Liquid Chromatography–Tandem Mass Spectrometry-Based Plasma

Feb 6, 2013 - Liquid Chromatography−Tandem Mass Spectrometry-Based Plasma. Metabonomics Delineate the Effect of Metabolites' Stability on. Reliabili...
0 downloads 0 Views 2MB Size
Technical Note pubs.acs.org/ac

Liquid Chromatography−Tandem Mass Spectrometry-Based Plasma Metabonomics Delineate the Effect of Metabolites’ Stability on Reliability of Potential Biomarkers Wei Yang,† Yanhua Chen,† Cong Xi,† Ruiping Zhang,† Yongmei Song,‡ Qimin Zhan,‡ Xiaofeng Bi,‡ and Zeper Abliz*,† †

State Key Laboratory of Bioactive Substance and Function of Natural Medicines, Institute of Materia Medica, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing 100050, P. R. China ‡ Cancer Institute and Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing 100021, P. R. China S Supporting Information *

ABSTRACT: Metabonomics is an important platform for investigating the metabolites of integrated living systems and their dynamic responses to changes caused by both endogenous and exogenous factors. A metabonomics strategy based on liquid chromatography−mass spectrometry/mass spectrometry in both positive and negative ion modes was applied to investigate the short-term and long-term stability of metabolites in plasma. Principal components analysis and ten types of identified metabolites were used to summarize the time-dependent change rules in metabolites systematically at different temperatures. The long-term stability of metabolites in plasma specimens stored at −80 °C for five years was also studied. Analysis of these subjects identified 36 metabolites with statistically significant changes in expression (p < 0.05) and found a kind of metabolite with a hundred-fold change. The stability of metabolites in blood at 4 °C for 24 h was also investigated. These studies show that a thorough understanding of the effects of metabolite stability are necessary for improving the reliability of potential biomarkers.

M

study the effect of temperature and delay in processing on biological sample quality by analyzing the scores plots of principal components analysis (PCA),4,19,20 but these studies were not in-depth enough to explore the metabolites’ stability in plasma samples fully. Thus, it is of importance to investigate simultaneously the time-dependent changes in various metabolites based on metabolic profile to assess stability of endogenous metabolites better. In the present study, metabolites’ stability was systematically studied to confirm the reliability of potential biomarkers in liquid chromatography tandem mass spectrometry (LC−MS/ MS)-based plasma metabonomics. The short-term stability of metabolites in plasma at 4 and 37 °C and the long-term stability of metabolites in plasma at −80 °C were investigated by rapid resolution (RR)LC−MS/MS-based metabonomics analysis in positive and negative ion modes. Changes in the metabolites in plasma samples kept at 37 and 4 °C after a delay in sample processing (0, 1, 2, 4, 8, 12, and 24 h) were analyzed. PCA was performed to reveal the clustering changes in the comprehensive metabolites profile. Time-dependent changes in identified metabolites at different temperatures were also

etabonomics is an important platform for understanding the metabolites produced by integrated living systems and their dynamic responses to the changes caused by both endogenous and exogenous factors.1−3 It is primarily based on the study of endogenous low-molecular weight compounds that are the substrates or products of various metabolic pathways.4−6 Therefore the valid data of metabolites is a fundamental prerequisite for a successful identification of potential biomarkers. For metabonomic analysis, a sufficient number of samples are required to reduce the influence of biological variability and obtain statistical significance.7−9 The acquisition of samples in such studies usually spans over months or years.8 In addition, it is usually impractical to process specimens immediately due to limitations in the clinical workflow.10 The metabolites’ instabilities in those processing delays of storage could frequently be ignored, which could seriously affect the research results. So, it is important to examine the stability of endogenous metabolites in metabonomics. Plasma obtained from mammals, especially humans, is a widely utilized source of specimens for metabonomics. Up to now, studies examining the stability of metabolites has focused on only a few types of metabolites, such as amino acids,11,12 acylcarnitines,13,14 lysophosphatidylcholines,15,16 and vitamins,17,18 which do not reflect the overall changes occurring in these samples. Some studies have applied metabonomics to © 2013 American Chemical Society

Received: December 10, 2012 Accepted: February 6, 2013 Published: February 6, 2013 2606

dx.doi.org/10.1021/ac303576b | Anal. Chem. 2013, 85, 2606−2610

Analytical Chemistry



RESULTS AND DISCUSSION Short-Term Stability of Metabolites in Plasma Specimens. The short-term stability of metabolites in plasma was studied by analyzing plasma specimens deposited at 4 and 37 °C for up to 24 h. As illustrated in Figure 1A, plasma specimens

explored. In addition, two sets of plasma specimens from healthy volunteers were analyzed in parallel to study the effect of long-term storage at −80 °C. In consideration of the clinical sampling of blood, the stability of metabolites in blood at 4 °C for 24 h was also investigated. A deeper understanding of the effect of metabolites’ stability on metabonomic analysis will be useful in the search for reliable potential biomarkers.



