changed Metabolites in Acute Graft Rejection Post-transplantation

Based on PCA score plot (Figure 3A), the subjects before and after transplantation were clearly separated, and the acute graft rejection subjects were clearly separated from nonacute graft rejection subjects after transplantation. The PCA result suggested that the metabolites were changed by renal transplant surgery, and the difference between the acute graft rejection group and the nonacute graft rejection group existed. A PLS2662

dx.doi.org/10.1021/pr5001048 | J. Proteome Res. 2014, 13, 2659−2667

Journal of Proteome Research

Article

Figure 4. Heat map of differential metabolites. Be and Af mean “before” and “after” transplantation, respectively. NR and AR mean “nonacute graft rejection” and “acute graft rejection”, respectively.

bilirubin, and fatty acid amides. Furthermore, some of the metabolites showed significantly increased only in the acute graft rejection group (middle part of Figure 4), including uric acid, dimethyluric acid and several bile acids. In the lower part of the heat map, the metabolites were shown at lower levels in the acute graft rejection group after transplantation, including kynurenine, xanthine, carnitines, polyunsaturated fatty acids (PUFAs), sphingomyelins (SMs), phosphatidylcholines (PCs), lysophosphatidylcholines (PCs), and lysophosphatidylethanolamine (LPEs), etc.; those metabolites were significantly increased only in the nonacute graft rejection group compared to those in the group before transplantation. Potential Biomarkers of Acute Graft Rejection and Biological Explanation

Based on our UPLC-MS metabolomics data, after transplantation, serum creatinine levels in the acute graft rejection group were significantly higher than those in the nonacute graft rejection group. In reverse, compared with before transplantation, serum creatinine levels were not changed in the acute graft rejection group but were significantly decreased in the nonacute graft rejection group. The result was comparable with the clinical chemical data (Table 2). Creatinine is a breakdown product of creatine phosphate in muscle, blood creatinine is eliminated from the body by kidney filtration, and the increasing of blood creatinine level is an indicator of renal function impairment.25 Similar to creatinine, the decrease of some metabolites, such as fatty acid amides, etc., was possible, owing to recovery of renal function in the nonacute graft rejection group. The serum level of tryptophan (Trp) was also found to be decreased in the nonacute graft rejection group after transplantation. However, kynurenine (Kyn), a metabolite of tryptophan, increased in the nonacute graft rejection group after transplantation. Kyn is synthesized from tryptophan by indoleamine 2,3-dioxygenase (IDO). Recently, more and more evidence showed that IDO is operated as an immunomodulatory enzyme, and the cells expressing IDO can suppress T-cell

Figure 5. Suggestion for a mechanistic model of metabolic alterations in acute graft rejection.

DA model was used (Figure 3B) to further study the metabolic characteristics of acute graft rejection, and an S-plot was used to find discriminative serum metabolites separating nonacute graft rejection and acute graft rejection subjects. The metabolites were identified following our recently published strategy,22 and FDR p < 0.05 was considered as significant. These differential metabolites are shown in a heat map (Figure 4) and in Supporting Information Table S1. In the upper part of the heat map, the metabolites were shown at higher levels in the acute graft rejection group after transplantation. Compared with before transplantation, some of the metabolites were significantly decreased only in the nonacute graft rejection group, but they were not changed in the acute graft rejection group, such as creatinine, valine, 2663

dx.doi.org/10.1021/pr5001048 | J. Proteome Res. 2014, 13, 2659−2667

Journal of Proteome Research

Article

Figure 6. (A) Heat map of sulfated steroids. Comparison of the intensities of sulfated steroids (B−E).

