Article Cite This: Anal. Chem. XXXX, XXX, XXX−XXX
pubs.acs.org/ac
Method for Quantification of Ribonucleotides and Deoxyribonucleotides in Human Cells Using (Trimethylsilyl)diazomethane Derivatization Followed by Liquid Chromatography−Tandem Mass Spectrometry Zheng Li,†,∥ Hui-Xia Zhang,†,∥ Yan Li,† Christopher Wai Kei Lam,† Cai-Yun Wang,† Wei-Jia Zhang,‡ Vincent Kam Wai Wong,† Su-Seng Pang,§ Mei-Cun Yao,*,‡ and Wei Zhang*,†
Anal. Chem. Downloaded from pubs.acs.org by YORK UNIV on 12/12/18. For personal use only.
†
State Key Laboratory of Quality Research in Chinese Medicines, Macau Institute for Applied Research in Medicine and Health, Macau University of Science and Technology, Taipa, Macau, China ‡ School of Pharmaceutical Sciences, Sun Yat-Sen University, Guang Zhou 510275, China § Louisiana State University, Baton Rouge, Louisiana 70803, United States S Supporting Information *
ABSTRACT: Investigation into intracellular ribonucleotides (RNs) and deoxyribonucleotides (dRNs) is important for studies of the mechanism of many biological processes, such as RNA and DNA synthesis and DNA repair, as well as metabolic and therapeutic efficacy of nucleoside analogues. However, current methods are still unsatisfactory for determination of nucleotides in complex matrixes. Here we describe a novel method for the determination of RN and dRN pools in cells based on fast derivatization with (trimethylsilyl)diazomethane (TMSD) followed by quantification using liquid chromatography−tandem mass spectrometry (LC-MS/MS). Derivatization was accomplished in 3 min, and each derivatized nucleotide not only had a sufficient retention on reversed-phase column by introduction of methyl groups but also exhibited a unique ion transition which consequently eliminated mutual interference in LC-MS/MS. Chromatographic separation was performed on a C18 column with a simple acetonitrile−water gradient elution system, which avoided contamination and ion suppression caused by ion-pairing reagents. The developed method was fully validated and applied to the analysis of RNs and dRNs in cell samples. Moreover, results demonstrated that the applicability of this method could be extended to nucleoside analogues and their metabolites and could facilitate many applications in future studies.
R
The common analytical methods for measuring intracellular RNs and dRNs include enzymatic assay,12,13 radioimmunoassay (RIA),14 and liquid chromatography (LC) with UV15 or tandem mass spectrometry (MS) detection.16,17 The enzymatic assay and RIA are commonly applicable only for deoxyribouncleoside triphosphates (dNTPs) and have limited application. Many approaches have been adopted to enhance the separation of nucleotides in reversed-phase liquid chromatography (RPLC), such as customized columns specific to nucleotides.18,19 However, these molecules are extremely polar and negatively charged which consequently show poor retention on RPLC columns. In addition, several derivatization methods have been introduced for the quantification of nucleotides using RPLC-MS/MS.20,21 However, those derivatization reactions are complex and time-consuming and present
ibonucleotides (RNs) and deoxyribonucleotides (dRNs) play very important roles in the regulation and modulation of many biological processes, and their biological interest has expanded considerably in recent decades. RNs are essential metabolites for the synthesis of lipids, proteins, and RNA and play very important roles in cell signaling.1,2 Intracellular dRN pool sizes are critical for the fidelity of DNA synthesis,3 and defects in dRN metabolism could lead to increased mutation rates3−5 and various human disorders.6,7 Moreover, nucleoside analogues are commonly used as antiviral, anticancer, and immunosuppressive drugs.8,9 Accordingly, many of the associated efficacies of these drugs originate from competing with or altering endogenous nucleotide pools in cells.10,11 Therefore, the quantification of cellular RN and dRN pools is fundamentally important for facilitating the understanding of mechanisms of pathological phenomena and pharmacological agents that are implicated in nucleotide metabolism. © XXXX American Chemical Society
Received: September 19, 2018 Accepted: November 21, 2018
A
DOI: 10.1021/acs.analchem.8b04281 Anal. Chem. XXXX, XXX, XXX−XXX
Article
Analytical Chemistry
