Subscriber access provided by READING UNIV
Article
Twins Derivatization Strategy for High-Coverage Quantification of Free Fatty Acids by Liquid Chromatography-Tandem Mass Spectrometry Ruiqi Jiang, Yu Jiao, Pei Zhang, Yong Liu, Xu Wang, Yin Huang, Zunjian Zhang, and Fengguo Xu Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.7b03020 • Publication Date (Web): 31 Oct 2017 Downloaded from http://pubs.acs.org on October 31, 2017
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 9
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry
Twins Derivatization Strategy for High-Coverage Quantification of Free Fatty Acids by Liquid Chromatography-Tandem Mass Spectrometry Ruiqi Jianga,b , Yu Jiaoa,c , Pei Zhanga,b, Yong Liua,c, Xu Wanga,b, Yin Huanga,b, Zunjian Zhanga,b**, Fengguo Xua,b* ‡
a
‡
Key Laboratory of Drug Quality Control and Pharmacovigilance (China Pharmaceutical University), Ministry of Education,
Nanjing 210009, China b
State Key Laboratory of Natural Medicine, China Pharmaceutical University, Nanjing 210009, China
c
Department of Organic Chemistry, China Pharmaceutical University, Nanjing 210009, China
Abstract: Free fatty acids (FFAs) are vitally important components of lipids that modulate biological metabolism in various ways. Although the molecular structures are simple, the analysis of FFAs is still challenging due to their unique properties and wide concentration range. In the present study, a high-coverage liquid chromatography-tandem mass spectrometry (LC-MS/MS) method was established for the quantification of FFAs in serum samples using two structural analogues 5-dimethylamino-naphthalene-1sulfonyl piperazine (Dns-PP) and diethylamino-naphthalene-1-sulfonyl piperazine (Dens-PP) as twins derivatization reagents. The Dns labeling of FFAs could significantly enhance their MS response via the introduction of easily ionizable moiety of tertiary aminecontaining part and aid fragmentation in the multiple reaction monitoring (MRM) mode. Our results demonstrated that the detection sensitivities of FFAs were increased by 50-1500 folds compared with non-derivatization method. At the same time, Dens labeled standards were used as one-to-one internal standards to ensure accurate quantifications. Thirty-eight FFAs, covering short-, mediumand long-chain, could be quantified in wide dynamic range with lower limit of quantification (LLOQ) varied from 2 to 20 nM. Using this method, we analyzed serum FFAs in rat models of cisplatin-induced nephrotoxicity and irinotecan-induced gastrointestinal toxicity, respectively. The findings were further compared with those revealed by previous untargeted metabolomics. The results indicate that twins derivatization based LC-MS provides more accurate view of global FFAs alternation and has great application potential in the fields of targeted metabolomics.
Free fatty acids (FFAs) could be functionally divided into short-chain fatty acids (SCFAs, ≤6 C atoms), medium-chain fatty acids (MCFAs, 7-12 C atoms) and long-chain fatty acids (LCFAs, >12 C atoms) depending on chain length. Despite quite simple molecular structures, they fulfil multiple critical functions in biological regulation: (1) FFAs especially SCFAs1 and LCFAs2 regulate energy metabolism; (2) FFAs are vital components of membrane lipids and involved in membranemediated cellular functions;3,4 and (3) FFAs play a crucial role in various signal pathways and modulate biological metabolism.5-7 Altered levels of FFAs have been observed within a variety of diseases including cardiovascular risk,8 hepatocellular carcinoma,9 insulin resistance,10 type 2 diabetes,11 schizophrenia12 and Alzheimer’s disease.13 SCFAs, in addition, as the products of dietary fibers fermented by gut microbiota, have been found to affect both colonic morphology and function.14 Recent studies have uncovered that they also modulate systemic immunity to play their anti-inflammatory role.15,16 Increasing interest has been taken in the interactions between gut microbes and host metabolism, where the effect of SCFAs is indispensable.17-19
Given the crucial roles of FFAs in physiological and pathological processes, growing attentions have been paid to the development of relevant analytical methodologies. Gas chromatography-mass spectrometry (GC-MS) is an excellent instrumental platform for FFAs analysis. In order to increase the volatility and thermal stability, methyl esterification or silylation derivatization is usually conducted prior to GC-MS analysis. Electron impact ionization (EI) is the most commonly used ionization mode in GC-MS, however it is too hard for most small molecules and has trap in FFAs analysis. Our previous study revealed that peaks of fatty acids in derivatized GC-MS chromatogram were the result of contributions from structurally related compounds, in both free and conjugated forms. This easily ignored multi-peak and multi-origination phenomenon may lead to perplexed interpretations.20 Moreover, the low vapor pressure of FFAs, particularly SCFAs, is a major pitfall for straightforward GC-MS analysis due to the losses during biological sample preparation.21,22 With flexible separation mechanisms, soft ionization technique, specific monitoring, liquid chromatography-mass spectrometry (LC-MS) is becoming a promising platform for FFAs analysis. Hellmuth et