responses and promote tolerance.26−28 IDO is considered important in transplanted organs, not only in understanding the regulation of graft rejection, but also as a potential therapeutic strategy.29,30 In our study, the increase of the ratio Kyn/Trp insinuated the increased IDO activity in the nonacute graft rejection group after transplantation, which may contribute a protective role to relieving the acute rejection after transplantation. The heat map (Figure 4) displays the accumulated serum levels of uric acid and dimethyluric acid and the lower serum levels of hypoxanthine and xanthine in the acute graft rejection group. In vivo, hypoxanthine is oxidized to xanthine by xanthine oxidase (XO). It is further oxidized to uric acid, and oxygen-free radical is generated at the same time.31 XO has been proven to be the source of oxygen radicals produced during ischemia.32 The increase of oxygen radical was related to oxidative stress and lead to inflammation and endothelial dysfunction, which was considered as a possible mechanism in the injury following reperfusion of transplanted organs.32−34 The longer warm ischemia time was one of the important reasons leading to acute graft rejection in the renal transplant recipients. Some lipids (carnitines, choline, several PUFAs, SMs, PCs, LPCs, and LPEs) and gut microbiota-associated metabolites (indoxyl sulfate, p-cresol sulfate, and hippuric acid) were found increased in the nonacute graft rejection group after transplantation, and the serum levels of those metabolites were significantly higher than those of the acute graft rejection group. PUFAs (FFA 20:4, FFA 22:5, and FFA 22:6) are essential fatty

acids. Carnitines are the intermediates for the transport of fatty acids from the cytosol into the mitochondria during fatty acids β-oxidation.35 The increase of FFAs and carnitines represented that fatty acid β-oxidation was accelerated for energy supply in patients of the nonacute graft rejection group. Choline is also an essential dietary nutrient, which composes the head groups of PC and SM.36 Indoxyl sulfate, p-cresol sulfate, and hippuric acid are degraded from dietary protein by gut microbiota.37 Gut microbiota-associated metabolites were closely related to energy metabolism and immune function.38,39 It is possible that the increase of lipids and gut microbiota-associated metabolites was caused by the enhancing nutrient intake and absorption. In addition, all patients were treated with immunosuppressants consisting of a combination of prednisolone, FK506, and cyclosporin A solution. Some studies showed that FK506 and Cyclosporin A can reduce phospholipase A2 activity40 and inhibit release of arachidonic acid.41 Therefore, the lower serum levels of LPCs, LPEs, and PUFAs in the acute graft rejection group may also be from the reduction of phospholipase A2 activity by high-dose immunosuppressants. Moreover, significant reduction of SM species was observed in the serum of the acute graft rejection patients. SM deficiency could be related to alteration in lipid raft formation and enhance the risk of oxidative damage.42,43 Furthermore, the serum levels of bile acids were increased in the acute graft rejection patients after transplant (middle part of Figure 4). Bile acids perform many important physiological functions. The concentrations of bile acids were used as a good 2664

dx.doi.org/10.1021/pr5001048 | J. Proteome Res. 2014, 13, 2659−2667

Journal of Proteome Research

■

indicator of hepatobiliary function.44,45 In our study, the increasing levels of bile acids, AST, and ALT may indicate liver dysfunction in the acute graft rejection group. Based on above results, the suggestion for a mechanistic model of metabolic alterations in acute graft rejection is shown in Figure 5.

ASSOCIATED CONTENT

S Supporting Information *

Table of differential metabolites of the acute graft rejection in human renal transplantation. This material is available free of charge via the Internet at http://pubs.acs.org.