Technical Note

EXPERIMENTAL SECTION

LC−MS Analysis. Mass spectrometry experiments were performed on the combination of Agilent 1200 Series Rapid Resolution Liquid Chromatography system and a quadrupoletime-of-flight (Q-TOF) LC−MS/MS system (QSTAR EliteTM, Applied Biosystem/MDS Sciex) with an electrospray ion (ESI) source. Chromatographic separation was performed on a Waters Hss C18 column (2.1 × 100 mm, 1.8 μm), and the column temperature was set at 35 °C. The mobile phase consisted of 0.1% formic acid and acetonitrile, with a linear gradient elution. The TOF scan in both positive and negative ion modes was applied for sample analysis. Preparation of Samples. All frozen samples were thawed at 4 °C and vortexed on ice before use. For each 150 μL aliquot of plasma sample, 600 μL of acetonitrile (in an ice-water bath) was added. The mixture was vortexed for 4 min on a vortexmixer (IKA, MS) at 3000 rpm and centrifuged at 10000 rpm at 4 °C for 5 min to remove the protein. The supernatant was collected and evaporated to dryness in a SpeedVac concentrator (Thermo Savant). The sample residues were dissolved in 100 μL of 2% acetonitrile and mixed for 4 min by a vortex-mixer on ice at 3000 rpm. The samples were then filtered with a 96-well plate filter (Agilent, CaptivaND). The preparation of the “mixed standard solution” (MSD) sample and “blank” (BK) sample is shown in the Supporting Information. Study Subjects. All plasma and blood specimens used for the study of short-term stability were taken from male Sprague−Dawley rats (250 ± 20 g) supplied by the Animal Care & Welfare Committee, Institute of Materia Medica, Chinese Academy of Medical Sciences and Peking Union Medical College (no. 0222382). The plasma specimens placed at 4 and 37 °C for 0, 1, 2, 4, 8, 12, and 24 h were prepared to evaluate the effect of delay time and temperature on the shortterm stability plasma specimens. The detailed information of the experimental design is outlined in Figure S1 of the Supporting Information. The stability of blood specimens placed at 4 °C for 24 h was also studied. Two batches of plasma specimens from health volunteers were analyzed simultaneously to study the long-term stability of plasma specimens at −80 °C. The first batch of specimens was collected between 2006 and 2007 (n = 76), and the second batch of specimens was collected in 2011 (n = 80) and analyzed after 2 months of storage at −80 °C. All plasma specimens were obtained from the Cancer Institute and Hospital of the Chinese Academy of Medical Sciences, Beijing, China. The detailed information of LC−MS/MS analysis, standards and reagents, preparation of standard solution, data handling, data quality assessment, study subjects, statistical analysis, and identification of metabolites were shown in the Supporting Information.

Figure 1. PCA scores plots of plasma specimens: (A) Plasma specimens deposited at different temperatures (blue ●: 4 °C, red +: 37 °C) and (B) plasma specimens deposited at 37 °C from 0 to 24 h (red ■: 0 h, orange *: 1 h, yellow▲: 2 h, green ⧫: 4 h, blue ■: 8 h, blue +: 12 h, and purple ○: 24 h).