1
ACS Paragon Plus Environment
Analytical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
al. determined the profile of nonesterified fatty acids (NEFAs) by “differential-energy” LC-MS multiple reaction monitoring (MRM) protocol, providing unbiased detection of NEFAs species.23 Zehethofer et al. utilized liquid chromatographytandem mass spectrometry (LC-MS/MS) to profile FFAs in human plasma via postcolumn addition of barium cations.24 In spite of the powerful capabilities of LC-MS methods, the following issues remain to be addressed when it comes to the analysis of FFAs: (1) low ionization efficiency of FFAs per se in the negative ion mode;23,25-27 (2) lack of characteristic MRM transitions based on the main fragmentation;28,29 (3) difficulty in large-scale quantification induced by various carbon chains and wide concentration range; (4) challenges in structural elucidation of FFAs species, particularly C=C location isomers.30 Chemical derivatization-based LC-MS is an emerging strategy for FFAs analysis, which might overcome the drawbacks mentioned above.31-34 Yang et al. attached a quaternary amine derived from 2-bromo-1-methylpyridinium iodide and 3-carbinol-1-methylpyridinium iodide to FFAs species, enhancing the detection sensitivities up to 2500 folds compared with underivatized ones.35 Zhang and co-workers developed a relative quantification LC-MS/MS strategy for FFAs species with 2,4-bis(diethylamino)-6-hydrazino-1,3,5triazine derivatization. General MRM conditions were performed, benefitting from the common fragment ion at m/z 200.0.36 Leng et al. compared a series of piperazine-based derivatization reagents referring to computer-calculated gasphase hydrogenation capacity and hydrophobicity, 2,4dimethoxy-6-piperazin-1-yl pyrimidine (DMPP) was selected to successfully measure 18 low-abundance FFAs in human urine.37 Another, ten SCFAs (C2-C6) were derivatized by 3nitrophenylhydrazine (3NPH), where phenyl group eased the baseline separation of pairs of isomers.38 Wang et al. developed a charge-remote fragmentation strategy via an amidation reaction. After being derivatized by a charge-carried reagent, different FFAs isomers could be identified based on distinguished fragements.30 Besides, reagent containing a labeled pyridinium side chain was exploited for the quantification of FFAs. The multiplexed method could resolve biological and analytical variance.39 Although many efforts have been taken, up to now there is no single LC-MS method available for simultaneous determination of short-, mediumand long-chain FFAs to quantify the global alternation accurately. In this paper, a high-coverage, sensitive and selective LCMS/MS absolute quantification method for thirty-eight FFAs (C2-C24) is described, based on 5-dimethylamino-naphthalene1-sulfonyl piperazine (Dns-PP) and diethylamino-naphthalene1-sulfonyl piperazine (Dens-PP) twins labeling. The targeted method was applied to determine the levels of FFAs in serum samples obtained from rat models of cisplatin-induced nephrotoxicity and irinotecan-induced gastrointestinal toxicity. The results were also compared with the ones obtained from our previous untargeted metabolomics.
Page 2 of 9
(HATU) were all purchased from Sigma-Aldrich (St. Louis, MO, USA). HPLC grade acetonitrile and methanol were obtained from Merck (Darmstadt, Germany). Ethyl acetate and formic acid were of analytical grade and purchased from Nanjing Chemical Reagent Co. (Jiangsu, China). Deionized water was provided by a Milli-Q purification system (Millipore, Watford, UK). Twins derivatization reagents, 5dimethylamino-naphthalene-1-sulfonyl piperazine (Dns-PP) and diethylamino-naphthalene-1-sulfonyl piperazine (Dens-PP) were both synthesized in house, as described in Note S1. Rat Models and Collection of Serum Samples. Orbital venous blood samples were collected from two types of rat models, both from our previous studies, namely cisplatininduced nephrotoxicity40 and irinotecan (CPT-11)-induced gastrointestinal toxicity.41 Samples of nephrotoxicity model were collected from groups of middle-dose cisplatin (NT, n = 6) and healthy control (NC, n = 6) at day 5 after cisplatin administration. As for gastrointestinal toxicity model, blood samples were collected from CPT-11 group (GIT, n = 10) and control group (GIC, n = 10). Serum was harvested by centrifugation (6000 g, 4 ℃, 10 min) after