■

Metabolic Characteristics of Acute Rejection Patients before Transplantation

AUTHOR INFORMATION

Corresponding Author

If we can find metabolic characteristics in the patients with acute rejection in advance before transplantation, it will be very helpful to develop different treatment plans. The clinical biochemical data before transplantation did not show a difference between the acute graft rejection group and the nonacute graft rejection group, and metabolite fingerprint also did not show a clear difference in the PCA score plot (Figure 3A). Based on the orthogonal signal correction (OSC) PLS-DA method, the subjects of nonacute graft rejection and acute graft rejection could be separated, and an S-plot was used to find serum discriminative metabolites (Figure 3C and D). Based on univariate t test analysis, we found that some sulfated steroids, such as DHEAS, dihydrotestosterone sulfate (DHTS), androsterone sulfate (ANDS), and etiocholanolone sulfate (ETIOS) showed higher serum levels in the nonacute graft rejection group than those in the acute graft rejection group before transplantation (Figure 6). Sulfated steroids are the sulfated forms of steroid hormones, and DHEAS is the major excreted steroid hormone. DHEA and DHEAS are reported to influence immune function.46,47 The low serum levels of DHEAS were found in the age-related decline in immunity47 and also in rheumatoid arthritis (RA) and systemic lupus erythematosus.48 DHEA can be used to activate immune function.48 In addition, serum DHEAS is a marker for adrenal function, which is closely related to glucocorticoids metabolism.49,50 DHEA and DHEAS were investigated to reduce damage of long-term glucocorticoids51 and as a replacement therapy of glucocorticoids.52 In the study, although the relationship between acute rejection and the serum levels of sulfated steroids is not clear, it is valuable to be further investigated. The possible causes may be associated with the patient’s own immune system or feedback of prednisolone.

■

Article

*Tel./ Fax: +86-411-84379530. E-mail address: [email protected]. cn. Author Contributions §

X.Z. and J.C. contributed equally to this wor.k

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This study has been supported by the foundation (No. 21175132), the creative research group project (No. 21321064) from the National Natural Science Foundation of China, and the science and technology project of Urumqi (No. Y131310009) of China.

■

REFERENCES

(1) Wolfe, R. A.; Ashby, V. B.; Milford, E. L.; Ojo, A. O.; Ettenger, R. E.; Agodoa, L. Y. C.; Held, P. J.; Port, F. K. Comparison of mortality in all patients on dialysis, patients on dialysis awaiting transplantation, and recipients of a first cadaveric transplant. N. Engl. J. Med. 1999, 341 (23), 1725−1730. (2) Silberberg, J. S.; Barre, P. E.; Prichard, S. S.; Sniderman, A. D. Impact of left-ventricular hypertrophy on survival in end-stage renaldisease. Kidney Int. 1989, 36 (2), 286−290. (3) Evans, R. W.; Manninen, D. L.; Garrison, L. P.; Hart, L. G.; Blagg, C. R.; Gutman, R. A.; Hull, A. R.; Lowrie, E. G. The quality of life of patients with end-stage renal-disease. N. Engl. J. Med. 1985, 312 (9), 553−559. (4) Tantravahi, J.; Womer, K. L.; Kaplan, B. Why hasn’t eliminating acute rejection improved graft survival? Annu. Rev. Med. 2007, 58, 369−385. (5) Kotanko, P.; Margreiter, R.; Pfaller, W. Urinary N-acetyl-beta-Dglucosaminidase and neopterin aid in the diagnosis of rejection and acute tubular necrosis in initially nonfunctioning kidney grafts. Nephron 2000, 84 (3), 228−235. (6) Aquino-Dias, E. C.; Joelsons, G.; da Silva, D. M.; Berdichewski, R. H.; Ribeiro, A. R.; Veronose, F. J. V.; Goncalves, L. F.; Manfro, R. C. Non-invasive diagnosis of acute rejection in kidney transplants with delayed graft function. Kidney Int. 2008, 73 (7), 877−884. (7) Tatomirovic, Z.; Bokun, R.; Dimitrijevic, J.; Ignjatovic, L.; Aleksic, A.; Hrvacevic, R. Value of urinary cytology findings in the diagnosis of acute renal graft rejection. Vojnosanit Pregl. 2003, 60 (1), 35−41. (8) Verhelst, X. P. D.; Troisi, R. I.; Colle, I.; Geerts, A.; van Vlierberghe, H. Biomarkers for the diagnosis of acute cellular rejection in liver transplant recipients: A review. Hepatol. Res. 2013, 43 (2), 165−178. (9) Girlanda, R.; Cheema, A. K.; Kaur, P.; Kwon, Y.; Li, A.; Guerra, J.; Matsumoto, C. S.; Zasloff, M.; Fishbein, T. M. Metabolomics of Human Intestinal Transplant Rejection. Am. J. Transplant. 2012, 12, S18−S26. (10) Legido-Quigley, C.; McDermott, L.; Vilca-Melendez, H.; Murphy, G. M.; Heaton, N.; Lindon, J. C.; Nicholson, J. K.; Holmes, E. Bile UPLC-MS fingerprinting and bile acid fluxes during human liver transplantation. Electrophoresis 2011, 32 (15), 2063−2070. (11) Nicholson, J. K.; Lindon, J. C.; Holmes, E. ’Metabonomics’: understanding the metabolic responses of living systems to pathophysiological stimuli via multivariate statistical analysis of