deposited at 4 °C gathered closely on the left, while the specimens placed at 37 °C were scattered. A dynamic shift in the score plots was also observed for the plasma specimens deposited at 37 °C, which was shown in Figure 1B. As illustrated in this figure, the plasma specimens clustered in three areas: samples frozen immediately concentrated on the left, plasma samples held at 37 °C for 1 and 2 h gathered closely in the middle, and specimens incubated for 4, 8, 12, and 24 h gathered together on the right. It indicated that an obvious change of plasma specimens could be observed within 4 h at 37 °C. Then 67 representative endogenous metabolites (no. 1− 67) in plasma specimens were identified to analyze metabolic changes in detail. The detailed information of these metabolites included 9 amino acids, 12 carnitines, 29 lysophosphatidylcholines (LPCs), 6 lysophosphatidylethanolamines (LPEs), 1 carbohydrate, 3 organic acid salts, 3 bile acids, 2 nucleosides, 1 hormone, and 1 choline, as outlined in Table S1 of the Supporting Information. The changes in metabolite levels in plasma specimens at 0, 1, 2, 4, 8, 12, and 24 h are shown in Figure 2 (change rate of metabolites >50%) and Figure S2 and Table S2 of the Supporting Information. As shown, LPCs, LPEs, carnitines, nucleosides, choline, and serotonin in plasma samples exhibited obvious changes at 37 °C. The concentration of all 29 LPCs in plasma accumulated gradually over 24 h, except for LPC(18:2) (no. 34) and two LPC(20:4). By 24 h, the levels of 20 LPCs had increased by over 100%, and seven had increased by over 500%. The changes observed in the levels of the six LPEs were different: two LPE(16:0) and LPE(18:2) obviously declined with increased storage time, while the two isomers of LPE(18:0) increased over 24 h. When the plasma samples were incubated at 37 °C, only the carnitine C8:0, carnitine C18:1, and carnitine C18:2 decreased with increased incubation time, while the other eight acylcarnitines and carnitine were stable over 24 h. The peak areas of the two nucleosides gradually decreased in plasma samples kept at 37 °C, and their acceptable delay in processing of these samples was found to be 2 and 1 h, respectively. The level of choline continued to increase up until 4 h and then gradually reached a stable state. Meanwhile, the peak areas of serotonin increased by approximately 1.5-fold over 24 h. The nine amino acids, glucose, three organic acid salts, and three bile acids were stable up to 24 h in plasma at 37 °C. Above all, when the plasma 2607

dx.doi.org/10.1021/ac303576b | Anal. Chem. 2013, 85, 2606−2610

Analytical Chemistry

Technical Note

Figure 2. The time-dependent changes of some metabolites after plasma specimens were incubated at 37 and 4 °C for 24 h.

Figure 3. The change of metabolites in plasma stored at −80 °C for 5 years. (A) The 3D scores plot of PCA analysis. (B) The peak areas of 36 identified endogenous metabolites. (C) The peak areas of a type of metabolites with great change.

specimens were placed at 37 °C, the analytes were not affected during the first hour, except for most LPCs, choline, and serotonin. When plasma specimens were held at 4 °C, most metabolites remained stable over 24 h, as shown in Table S2 of the Supporting Information. The only relatively obvious change trend could be seen in LPCs, most of which continued to increase for 24 h (Figure 2). In addition, the level of serotonin also began to increase after 8 h and increased by 79.1% up to 24 h (Figure 2). The rate of change was obviously lower at 4 °C than at 37 °C. Some metabolites, including some amino acids, most LPCs, deoxycytidine, and serotonin, exhibited change statistically when plasma specimens were placed at 4 °C for 24 h, as shown in Figure S3 of the Supporting Information. If the plasma specimen was deposited at 4 °C for 4 h, only valine and most LPCs were unstable. Most LPCs were unstable and increased during the first hour, even if the plasma samples were placed at 4 °C immediately.

Long-Term Stability of Metabolites in Plasma Specimens at −80 °C. The three-dimensional (3D) scores plot of PCA analysis for the two batches of specimens stored for either 2 months or ∼5 years at −80 °C, is shown in Figure 3A. The specimens clustered together by storage time, implying a significant difference between specimens kept at −80 °C for different storage times. The endogenous metabolites that exhibited statistically significant differences are shown in Table S3 of the Supporting Information. In addition, 36 of the metabolites were identified, and detailed information on the determination of these metabolites is provided in Table S4 of the Supporting Information. Five types of metabolites were identified: 27 lysophosphatidylcholines (LPCs), three lysophosphoethanolamines (LPEs), four acylcarnitines, serotonin, and hypoxanthine. The levels of all 36 metabolites were increased, except for the acylcarnitines and hypoxanthine, and the changed peak areas are shown in Figure 3B. The change trends of LPCs, acylcarnitines, and serotonin were the same as that of those of plasma samples stored at 37 °C for 24 h. 2608

dx.doi.org/10.1021/ac303576b | Anal. Chem. 2013, 85, 2606−2610

Analytical Chemistry

Technical Note

sample instability should be considered and removed with the time-dependent rules of metabolites. (4) Changes of metabolites in 4 °C blood anticoagulated with K-EDTA are worthy of attention. Overall, these studies show that awareness of the stability of metabolites is essential to identify reliable potential biomarkers and should not be ignored when performing metabonomics analysis.