coagulation for 1 h and then stored at -80 ℃ prior to analysis. Analytical Procedure. Stock and Working Solutions. Standard stock solutions of FFAs were prepared in methanol to obtain concentrations of 10 mM, except for behenic acid and lignoceric acid of 1 mM due to poor solubility. Fourteen-level mixed standard working solutions, also served as calibrators, were prepared with acetonitrile through serial dilution of stock solutions to provide concentrations of 0.01, 0.02, 0.04, 0.1, 0.2, 0.4, 1, 2, 4, 10, 20, 40, 100 and 200 µM for method validation. Likewise, five-level quality control (QC) working solutions, namely 0.05, 0.5, 5, 25, 75 µM, were prepared in a similar manner to provide low, medium and high situations. Solutions of Dns-PP, Dens-PP and HATU were all prepared in acetonitrile at the concentrations of 12 mM, 12 mM and 9 mM, respectively. They were diluted with acetonitrile to desired level before use. All stock and working solutions were prepared freshly. Derivatization Optimization. To optimize the reaction condition, a mixed standard solution containing 5 μM of each of the thirty-eight FFAs was utilized as the typical example. Twenty microliters of mixed standard solution, 20 μL of DnsPP solution, 20 μL of HATU solution and 40 μL of 1:1 water/acetonitrile (v/v) were mixed. The mixtures were allowed to react under different concentrations of Dns-PP (3, 6, 9, 12 mM), HATU (3, 6, 9 mM) and at different time points (10, 30, 60, 90, 120, 150, 180 min) individually, where the temperature of incubation was maintained at 37 ℃. Then, the resulting solutions were centrifuged (20000 g, 4 ℃, 10 min) before LCMS/MS determination. Twins Derivatization-Internal Standards. Briefly, 1 mL of mixed standard solution containing 0.5 μM of each FFA were pipetted into a 10-mL glass tube. And then 1 mL of 9 mM DensPP solution, 1 mL of 3 mM HATU solution and 2 mL of 1:1 water/acetonitrile (v/v) were sequentially added. After vortex mixing, the mixture was incubated at 37 ℃ for 150 min. The resulting solution was used as the twins derivatization-internal standards (TD-ISs) mix. Sample Preparation. FFAs were extracted from serum samples on the basis of the approach utilized by Zhu et al.,42 with a slight modification. An aliquot of 50 μL of serum sample
EXPERIMENTAL SECTION Chemicals and Reagents. Standard compounds of FFAs (listed in Table 1) and 1-[Bis(dimethylamino)methylene]-1H1,2,3-triazolo[4,5-b]pyridinium 3-oxid hexafluorophosphate
2
ACS Paragon Plus Environment
Page 3 of 9
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry
was transferred into a 1.5 mL Eppendorf tube containing 10 μL of 0.5% formic acid in water and 200 μL of ethyl acetate. After vortex mixing for 5 min, the resulting mixture was centrifuged (6000 g, 4 ℃, 10 min), then the organic layer was pipetted into a glass tube and evaporated to dryness at 37 ℃ under nitrogen. All of the dried samples were stored at -80 ℃ for further derivatization. FFAs Derivatization. The derivatization mechanism is shown in Scheme 1. Briefly, 20 μL of acetonitrile, 20 μL of 9 mM DnsPP solution, 20 μL of 3 mM HATU solution and 40 μL of 1:1 water/acetonitrile (v/v) were sequentially added to each dried sample, followed by vortex mixing for 5 min. The mixture was then allowed to be incubated at 37 ℃ for 150 min for derivatization. Each Dns-PP derivatized serum sample was mixed with TD-ISs equally before determination. The mixed solution was then centrifuged (20000 g, 4 ℃, 10 min) and an aliquot of supernatant (2 μL) was directly analyzed by LCMS/MS. Scheme 1. FFAs Derivatization Mechanism
LC-MS/MS Analysis. A Shimadzu Nexera UFLC system coupled to MS-8040 triple quadrupole mass spectrometer (Shimadzu Co., Tokyo, Japan) equipped with an electrospray ionization (ESI) source was applied for the analysis of FFAs. The autosampler was kept at 4 ℃. Chromatographic separations were implemented on an Agilent Zorbax Eclipse XDB-C18 column (2.1 × 100 mm, 1.8 μm, Agilent Technologies, Santa Clara, CA) with a flow rate of 0.4 mL/min at 50 ℃. The mobile phase A was 0.1% formic acid in water, and B was methanol for gradient elution. The binary solvent gradient condition was optimized as follows: 0-5 min at 52% B, 5-20 min from 52 to 78%, 20-23 min at 78%, 23-29 min from 78 to 100%, 29-37 min at 100%, and maintained at 52% mobile phase B for additional 2 min for re-equilibration. Sensitive and selective detection of derivatized FFAs was performed by MRM in positive ion mode. The optimal parameters of mass spectrometer were as follows: spray voltage, 4.5 kV; nebulizing gas, 3 L/min; drying gas, 15 L/min; heat block temperature, 400 ℃; desorption line (DL) temperature, 250 ℃. The optimized MRM transitions and parameters of analytes are given in Table 1. All data were acquired utilizing LabSolutions LCMS software version 5.53 (Shimadzu Co., Tokyo, Japan).