CONCLUSIONS

In this study, a metabolomics approach including RPLC and HILIC separations coupled with mass spectrometry was used to investigate the metabolic characteristics of the acute graft rejection in human renal transplantation. The metabolite fingerprint revealed comprehensive metabolic abnormalities in acute graft rejection, prompting kidney injury, liver damage, oxidative stress, immune and drug response, etc. In addition, the lower serum DHEAS levels before transplantation may suggest higher risk of acute graft rejection. The results indicate that the LC-MS metabolomics approach was a powerful tool in the metabolic abnormality investigation of the acute graft rejection in renal transplantation. The differential metabolites were useful to understand the mechanisms of rejection, and they were valuable for the clinic noninvasive diagnosis or therapy of acute graft rejection. 2665

dx.doi.org/10.1021/pr5001048 | J. Proteome Res. 2014, 13, 2659−2667

Journal of Proteome Research

Article

biological NMR spectroscopic data. Xenobiotica 1999, 29 (11), 1181− 1189. (12) Zhao, X.; Peter, A.; Fritsche, J.; Elcnerova, M.; Fritsche, A.; Haering, H.-U.; Schleicher, E. D.; Xu, G.; Lehmann, R. Changes of the plasma metabolome during an oral glucose tolerance test: is there more than glucose to look at? Am. J. Physiol. Endocrinol. Metab. 2009, 296 (2), E384−E393. (13) Serkova, N.; Fuller, T. F.; Klawitter, J.; Freise, C. E.; Niemann, C. U. H-1-NMR-based metabolic signatures of mild and severe ischemia/reperfusion injury in rat kidney transplants. Kidney Int. 2005, 67 (3), 1142−1151. (14) Kim, C.-D.; Kim, E.-Y.; Yoo, H.; Lee, J. W.; Ryu, D. H.; Noh, D. W.; Park, S.-H.; Kim, Y.-L.; Hwang, G.-S.; Kwon, T.-H. Metabonomic Analysis of Serum Metabolites in Kidney Transplant Recipients With Cyclosporine A- or Tacrolimus-Based Immunosuppression. Transplantation 2010, 90 (7), 748−756. (15) Li, L.; Sui, W.; Che, W.; Li, W.; Chen, J.; Li, H.; Tang, D. e.; Li, M.; Dai, Y. 1H NMR-based metabolic profiling of human serum before and after renal transplantation. ASAIO J. 2013, 59 (3), 286−93. (16) Wang, J.-N.; Zhou, Y.; Zhu, T.-Y.; Wang, X.; Guo, Y.-L. Prediction of acute cellular renal allograft rejection by urinary metabolomics using MALDI-FTMS. J. Proteome Res. 2008, 7 (8), 3597−3601. (17) Wishart, D. S. Metabolomics in monitoring kidney transplants. Curr. Opin. Nephrol. Hypertens. 