oboric acid (HBF4) and TMSD (2 M in hexanes) were provided by Alfa Aesar Co. (Ward Hill, MA, USA). Ultrapure water was produced from a Milli-Q water system (Millipore Co., Burlington, MA, USA). The SPE column Oasis WAX 3 cm3 Cartridge (30 mg, 60 μm) was purchased from Waters Co. (Milford, MA, USA), and the chromatographic column Sepax GP-C18 (2.1 × 150 mm2, 1.8 μm) was obtained from Sepax Technologies (Newark, DE, USA). For culturing cells, phosphate buffered saline (PBS), RPMI medium 1640, 0.25% trypsin−EDTA solution, penicillin− streptomycin solution, and fetal bovine serum (FBS) were obtained from GIBCO Invitrogen Co. (Carlsbad, CA, USA). Human colorectal carcinoma cell line (HCT116) and human non-small-cell lung cancer cell line (A549) were supplied by American Type Culture Collection (ATCC, Rockville, MD, USA). Preparation of Stock Standards and Working Samples. Initial stock solutions at 400 μM (dATP, dGTP, dCTP, and TTP), 500 μm (dAMP, dGMP, dCMP, TMP, dADP, dGDP, dCDP, and TDP), 1 mM (ADP, GDP, CDP, UDP, ATP, GTP, CTP, and UTP), 2 mM (AMP, GMP, CMP, and UMP) were prepared in 50% methanol. These stock solutions were stored at −20 °C in brown glass bottles. The isotope-labeled internal standard solutions (100 mM in 5 mM Tris HCl) were diluted to a concentration of 20 μM in 80% methanol and stored at −20 °C in brown glass bottles. AMP−13C10,15N5 was selected as internal standard (IS) for ribouncleoside monophosphates (NMPs), ribouncleoside diphosphates (NDPs), dNMPs, and dNDPs while ATP−13C10,15N5 was selected as IS for ribouncleoside triphosphates (NTPs) and dNTPs. The working solutions for quality control (QC) samples and calibration curve samples were prepared daily by diluting stock solutions to appropriate concentrations finally in 80% methanol containing an analytefree cell matrix, 4 μM AMP− 13C 10 ,15 N 5 , and 2 μM ATP−13C10,15N5. The analyte-free cell matrix was prepared from cell lysate, by stripping with the activated carbon. Carbon was added to cell lysate at the ratio of 60 mg/mL to remove intracellular nucleotides according to published methods.24,41 The mixture was centrifuged and loaded on an Oasis WAX 3 cm3 Cartridge for SPE, and the eluate was evaporated to dryness under nitrogen. The residue was reconstituted in 80% methanol and stored at −20 °C. Cell Culture and Preparation of Cell Pellets. For development and validation of the method, A549 and HCT116 cells were maintained in RPMI 1640 medium containing 10% (v/v) FBS, 100 UI/mL penicillin, 100 μg/mL streptomycin in a 37 °C humidified incubator with 5% CO2. For testing the potential applicability of this proposed method, A549 cells were treated with gemcitabine (10 μM) and fluorouracil (10 μM) for 24 h, respectively. Cells were collected when they reached approximately 80% confluence. Cells were washed with ice-cold PBS once and then treated with trypsin. Each digested sample was resuspended in 10 mL of ice-cold PBS, followed by centrifugation for 5 min (1500g, 4 °C); then, the pellets were frozen in liquid nitrogen and stored at −80 °C for further analysis. Sample Preparation and Derivatization. Cell pellets were treated with 150 μL of 80% methanol containing 4 μM AMP−13C10,15N5 and 2 μM ATP−13C10,15N5 as IS. Then samples were vortexed for 1 min and placed on ice for 10 min. After centrifugation (13000g) for 15 min at 4 °C, the supernatant was diluted with 1350 μL of 60% methanol
limited applicability. Another widely used approach for nucleotide analysis is RPLC-MS/MS with ion-pairing agents.22−30 We have also developed an ion-pairing LC-MS/ MS method to determine RNs and dRNs and investigated their intracellular levels perturbed by different nucleoside analogues.31−33 Although ion-pairing method demonstrates great reproducibility and good peak shapes for the nucleotides, ion-pairing agents are unfriendly to the LC-MS system due to diminished column life, inevitable system contamination, and notable ion suppression in the electrospray ionization source (ESI). Recently, a few hydrophilic interaction chromatography (HILIC) studies on the separation of nucleotides have been reported using different types of stationary phases.34−40 However, to date HILIC still encounters some inevitable defects in analysis of nucleotides. Nucleotides usually show severe peak tailing and unsatisfactory peak shapes in HILIC. In addition, these methods have not been verified for the applicability for the quantification of deoxyribouncleoside monophosphates (dNMPs) and deoxyribouncleoside diphosphates (dNDPs). Furthermore, the separation of adenosine triphosphate (ATP) and deoxyguanosine triphosphate (dGTP) with the same molecular weight and transition is still challenging and remains the major obstacle in nucleotide analysis since the intracellular concentration of ATP is about 1000-fold greater than that of dGTP.16 We report in this work a novel method for simultaneous quantification of intracellular RNs and dRNs using simple RPLC coupled to MS. A sample preparation based on protein precipitation and solid-phase extraction (SPE) was used to remove the main interferents from the cell matrix. The generated clean cell extract was subjected to a fast derivatization with (trimethylsilyl)diazomethane (TMSD) before LC-MS/MS analysis. The proposed method was validated and applied to determine RNs and dRNs in different cell samples. Furthermore, we believe that the developed method could have a broad application potential for analysis of a variety of highly polar metabolites that are unsuitable for analysis with RPLC.