Figure 1. Comparison of Dns-PP derivatization efficiency for thirty-eight FFAs under different reaction conditions: (A) effect of Dns-PP concentration, (B) effect of HATU concentration, (C) effect of reaction time. The concentration of mixed FFAs standard solution was 0.5 µM. The error bars were plotted according to standard errors of normalized peak areas determined in triplicates.
Method Validation. The proposed method was validated according to the criteria described in the FDA guidelines for bioanalytical method validation.43
Linearity and Sensitivity. Calibration was conducted based on the analysis of derivatized fourteen-level FFAs standard working solutions. Calibration curves were established by
3
ACS Paragon Plus Environment
Analytical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
employing a linear regression with a weighting factor of 1/x2. The lower limit of quantification (LLOQ), estimated at signalto-noise ratio near or above 10, was determined in five replicates for each FFA. Precision and Accuracy. Intra- and inter-assay precision and accuracy were assessed by analyzing QC standards at five different levels. The intra-assay accuracy and precision were estimated using recovery and coefficient of variation (CV) of selected QC concentration within a run (n = 5), respectively. Inter-assay accuracy and precision were determined on three consecutive days (n = 15). Stability. Autosampler storage stability of all FFAs derivatives was evaluated by analysis of the QC samples of 0.5 and 25 μM at 8 h intervals over 48 h (4 ℃). The stabilities of FFAs standard stock solutions at room temperature for 6 h and at -20 ℃ for 72 h were also assessed. Extraction Efficiency. In the absence of truly FFAs-free matrix, biological blank samples were prepared by adsorption on activated charcoal.44,45 To evaluate the extraction efficiency, processed blank serum samples spiked at 0.5 and 25 μM were prepared. Extraction efficiency was defined as the ratio of response of FFAs extracted from spiked sample to the one of extracted blank serum sample to which same concentrations of FFAs were spiked post-extraction representing 100% efficiency of extraction. Five replicates for each measurement were performed. Matrix Effect. Pooled serum sample was processed as described in Sample Preparation section, the organic layer was evaporated to prepare serum matrix. FFAs standard working solutions at 0.5 and 25 μM were derivatized by Dns-PP, respectively. The resulting solutions were then added to the above dried serum matrix after mixed equally with TD-ISs to estimate the influence of matrix ions. Here, IS-normalized matrix factor (MF) of each FFA is defined as below, where R represents the response of FFA or IS:46 ∙
Page 4 of 9
responses of intact FFAs before and after derivatization, taking cis-5,8,11,14,17-eicosapentaenoic acid, heneicosanoic acid, cis4,7,10,13,16,19-docosahexaenoic acid and nervonic acid as examples. No intact FFAs could be detected after the standard solution was derivatized by Dns-PP (Figure S1). Strategy of Twins Derivatization-Internal Standards. Because of the large diversity of endogenous FFAs, it is of great difficulty to synthesize IS for each FFA. It appears to be a belief that a stable isotopically labeled (SIL) analogue is the preferred internal standard utilized in LC-MS/MS method due to its almost identical physicochemical properties to target compound.48,49 The SIL theoretically co-elutes with analyte, hence compensating the degree of ionization enhancement or suppression caused by co-eluting sample matrix. However, SIL analogues are less likely to be available or very expensive, especially synthesizing exclusively non-deuterium labeled SIL internal standards.
Figure 2. Workflow of twins derivatization for absolute quantification of FFAs.
In view of the structural similarities of internal standards to target analytes, here, we synthesized Dens-PP as the twins derivatization reagent of Dns-PP. A novel strategy of twins derivatization was performed for the quantification of FFAs in serum samples, as an alternative of SIL (Figure 2). As the mix of Dns-PP-derivatized serum sample with Dens-PP-derivatized FFAs standard compounds, thirty-eight structural analogues were added as one-to-one internal standards. The stability of all TD-ISs was also evaluated as they were stored in autosampler for 48 h (4 ℃). The CVs of six injections were ≤7% (Table S1).
.
RESULTS AND DISCUSSION Derivatization Optimization. This study was aimed to develop a sensitive and selective chemical derivatization-based LC-MS/MS method for the quantification of thirty-eight FFAs (C2-C24). Here, we developed a well-designed LC-MSoriented derivatization reagent, 5-dimethylamino-naphthalene1-sulfonyl piperazine (Dns-PP), based on the subunits as follows: (1) piperazine group has been shown excellent reactivity toward carboxyl group;47 (2) naphthalene nucleus of Dns-PP facilitates chromatographic retention of FFAs derivatives per se and separates them from other underivatized small molecules to eliminate the competition for ionization; and (3) both tertiary amine and piperazine group with high proton affinity could promote the ionization process, resulting in enhanced sensitivity. As shown in Figure 1A, the effect of 6-12 mM Dns-PP on the derivatization reaction did not show significant difference, thus 9 mM Dns-PP was used for all the subsequent analysis. The concentration of HATU did not affect the reaction as indicated by the plots in Figure 1B, thus 3 mM HATU was chosen. Figure 1C shows the time period of derivatization. All FFAs derivatives have been to the maximum at the point of 150 min. In addition, we also compared the
Figure 3. Fragmentation of Dns-PP labeled FFAs.