2006, 15 (6), 637−642. (18) Chen, J.; Wen, H.; Liu, J.; Yu, C.; Zhao, X.; Shi, X.; Xu, G. Metabonomics study of the acute graft rejection in rat renal transplantation using reversed-phase liquid chromatography and hydrophilic interaction chromatography coupled with mass spectrometry. Mol. Biosyst. 2012, 8 (3), 871−878. (19) Racusen, L. C.; Solez, K.; Colvin, R. B.; Bonsib, S. M.; Castro, M. C.; Cavallo, T.; Croker, B. P.; Demetris, A. J.; Drachenberg, C. B.; Fogo, A. B.; Furness, P.; Gaber, L. W.; Gibson, I. W.; Glotz, D.; Goldberg, J. C.; Grande, J.; Halloran, P. F.; Hansen, H. E.; Hartley, B.; Hayry, P. J.; Hill, C. M.; Hoffman, E. O.; Hunsicker, L. G.; Lindblad, A. S.; Marcussen, N.; Mihatsch, M. J.; Nadasdy, T.; Nickerson, P.; Olsen, T. S.; Papadimitriou, J. C.; Randhawa, P. S.; Rayner, D. C.; Roberts, I.; Rose, S.; Rush, D.; Salinas-Madrigal, L.; Salomon, D. R.; Sund, S.; Taskinen, E.; Trpkov, K.; Yamaguchi, Y. The Banff 97 working classification of renal allograft pathology. Kidney Int. 1999, 55 (2), 713−723. (20) Holt, D. W.; Johnston, A.; Marsden, J. T.; Vernillet, L.; Keown, P. A.; Rosano, T. G.; Shaw, L. M.; Rosenthaler, J. Monoclonalantibodies for radioimmunoassay of cyclosporine - a multicenter comparison of their performance with the sandoz polyclonal radioimmunoassay kit. Clin. Chem. 1988, 34 (6), 1091−1096. (21) Yin, P.; Peter, A.; Franken, H.; Zhao, X.; Neukamm, S. S.; Rosenbaum, L.; Lucio, M.; Zell, A.; Haering, H.-U.; Xu, G.; Lehmann, R. Preanalytical Aspects and Sample Quality Assessment in Metabolomics Studies of Human Blood. Clin. Chem. 2013, 59 (5), 833−845. (22) Chen, J.; Zhao, X.; Fritsche, J.; Yin, P.; Schmitt-Kopplin, P.; Wang, W.; Lu, X.; Haring, H. U.; Schleicher, E. D.; Lehmann, R.; Xu, G. Practical approach for the identification and isomer elucidation of biomarkers detected in a metabonomic study for the discovery of individuals at risk for diabetes by integrating the chromatographic and mass spectrometric information. Anal. Chem. 2008, 80 (4), 1280− 1289. (23) Bijlsma, S.; Bobeldijk, L.; Verheij, E. R.; Ramaker, R.; Kochhar, S.; Macdonald, I. A.; van Ommen, B.; Smilde, A. K. Large-scale human metabolomics studies: A strategy for data (pre-) processing and validation. Anal. Chem. 2006, 78 (2), 567−574. (24) Gika, H. G.; Theodoridis, G. A.; Wingate, J. E.; Wilson, I. D. Within-day reproducibility of an HPLC-MS-Based method for metabonomic analysis: Application to human urine. J. Proteome Res. 2007, 6 (8), 3291−3303. (25) Bhayani, S. B.; Rha, K. H.; Pinto, P. A.; Ong, A. M.; Allaf, M. E.; Trock, B. J.; Jarrett, T. W.; Kavoussi, L. R. Laparoscopic partial