■
EXPERIMENTAL SECTION Chemicals and Reagents. Adenosine monophosphate (AMP), adenosine diphosphate (ADP), ATP, guanosine monophosphate (GMP), guanosine diphosphate (GDP), guanosine triphosphate (GTP), cytidine monophosphate (CMP), cytidine diphosphate (CDP), cytidine triphosphate (CTP), uridine monophosphate (UMP), uridine diphosphate (UDP), uridine triphosphate (UTP), deoxyadenosine monophosphate (dAMP), deoxyadenosine diphosphate (dADP), deoxyadenosine triphosphate (dATP), deoxyguanosine monophosphate (dGMP), deoxyguanosine diphosphate (dGDP), dGTP, deoxycytidine monophosphate (dCMP), deoxycytidine diphosphate (dCDP), deoxycytidine triphosphate (dCTP), thymidine monophosphate (TMP), thymidine diphosphate (TDP), thymidine triphosphate (TTP), stable isotope labeled adenosine-13C10,15N5-triphosphate (ATP−13C10,15N5), adenosine-13C10,15N5-monophosphate (AMP−13C10,15N5), 30% ammonium hydroxide aqueous solution (NH4OH), and ammonium acetate (NH4OAc) were purchased from Sigma-Aldrich Inc. (St. Louis, MO, USA). MS grade methanolacetonitrile and acetic acid (AcOH) were obtained from J. T. Baker Chemical Co. (Phillipsburg, NJ, USA). Formic acid was bought from Fisher Scientific Co. (Fair Lawn, NJ, USA). Diethyl ether was purchased from Tedia Co. (Fairfield, OH, USA). TetrafluorB
DOI: 10.1021/acs.analchem.8b04281 Anal. Chem. XXXX, XXX, XXX−XXX
Article
Analytical Chemistry
Table 1. MS Parameters for the Quantification of Each Nucleotide Derivative and Internal Standard in the LC-MS/MS Method compd
prototype (m/z)
no. of added methyl
precursor ion (m/z)
product ion (m/z)
tube lens (V)
collision energy (eV)
linear range (nM)
LLOQ (nM)
AMP ADP ATP GMP GDP GTP CMP CDP CTP UMP UDP UTP dAMP dADP dATP dGMP dGDP dGTP dCMP dCDP dCTP TMP TDP TTP AMP−13C1015N5 ATP−13C1015N5
346 426 506 362 442 522 322 402 482 323 403 483 330 410 490 346 426 506 306 386 466 321 401 481 364 521
2 3 4 3 4 5 2 3 4 2 3 4 2 3 4 3 4 5 2 3 4 2 3 4 2 4
374 468 562 404 498 592 350 444 538 351 445 539 358 452 546 388 482 576 334 428 522 349 443 537 392 577
148 374 413 178 404 413 124 350 350 125 351 413 148 358 173 178 303 397 124 334 397 139 349 397 158 418
100 85 102 90 82 98 59 71 85 105 90 102 90 86 97 87 92 93 85 83 59 91 85 85 100 102
25 23 25 27 24 24 25 25 30 35 23 25 20 21 26 23 20 26 24 22 29 26 23 26 25 25
50−5000 0.1−10a 0.5−50a 25−2500 0.1−10a 0.5−50a 25−2500 0.1−10a 0.25−25a 12.5−1250 0.1−10a 0.125−12.5a 1−200 5−1000 10−2000 1−200 5−1000 10−2000 0.5−100 2.5−500 10−2000 0.125−25 1.25−250 10−2000
50 100 500 25 100 500 25 100 250 12.5 100 125 1 5 10 1 5 10 0.5 2.5 10 0.125 1.25 10
a
Unit of analyte is micromoles per liter.