In addition, 15 metabolites exhibited significant changes after the plasma specimens had been frozen at −80 °C for five years, but their structures have not yet been identified. As shown in Table S5 of the Supporting Information, these metabolites had the same fragment ions (m/z 88.1, m/z 70.1, and m/z 57.1) and similar fragmentation patterns. As described in “Data handling” of the Supporting Information, these metabolites are preliminarily presumed to be alkylamines. The fold changes in most of the metabolites were dozens of times, and three of the metabolites (compounds 6, 10, and 12) increased over 100 times. The 15 metabolites that were significantly changed are shown in Figure 3C. It is important to note that these metabolites were barely detected in the fresh plasma samples, indicating that these kinds of metabolites should be excluded as potential biomarkers. Explanation of Metabolites’ Change Trends in Plasma Specimens. Most LPCs and serotonin dramatically increased in plasma specimens, even when they were stored at 4 °C or −80 °C. It has been long reported that Lecithin-cholesterolacyltransferase (LCAT) is active in the plasma and characterized by the increase of LPCs in plasma.21 Grant et al. have reported that ethylenediaminetetraacetic acid (EDTA) treatment to platelets enhanced release of serotonin induced by thrombin.22 It was likely that the elevated serotonin in plasma was related to the residual platelets. Slow declination of a few acylcarnitines and nucleosides in plasma could be attributed to the decompostion reaction activated. Short-Term Stability of Metabolites in Blood Specimens at 4 °C. On the PCA score-plot, time-dependent changes in blood specimens that were not obvious were observed when the blood specimens were held at 4 °C. After the blood specimens were placed at 4 °C for 24 h, timedependent changes and acceptable delay time in the 67 metabolites (Table S1 of the Supporting Information, no. 1− 67) that have been identified in plasma specimens are listed in Table S6. Some metabolites, such as some amino acids, carnitine, some long-chain acylcarnitines, some LPCs, LPE(18:0) (no. 55), glucose, lactate, two nucleosides, and serotonin, exhibited changes within 24 h. The time-dependent changes of those metabolites are shown in Figure S4 of the Supporting Information. It is worth noting that some metabolites in 4 °C blood, including valine, three LPCs, glucose, lactate, cytidine, and serotonin, changed during the first hour. It was likely that the blood cells were still active, even when blood specimens were kept at 4 °C for 1 h. The human plasma metabonomics specimens are from clinical, which are always anticoagulated with K-EDTA. Therefore, K-EDTA may not be an apporpriate anticoagulant for blood specimens in metabonomics when delayed procssing of sampling is unavoidable.



ASSOCIATED CONTENT

S Supporting Information *

This material is available free of charge via the Internet at http://pubs.acs.org.



AUTHOR INFORMATION

Corresponding Author

*Address: State Key Laboratory of Bioactive Substance and Function of Natural Medicines, Institute of Materia Medica, Chinese Academy of Medical Sciences and Peking Union Medical College, 1#Xian Nong Tan Street, Beijing 100050, China. Tel/Fax: (+86) 01063165218. E-mail: [email protected]. cn. Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS The study has been supported by the National Natural Science Foundation of China (Grant 21175154) and the National Instrumentation Program (Grant 2011YQ170067).