nephrectomy: Effect of warm ischemia on serum creatinine. J. Urol. 2004, 172 (4), 1264−1266. (26) Frumento, G.; Rotondo, R.; Tonetti, M.; Damonte, G.; Benatti, U.; Ferrara, G. B. Tryptophan-derived catabolites are responsible for inhibition of T and natural killer cell proliferation induced by indoleamine 2,3-dioxygenase. J. Exp. Med. 2002, 196 (4), 459−468. (27) Taylor, M. W.; Feng, G. S. Relationship between interferongamma, indoleamine 2,3-dioxygenase, and tryptophan catabolism. FASEB. J. 1991, 5 (11), 2516−2522. (28) Terness, P.; Bauer, T. M.; Rose, L.; Dufter, C.; Watzlik, A.; Simon, H.; Opelz, G. Inhibition of allogeneic T cell proliferation by indoleamine 2,3-dioxygenase-expressing dendritic cells: Mediation of suppression by tryptophan metabolites. J. Exp. Med. 2002, 196 (4), 447−457. (29) Tan, P. H.; Bharath, A. K. Manipulation of indoleamine 2,3 dioxygenase; a novel therapeutic target for treatment of diseases. Expert Opin. Ther. Targets 2009, 13 (8), 987−1012. (30) Uyttenhove, C.; Pilotte, L.; Theate, I.; Stroobant, V.; Colau, D.; Parmentier, N.; Boon, T.; Van den Eynde, B. J. Evidence for a tumoral immune resistance mechanism based on tryptophan degradation by indoleamine 2,3-dioxygenase. Nat. Med. 2003, 9 (10), 1269−1274. (31) Hille, R. Structure and function of xanthine oxidoreductase. Eur. J. Inorg. Chem. 2006, 10, 1913−1926. (32) Meneshian, A.; Bulkley, G. B. The physiology of endothelial xanthine oxidase: From urate catabolism to reperfusion injury to inflammatory signal transduction. Microcirculation 2002, 9 (3), 161− 175. (33) Peglow, S.; Toledo, A. H.; Anaya-Prado, R.; Lopez-Neblina, F.; Toledo-Pereyra, L. H. Allopurinol and xanthine oxidase inhibition in liver ischemia reperfusion. J. Hepatobiliary Pancreatic Sci. 2011, 18 (2), 137−146. (34) Yamakawa, Y.; Takano, M.; Patel, M.; Tien, N.; Takada, T.; Bulkley, G. B. Interaction of platelet activating factor, reactive oxygen species generated by xanthine oxidase, and leukocytes in the generation of hepatic injury after shock/resuscitation. Ann. Surg. 2000, 231 (3), 387−398. (35) Vaz, F. M.; Wanders, R. J. A. Carnitine biosynthesis in mammals. Biochem. J. 2002, 361, 417−429. (36) Zeisel, S. H.; da Costa, K.-A. Choline: an essential nutrient for public health. Nutr. Rev. 2009, 67 (11), 615−623. (37) Menon, V.; Kopple, J. D.; Wang, X.; Beck, G. J.; Collins, A. J.; Kusek, J. W.; Greene, T.; Levey, A. S.; Sarnak, M. J. Effect of a Very Low-Protein Diet on Outcomes: Long-term Follow-up of the Modification of Diet in Renal Disease (MDRD) Study. Am. J. Kidney. Dis. 2009, 53 (2), 208−217. (38) Nicholson, J. K.; Holmes, E.; Wilson, I. D. Gut microorganisms, mammalian metabolism and personalized health care. Nat. Rev. Microbiol. 2005, 3 (5), 431−438. (39) Round, J. L.; Mazmanian, S. K. The gut microbiota shapes intestinal immune responses during health and disease. Nat. Rev. Immunol. 2009, 9 (5), 313−323. (40) Anderson, R.; Vanrensburg, C. E. J.; Myer, M. S. Alphatocopherol prevents cyclosporine a-mediated activation of phospholipase a(2) and inhibition of Na+,K+-adenosine triphosphatase-activity in cultured hamster renal tubular cells. Toxicol. Appl. Pharmacol. 1994, 125 (2), 176−183. (41) Gabryel, B.; Chalimoniuk, M.; Stolecka, A.; Waniek, K.; Langfort, J.; Malecki, A. Inhibition of arachidonic acid release by cytosolic phospholipase A(2) is involved in the antiapoptotic effect of FK506 and cyclosporin A on astrocytes exposed to simulated ischemia in vitro. J. Pharmacol. Sci. 2006, 102 (1), 77−87. (42) Pena, L. A.; Fuks, Z.; Kolesnick, R. Stress-induced apoptosis and the sphingomyelin pathway. Biochem. Pharmacol. 1997, 53 (5), 615− 621. (43) Ballou, L. R.; Laulederkind, S. J. F.; Rosloniec, E. F.; Raghow, R. Ceramide signalling and the immune response. Biochim. Biophys. Acta, Lipids Lipid Metab. 1996, 1301 (3), 273−287. (44) Ferraris, R.; Colombatti, G.; Fiorentini, M. T.; Carosso, R.; Arossa, W.; Delapierre, M. Diagnostic-value of serum bile-acids and 2666