High-coverage quantification. Low ionization efficiency and wide concentration range of FFAs hindered high-coverage quantification. In this paper, Dns-PP labeling could remove the above obstacles. After derivatization, the sensitivities were significantly enhanced, approximately 50-1500 folds higher than intact FFAs detected in negative ion mode. Thus, the FFAs could be detected as many as possible, especially for lowabundance ones. In addition, the derivatives could produce alternative product ions at m/z 170.1 and 320.1, as shown in Figure 3. As for abundant FFAs, the ion at m/z 170.1 with lower response was chosen as the product ion in MRM mode to avoid
4
ACS Paragon Plus Environment
Page 5 of 9
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry
saturation (Table 1). Thus, low- and high-abundance FFAs (C2C24) could be determined simultaneously. LC-MS/MS Analysis. The monitored MRM parameters of analytes are detailed in Table 1. Representative product ions at m/z 170.1 and 320.1 derived from Dns-PP moiety were generated for all analytes (Figure S2). Similarly, fragment ion at m/z 348.2 related to Dens-PP was also observed for all
internal standards (Figure S3). The separations of thirty-eight derivatized-FFAs, including six pairs of structural isomers (Table S2), were optimized on a C18 column. Representative extracted ion chromatograms acquired from a mixed standard solution of FFAs after derivatization are shown in Figure 4. Averagely, one minute-analysis for each analyte was achieved despite pairs of isomers.
Figure 4. Representative extracted ion chromatograms of Dns-PP derivatized FFAs: (1) acetic acid, (2) propionic acid, (3) isobutyric acid, (4) butyric acid, (5) 2-methylbutyric acid, (6) isovaleric acid, (7) valeric acid, (8) 3-methylpentanoic acid, (9) 4-methylvaleric acid, (10) hexanoic acid, (11) 2-methylhexanoic acid, (12) heptanoic acid, (13) 2-ethylhexanoic acid, (14) 2-methylheptanoic acid, (15) octanoic acid, (16) 4-methyloctanoic acid, (17) nonanoic acid, (18) decanoic acid, (19) dodecanoic acid, (20) tridecanoic acid, (21) myristic acid, (22) cis5,8,11,14,17-eicosapentaenoic acid, (23) linolenic acid, (24) palmitoleic acid, (25) pentadecanoic acid, (26) cis-4,7,10,13,16,19docosahexaenoic acid, (27) arachidonic acid, (28) linoleic acid, (29) palmitic acid, (30) oleic acid, (31) stearic acid, (32) cis-11-eicosenoic acid, (33) nonadecanoic acid, (34) erucic acid, (35) heneicosanoic acid, (36) behenic acid, (37) nervonic acid, (38) lignoceric acid.
Method Validation. To validate the proposed method, a series of attributes were estimated, including sensitivity, calibration range, linearity, accuracy, precision, stability, extraction efficiency and matrix effect. Here, three types of stability were evaluated during the validation (Table S3). The derivatized FFAs were stable in the autosampler at 4 ℃. The FFAs standard stock solution were also stable at room temperature for 6 h or stored at -20 ℃ for 72 h. Sensitivities were evaluated by the determination of LLOQ for all derivatives. As shown in Table 1, all analytes had LLOQ lower than 20 nM, with ten of them at 2 nM, showing similar capability in MS response enhancement with DMPP and 3NPH.37,38 The linearities of calibration curves for all analytes were excellent among the validated concentration range of 0.02 µM to 100 µM, with correlation coefficients (R2) greater than 0.99 (Table 1). The intra- and inter-assay performance was
assessed by analyzing five-level QC standards, representing the entire calibration range. For intra- and inter-assay accuracy, recoveries of all analytes were about 92-111% and 91-113%, respectively (Table S4). The CVs of intra- and inter-assay were ranged from 1.0 to 7.8% and 1.2 to 10.4%, respectively (Table S4). Extraction efficiencies of FFAs, as determined in five replicates, were about 82-97% at 0.5 µM and 76-94% at 25 µM. The CVs for two levels were blow 15%, demonstrating favorable sample preparation (Table S5). As for LC-MS, the presence of invisible co-eluting components in matrix may affect the quantification of analytes, namely matrix effect. Here, IS-normalized MFs were determined to evaluate ionization enhancement or suppression. As shown in Table S5, the interferences of matrix to analytes were almost ≤20%. In conclusion, the FFAs-targeted method not only showed excellent reliability of quantification but also encompassed higher coverage compared with previous methods.23,35,37,38 Table 1. Mass spectrometer parameters, sensitivity and linearity of LC-MS/MS assay CE (V)
LLOQa (nM)
LLOQb (µM)
Linear range (µM)
R2
241
-18
10
4
0.1-100
0.9972
320.1
159
-18
20
4
0.2-100
0.9959
390.3
320.1
159
-20
10
8
0.1-100
0.9966
390.3
320.1
159
-20
10
6
0.1-40
0.9979
No.