REFERENCES

(1) Tweeddale, H.; Notley-Mcrobb, L.; Ferenci, T. J. Bacteriol. 1998, 180, 5109−5116. (2) Nicholson, J. K.; Wilson, I. D. Nat. Rev. Drug Discovery 2003, 2, 668−676. (3) Tang, H. R.; Wang, Y. L. Prog. Biochem. Biophys. 2006, 33, 401− 417. (4) Bernini, P.; Bertini, I.; Luchinat, C.; Nincheri, P.; Staderini, S.; Turano, P. J. Biomol. NMR 2011, 49, 231−243. (5) Dettmer, K.; Aronov, P. A.; Hammock, B. D. Mass Spectrom. Rev. 2007, 26, 51−78. (6) Weiss, R. H.; Kim, K. Nat. Rev. Nephrol. 2012, 8, 22−33. (7) Sreekumar, A.; Poisson, L. M.; Rajendiran, T. M.; Khan, A. P.; Cao, Q.; Yu, J.; Laxman, B.; Mehra, R.; Lonigro, R. J.; Li, Y.; Nyati, M. K.; Ahsan, A.; Kalyana-Sundaram, S.; Han, B.; Cao, X.; Byun, J.; Omenn, G. S.; Ghosh, D.; Pennathur, S.; Alexander, D. C.; Berger, A.; Shuster, J. R.; Wei, J. T.; Varambally, S.; Beecher, C.; Chinnaiyan, A. M. Nature 2009, 457, 910−914. (8) Dunn, W. B.; Broadhurst, D.; Begley, P.; Zelena, E.; FrancisMcIntyre, S.; Anderson, N.; Brown, M.; Knowles, J. D.; Halsall, A.; Haselden, J. N.; Nicholls, A. W.; Wilson, I. D.; Kell, D. B.; Goodacre, R. Nat. Protoc. 2011, 6, 1060−1083. (9) Chen, J.; Zhang, X.; Cao, R.; Lu, X.; Zhao, S.; Fekete, A.; Huang, Q.; Schmitt-Kopplin, P.; Wang, Y.; Xu, Z.; Wan, X.; Wu, X.; Zhao, N.; Xu, C.; Xu, G. J. Proteome Res. 2011, 10, 2625−2632. (10) Guo, X. Z; Qiu, L. J. Mod. Lab. Med. 2007, 22, 100−101. (11) Zwart, S. R.; Wolf, M.; Rogers, A.; Rodgers, S.; Gillman, P. L.; Hitchcox, K.; Ericson, K. L.; Smith, S. M. Clin. Biochem. 2009, 42, 907−910. (12) Rosenling, T.; Slim, C. L.; Christin, C.; Coulier, L.; Shi, S.; Stoop, M. P.; Bosman, J.; Suits, F.; Horvatovich, P. L.; StockhofeZurwieden, N.; Vreeken, R.; Hankemeier, T.; van Gool, A. J.; Luider, T. M.; Bischoff, R. J. Proteome Res. 2009, 8, 5511−5522. (13) Fingerhut, R.; Ensenauer, R.; Röschinger, W.; Arnecke, R.; Olgemöller, B.; Roscher, A. A. Anal. Chem. 2009, 81, 3571−3575.



CONCLUSIONS The metabolites’ instabilities were systematically studied by LC−MS/MS-based plasma metabonomics. The following recommendations for improving the reliability of biomarkers in LC−MS/MS-based plasma metabonomics can be derived from the results described above: (1) When plasma specimens cannot be processed immediately due to a restriction in workflow, they should be stored at 4 °C. But caution must be used to these metabolites changed in 4 °C plasma. (2) Metabonomics specimens stored at −80 °C for 5 years present significant concerns when used for later analysis. (3) When metabolites are screened as potential biomarkers, changes from 2609

dx.doi.org/10.1021/ac303576b | Anal. Chem. 2013, 85, 2606−2610

Analytical Chemistry

Technical Note

(14) Mancinelli, A.; Iannoni, E.; Duran, M.; Calvani, M. Clin. Chim. Acta 2007, 375, 169−170. (15) Maślanka, K.; Smoleńska-Sym, G.; Michur, H.; Wróbel, A.; Lachert, E.; Brojer, E. Arch. Immunol. Ther. Exp. 2011, 60, 55−60. (16) Vlaar, A. P. J.; Kulik, W.; Nieuwland, R.; Peters, C. P.; Tool, A. T. J.; van Bruggen, R.; Juffermans, N. P.; de Korte, D. Transfusion 2011, 51, 2358−2366. (17) Hustad, S.; Eussen, S.; Midttun, O.; Ulvik, A.; van de Kant, P. M.; Morkrid, L.; Gislefoss, R.; Ueland, P. M. Clin. Chem. 2011, 58, 402−410. (18) Drammeh, B. S.; Schleicher, R. L.; Pfeiffer, C. M.; Jain, R. B.; Zhang, M.; Nguyen, P. H. Clin. Chem. 2008, 54, 1883−1891. (19) Gika, H. G.; Theodoridis, G. A.; Wilson, I. D. J. Chromatogr. A 2008, 1189, 314−322. (20) Lauridsen, M.; Hansen, S. H.; Jaroszewski, J. W.; Cornett, C. Anal. Chem. 2007, 79, 1181−1186. (21) Glomset, J. A. J. Lipid Res. 1968, 9, 155−167. (22) Percy, A. K.; Schmell, E.; Earles, B. J.; Lennarz, W. J. Biochemistry 1973, 12, 2456−2461.

2610

dx.doi.org/10.1021/ac303576b | Anal. Chem. 2013, 85, 2606−2610