dx.doi.org/10.1021/pr5001048 | J. Proteome Res. 2014, 13, 2659−2667

Journal of Proteome Research

Article

routine liver-function tests in hepatobiliary diseasessensitivity, specificity, and predictive value. Dig. Dis. Sci. 1983, 28 (2), 129−136. (45) Gadaleta, R. M.; van Mil, S. W. C.; Oldenburg, B.; Siersema, P. D.; Klomp, L. W. J.; van Erpecum, K. J. Bile acids and their nuclear receptor FXR: Relevance for hepatobiliary and gastrointestinal disease. Biochim. Biophys. Acta, Mol. Cell Biol. Lipids 2010, 1801 (7), 683−692. (46) Buford, T. W.; Willoughby, D. S. Impact of DHEA(S) and cortisol on immune function in aging: a brief review. Appl. Physiol., Nutr., Metab. 2008, 33 (3), 429−433. (47) Khorram, O.; Vu, L.; Yen, S. S. C. Activation of immune function by dehydroepiandrosterone (DHEA) in age-advanced men. J. Gerontol., A: Biol. Sci. Med. Sci. 1997, 52 (1), M1−M7. (48) Straub, R. H.; Weidler, C.; Demmel, B.; Herrmann, M.; Kees, F.; Schmidt, M.; Scholmerich, J.; Schedel, J. Renal clearance and daily excretion of cortisol and adrenal androgens in patients with rheumatoid arthritis and systemic lupus erythematosus. Ann. Rheum. Dis. 2004, 63 (8), 961−968. (49) Arlt, W.; Callies, F.; van Vlijmen, J. C.; Koehler, I.; Reincke, M.; Bidlingmaier, M.; Huebler, D.; Oettel, M.; Ernst, M.; Schulte, H. M.; Allolio, B. Dehydroepiandrosterone replacement in women with adrenal insufficiency. N. Engl. J. Med. 1999, 341 (14), 1013−1020. (50) Tritos, N. A. Serum Dehydroepiandrosterone Sulfate in the Diagnosis of Adrenal Insufficiency: Ready for Prime Time? Endocr. Pract. 2011, 17 (2), 167−169. (51) Ahmed, H. H.; Morcos, N. Y. S.; Eskander, E. F.; Seoudi, D. M. S.; Shalby, A. B. Role of dehydroepiandrosterone in management of glucocorticoid-induced secondary osteoporosis in female rats. Exp. Toxicol. Pathol. 2012, 64 (6), 659−664. (52) Saponaro, S.; Guarnieri, V.; Pescarmona, G. P.; Silvagno, F. Long-term exposure to dehydroepiandrosterone affects the transcriptional activity of the glucocorticoid receptor. J. Steroid Biochem. Mol. Biol. 2007, 103 (2), 129−136.

2667

dx.doi.org/10.1021/pr5001048 | J. Proteome Res. 2014, 13, 2659−2667