Analytes
Parent (m/z)
Fragment (m/z)
1
acetic acid
362.2
320.1
2
propionic acid
376.2
3
isobutyric acid
4
butyric acid
Dwell time (ms)
5
ACS Paragon Plus Environment
Analytical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 6 of 9
5
2-methylbutyric acid
404.1
320.1
159
-19
10
4
0.1-40
0.9961
6
isovaleric acid
404.1
320.1
159
-19
4
6
0.04-40
0.9982
7
valeric acid
404.1
320.1
159
-19
10
4
0.1-100
0.9932
8
3-methylpentanoic acid
418.3
320.1
159
-22
4
6
0.04-100
0.9924
9
4-methylvaleric acid
418.3
320.1
159
-22
10
2
0.1-100
0.9961
10
hexanoic acid
418.3
320.1
159
-22
4
1
0.04-40
0.9970
11
2-methylhexanoic acid
432.3
320.1
119
-19
10
2
0.1-40
0.9969
12
heptanoic acid
432.3
320.1
119
-19
4
2
0.04-100
0.9976
13
2-ethylhexanoic acid
446.3
320.1
119
-21
2
2
0.02-100
0.9966
14
2-methylheptanoic acid
446.3
320.1
119
-21
2
2
0.02-100
0.9970
15
octanoic acid
446.3
320.1
119
-21
10
0.5
0.1-40
0.9943
16
4-methyloctanoic acid
460.3
320.1
94
-22
4
2
0.04-40
0.9950
17
nonanoic acid
460.3
320.1
94
-22
20
2
0.2-100
0.9970
18
decanoic acid
474.1
320.1
66
-23
4
2
0.04-40
0.9964
19
dodecanoic acid
502.2
320.1
66
-22
4
1
0.04-100
0.9961
20
tridecanoic acid
516.4
320.1
41
-23
2
1
0.02-40
0.9973
21
myristic acid
530.4
170.1
37
-23
4
0.5
0.04-40
0.9986
22
cis-5,8,11,14,17eicosapentaenoic acid
604.2
320.1
37
-24
10
1
0.1-100
0.9967
23
linolenic acid
580.2
170.1
27
-23
2
2
0.02-40
0.9972
24
palmitoleic acid
556.4
170.1
27
-23
2
2
0.02-100
0.9947
25
pentadecanoic acid
544.2
320.1
27
-25
4
2
0.04-40
0.9958
26
cis-4,7,10,13,16,19docosahexaenoic acid
630.3
170.1
27
-21
10
1
0.1-100
0.9961
27
arachidonic acid
606.2
170.1
27
-24
20
1
0.2-100
0.9963
28
linoleic acid
582.3
170.1
27
-24
4
0.5
0.04-100
0.9991
29
palmitic acid
558.2
170.1
27
-25
4
2
0.04-100
0.9981
30
oleic acid
584.3
170.1
27
-27
2
1
0.02-40
0.9962
31
stearic acid
586.3
170.1
31
-26
2
0.5
0.02-40
0.9945
32
cis-11-eicosenoic acid
612.3
320.1
31
-27
2
0.5
0.02-40
0.9951
33
nonadecanoic acid
600.5
320.1
31
-26
2
1
0.02-40
0.9975
34
erucic acid
640.5
320.1
34
-27
2
1
0.02-40
0.9941
35
heneicosanoic acid
628.5
320.1
34
-25
4
1
0.04-100
0.9962
36
behenic acid
642.5
320.1
45
-29
4
2
0.04-100
0.9958
37
nervonic acid
668.6
320.1
45
-29
4
1
0.04-40
0.9986
38
lignoceric acid
670.5
320.1
45
-30
4
1
0.04-40
0.9947
a
b
LLOQs of Dns-PP labeled FFAs detected in MRM (+) mode. LLOQs of intact FFAs detected in SIM (-) mode.
metabolomics.40,41 Lipid metabolism was interrupted during both pathologies. Here, the FFAs-targeted quantification method was utilized for reanalysis of changed levels. A higher coverage of FFAs, ranging from C2 to C24, were determined in the MRM mode, compared with previously reported methods.23,35,37,38 Consequently, thirty FFAs in serum samples from both models were quantified successfully (Table S6).