before loading on the SPE column, which was conditioned with 2 mL of methanol and 2 mL of 50 mM NH4OAc solution (pH adjusted to 4.5 with AcOH). After washing with 2 mL of 60% methanol to remove interfering compounds, the analytes were eluted from the cartridge with 2 mL of methanol/water/ ammonia solution (80/15/5, v/v/v) and then evaporated to dryness under nitrogen at room temperature. The residue was reconstituted in 150 μL of 80% methanol and vortexed for 1 min. Then 20 μL of HBF4 (50 μg/mL in methanol), 50 μL of diethyl ether, and 50 μL of TMSD were added and vortexed for 3 min for methylation. Subsequently this reaction was terminated by 10 μL of 50% formic acid in methanol. After centrifugation (13000g) at 4 °C for 10 min, 25 μL of the sample was injected into the LC-MS/MS system for analysis. LC-MS/MS Analysis. All experiments were carried out on a Thermo Fisher TSQ LC-MS/MS system (Thermo Fisher, San Jose, CA, USA). Chromatographic separation was performed on the Sepax GP-C18 column. The column was maintained at 35 °C, and the flow rate was 0.2 mL/min. The mobile phase was composed of water (A) and acetonitrile (B). The column was eluted with a linear gradient system: 0−5−10−12−13−18 min, 12%−30%−40%−55%−12%− 12% B. The autosampler was set at 4 °C. For all RNs and dRNs derivatives, mass detection was carried out using ESI in the negative mode using the following optimized parameters: ion spray voltage, 2800 V; vaporizer temperature, 300 °C; sheath gas pressure, 50 psi; capillary temperature, 320 °C; auxiliary gas pressure, 15 psi. Quantification was performed using multiple reactions monitoring (MRM) as listed in Table 1. For each analyte, the linear range and lower limit of quantitation (LLOQ) were determined on the basis of the expected concentrations of
nucleotides in cell samples. For LLOQ, these values could be set lower if necessary. Method Validation. The linearity, lower limit of detection (LOD), LLOQ, intraday and interday precision, accuracy, matrix effect, SPE extraction recovery, and stability were validated. The details for performing method validation are available in the Supporting Information. Application To Determine RN and dRN Pools in Cells. In order to prove its versatility for several cell classes, this method was applied to determine the concentrations of RN and dRN pools in different human cancer cells. For HCT116 and A549 cells, about 4 million cells contained in each sample were used for the quantitation of RNs and dRNs.
■
RESULTS Method Development. The phosphate groups on the nucleotide make the compound extremely polar and ionic. For nucleotides, alkylation of the phosphonic acid group can decrease their polarity and tendency to bind to metal ions, as well as improve peak shape and the ESI response in MS. Here we used an efficient derivatization reagent TMSD to methylate nucleotides in cell samples. This derivatization reaction was performed at mild reaction condition. The nucleotides reacted completely in the derivatization reaction, and the mass spectrum of the reaction liquid of each nucleotide was provided in Supporting Information Figures S-1 and S-2. Figure 1A shows the secondary mass spectra of methyl derivatives of the representative nucleotides. Based on the mass spectral fragmentation pattern, we confirmed that the methyl group could bond with the phosphate, primary amine, and secondary amine groups. The number of added methyl groups C
DOI: 10.1021/acs.analchem.8b04281 Anal. Chem. XXXX, XXX, XXX−XXX
Article
Analytical Chemistry
Figure 1. (A) Mass spectra of the derivatives of representative nucleotides (AMP, ADP, ATP, GMP, GDP, and GTP) in negative ion mode; (B) general reaction scheme for synthesis of nucleotide derivatives; (C) selected reconstructed MRM chromatograms of derivatives of targeted nucleotides using the LC-MS/MS method.