Method Application. The validated method was applied to determine the levels of FFAs in serum samples from two types of rat models, namely cisplatin-induced nephrotoxicity and irinotecan-induced gastrointestinal toxicity. Both cisplatin and irinotecan are effective chemotherapeutic agents with severe side effects. In our previous studies, we have screened out and identified potential biomarkers related to toxicity via untargeted
6
ACS Paragon Plus Environment
Page 7 of 9
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry
Interestingly, concentrations of tridecanoic acid, myristic acid, cis-5,8,11,14,17-eicosapentaenoic acid, linolenic acid, palmitoleic acid, pentadecanoic acid, cis-4,7,10,13,16,19docosahexaenoic acid and arachidonic acid were observed to be declined significantly during both pathologies (Figure S4A). Declined concentrations of additional five FFAs, including decanoic acid, dodecanoic acid, linoleic acid, FFA 18:1 and nonadecanoic acid, were found only in GIT model, indicating different mechanisms of two types of toxicities (Figure S4B). FFA 18:1, here, represents all C=C location isomers of FFA 18:1 in serum samples. Additionally, we compared the potential biomarkers related to irinotecan-induced gastrointestinal toxicity screened out by untargeted metabolomics41 with significantly changed FFAs determined using our targeted method. Untargeted metabolomics is normally preformed to screen out potential biomarkers. This “one size fits all” protocol, however, typically provides semi-quantitative data (fold changes), affecting the subsequent interpretation inevitably.50-52 By contrast, targeted metabolomics is carried out to absolutely quantify a defined group of metabolites. Recent studies revealed that results obtained from these two strategies were not consistent substantially, despite partial overlap, which indicated targeted metabolomics should be conducted to validate and confirm untargeted findings.53,54 As shown in Figure S5 and Table S6, three abundant FFAs, namely palmitoleic acid (16.05-42.13 µM), arachidonic acid (45.02-85.34 µM), linoleic acid (90.11184.03 µM), were found to be declined in CPT-11 group, corresponding to the results of untargeted analysis. Another nine FFAs were also observed to be altered largely in GIT model via FFAs-targeted determination, including relatively lowabundance decanoic acid (0.36-0.92 µM), dodecanoic acid (2.19-3.35 µM), tridecanoic acid (0.10-0.14 µM), myristic acid (4.46-17.96 µM), linolenic acid (5.21-15.60 µM), pentadecanoic acid (1.71-4.35 µM), nonadecanoic acid (0.200.43 µM) and polyunsaturated fatty acids (PUFAs), such as cis5,8,11,14,17-eicosapentaenoic acid (0.89-4.29 µM), cis4,7,10,13,16,19-docosahexaenoic acid (21.58-60.97 µM) (Figure S4). Specific sample preparation, high-performance chromatographic separation, soft ionization, selective MS detection and high sensitivity all contributed to more excellent performance of FFAs-targeted methodology in comparison with untargeted one.
metabolic profiling studies and accurate describe of global view about FFAs species.
ASSOCIATED CONTENT Supporting Information The Supporting Information is available free of charge on the ACS Publications website. Synthesis of twins derivatization reagents; chromatograms of intact FFAs detected before and after derivatization; representative product ion (MS/MS) spectra of FFAs derivatives and TD-ISs; altered FFAs determined in two toxicity models; heatmap of differential metabolites screened out by untargeted metabolomics; summary of TDISs’ stability; basic information of FFAs quantified in this study; summary of stability of analytes; intra- and inter-assay performance of this method; extraction efficiency and matrix effect from rat serum samples; concentrations of FFAs in rat models. (PDF)
AUTHOR INFORMATION Corresponding Authors *Fax: +86 025 83271021; Tel.: +86 025 83271021; E-mail:
[email protected] **Fax: +86 025 83271454; Tel.: +86 025 83271454; E-mail:
[email protected] Author Contributions ‡These authors contributed equally. All authors have given approval to the final version of the manuscript.
Notes The authors declare no competing financial interest.
ACKNOWLEDGEMENTS This study was financially supported by the NSFC (No.81573385, 81573626 and 81430082), the Program for Jiangsu province Innovative Research Team, the Program for New Century Excellent Talents in University (No. NCET-13-1036) and a project funded by the Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD).