acentriontrile was adopted. The nucleotide derivatives showed good chromatographic properties on the column. As shown in Figure 1C, all analytes were detected with sufficient retentions, good peak shapes, and satisfactory separations. One limitation of previous methods was the interference resulting from similar or same ion transitions of some nucleotides which hardly achieved chromatographic separation. For example, chromatographic separation of ATP and dGTP was difficult to most investigators. Additionally, ATP and dGTP exhibited the same ion transition (m/z: 506→159) in negative ion mode. Moreover, the concentration of ATP is about 1000-fold greater than that of dGTP in cells. Therefore, the interference of ATP on dGTP quantification is not negligible. In this method, dGTP derivative had one more methyl group than
to the monophosphate group, diphosphate group, triphosphate group, primary amine, and secondary amine was 1, 2, 3, 1, and 1, respectively. The number of added methyl groups depended on the number of phosphate groups, primary amine, and secondary amine on nucleotides, which are listed in Table 1 and Figure 1B. GMP derivative had one more methyl group than other monophosphate nucleotide derivatives because GMP contained both one primary amine and one secondary amine. The same results were confirmed in other nucleotide derivatives based on the above pattern, and detailed mass spectra analysis of all 24 analytes was provided in Supporting Information Figures S-3 and S-41. In this study, a simple LC-MS/MS method employing a typical C18 column and a mobile phase composed of water and D
DOI: 10.1021/acs.analchem.8b04281 Anal. Chem. XXXX, XXX, XXX−XXX
Article
Analytical Chemistry
consequently facilitating pharmaceutical metabolism study of this class of drugs. Furthermore, we believe that other classes of metabolites, such as sugar bisphosphates and organic acids, can also be analyzed satisfactorily by this method.
that of ATP, so chromatographic separation was achieved. More importantly, they exhibited different ion transitions (m/ z: 562→413 (ATP derivative) and 576→397 (dGTP derivative)) in this method (Table 1). Therefore, the interference described above was successfully overcome. Method Validation. The LC-MS/MS method was fully validated in terms of linearity range, LLOQ, intraday and interday precision, accuracy, matrix effect, SPE extraction recovery, and stability and proven to meet the requirement of bioanalysis. The results of method validation are described in the Supporting Information in detail. Application To Determine RN and dRN Pools in Cells. In order to assess the applicability and robustness of the newly developed method, it was applied to determine intracellular RNs and dRNs pools in different cell lines (HCT116 and A549 cells). Since the pool sizes of cellular nucleotides are dynamic and closely linked to various factors, especially the cell cycle, the cell cycle profiles were verified using flow cytometry analysis (Figure 2A). Intracellular RNs and dRNs were
■
DISCUSSION Derivatization. RNs and dRNs are extremely polar compounds with several pKa values, making it difficult to retain them on a common reversed-phase column and to maintain a single species for chromatographic separation. Several methods with derivatization have been reported for the quantification of nucleotides in biological samples by LC-MS/ MS. For example, Nordstrom et al.20 reported the use of propionyl or benzoyl derivatives to increase hydrophobicity and ESI response of AMP, ADP, and ATP. The derivatization was performed in acetonitrile and N-methylimidazole at 37 °C for 30 min. However, this method is useable only for very small amounts of NTPs and dNTPs because of their poor solubility in these relatively hydrophobic solvents. Flarakos et al.21 then developed a method for the synthesis of dGMP derivatives utilizing coupling reagents typically employed in peptide synthesis. The hydrophobic derivatives showed increases in ionization efficiency and improved peak shape in a LC-MS/MS method. Unfortunately, this method is complex and timeconsuming. We intended to find a derivatization reagent with the following properties: (1) simple, mild, and rapid; (2) amenable to small sample sizes and reagent volumes; (3) quantitative derivatization reaction; (4) commercial availability; (5) low cost and low toxicity; (6) high stability of derivatization reagents and products. To satisfy these requirements, TMSD and bis(trimethylsilyl)trifluoroacetamide (BSTFA) were selected as the derivatization reagents. A mixture solution of AMP, ADP, and ATP was prepared as a test example for evaluating the derivatization reagents. AMP, ADP, and ATP can be completely derivatized by TMSD with good yields (Figure 4A). Although AMP was almost derivatized by silylating reaction with BSTFA, ADP and ATP were partly derivatized by BSTFA (Figure 4B). Furthermore, as described above, the number of added methyl groups depended on the number of phosphate groups, primary amines and secondary amines on nucleotides; thus, each derivatized nucleotide not only had a higher retention time on the reversed-phase column, but also exhibited a unique ion transition which consequently eliminated mutual interference in LC-MS/MS. Therefore, TMSD was finally adopted in this method. The key points of TMSD derivatization are methanol and HBF4 for the high-yielding and rapid (