REFERENCES (1) den Besten, G.; van Eunen, K.; Groen, A. K.; Venema, K.; Reijngoud, D. J.; Bakker, B. M. J. Lipid Res. 2013, 54, 2325-2340. (2) Nakamura, M. T.; Yudell, B. E.; Loor, J. J. Prog. Lipid Res. 2014, 53, 124-144. (3) Klausner, R.; Kleinfeld, A.; Hoover, R.; Karnovsky, M. J. J. Biol. Chem. 1980, 255, 1286-1295. (4) Van Meer, G.; Voelker, D. R.; Feigenson, G. W. Nat. Rev. Mol. Cell Biol. 2008, 9, 112-124. (5) Carracedo, A.; Cantley, L. C.; Pandolfi, P. P. Nat. Rev. Cancer 2013, 13, 227-232. (6) Nguyen, C.; Haushalter, R. W.; Lee, D. J.; Markwick, P. R.; Bruegger, J.; Caldara-Festin, G.; Finzel, K.; Jackson, D. R.; Ishikawa, F.; O’Dowd, B. Nature 2014, 505, 427-431. (7) Seton-Rogers, S. Nat. Rev. Cancer 2016, 16, 677. (8) Kim, S. R.; Jeon, S. Y.; Lee, S. M. Lipids Health Dis. 2015, 14, 133. (9) Lu, Y.; Huang, C.; Gao, L.; Xu, Y. J.; Chia, S. E.; Chen, S.; Li, N.; Yu, K.; Ling, Q.; Cheng, Q. Metabolomics 2015, 11, 1526-1538. (10) Singh, S. B.; Kang, L.; Nawrocki, A. R.; Zhou, D.; Wu, M.; Previs, S.; Miller, C.; Liu, H.; Hines, C.; Madeira, M. PLoS One 2017, 12, e0170721.
CONCLUSIONS In this paper, a high-coverage FFAs quantification method was established based on LC-MS-oriented 5-dimethylaminonaphthalene-1-sulfonyl piperazine (Dns-PP) derivatization, providing enhanced sensitivity and selectivity. Twins derivatization strategy ensured accurate quantification by providing one-to-one internal standards. Thirty-eight FFAs (C2C24) with wide concentration range were quantified successfully. The FFAs-targeted method was then applied to determine the changed levels in rat serum samples from different adverse drug reaction models with only 50 µL of samples. Noticeably, several low-abundance FFAs and another two PUFAs were also observed to be changed largely in GIT model, except for abundant FFAs screened out by untargeted metabolomics. This high-coverage targeted methodology might contribute to subsequent validation of alternations detected by
7
ACS Paragon Plus Environment
Analytical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
(11) Blaak, E. Proc. Nutr. Soc. 2003, 62, 753-760. (12) Solberg, D.; Bentsen, H.; Refsum, H.; Andreassen, O. Acta Psychiatr. Scand. 2015, 132, 293-300. (13) Hennebelle, M.; Harbeby, E.; Tremblay, S.; Chouinard-Watkins, R.; Pifferi, F.; Plourde, M.; Guesnet, P.; Cunnane, S. C. Clin. Lipidol. 2015, 10, 91-102. (14) Scheppach, W. Gut 1994, 35, S35-S38. (15) Smith, P. M.; Howitt, M. R.; Panikov, N.; Michaud, M.; Gallini, C. A.; Bohlooly-y, M.; Glickman, J. N.; Garrett, W. S. Science 2013, 341, 569-573. (16) Meijer, K.; de Vos, P.; Priebe, M. G. Curr. Opin. Clin. Nutr. Metab. Care 2010, 13, 715-721. (17) Tremaroli, V.; Bäckhed, F. Nature 2012, 489, 242-249. (18) Nicholson, J. K.; Holmes, E.; Kinross, J.; Burcelin, R.; Gibson, G.; Jia, W.; Pettersson, S. Science 2012, 336, 1262-1267. (19) Dickson, I. Nat. Rev. Gastroenterol. Hepatol. 2017, 14, 67. (20) Xu, F.; Zou, L.; Ong, C. N. TrAC, Trends Anal. Chem. 2010, 29, 269-280. (21) Zheng, X.; Qiu, Y.; Zhong, W.; Baxter, S.; Su, M.; Li, Q.; Xie, G.; Ore, B. M.; Qiao, S.; Spencer, M. D. Metabolomics 2013, 9, 818827. (22) Husek, P. J. Chromatogr., Biomed. Appl. 1998, 717, 57−91. (23) Hellmuth, C.; Weber, M.; Koletzko, B.; Peissner, W. Anal. Chem. 2012, 84, 1483-1490. (24) Zehethofer, N.; Pinto, D. M.; Volmer, D. A. Rapid Commun. Mass Spectrom. 2008, 22, 2125-2133. (25) Jemal, M.; Ouyang, Z.; Teitz, D. S. Rapid Commun. Mass Spectrom. 1998, 12, 429-434. (26) Sajiki, J.; Yonekubo, J. Anal. Chim. Acta 2002, 465, 417-426. (27) Thomas, M. C.; Dunn, S. R.; Altvater, J.; Dove, S. G.; Nette, G. W. Anal. Chem. 2012, 84, 5976-5983. (28) Schiesel, S.; Lämmerhofer, M.; Lindner, W. Anal. Bioanal. Chem. 2010, 397, 147-160. (29) Semeraro, M.; Rizzo, C.; Boenzi, S.; Cappa, M.; Bertini, E.; Antonetti, G.; Dionisi-Vici, C. Clin. Chim. Acta 2016, 458, 159-164. (30) Yang, K.; Zhao, Z.; Gross, R. W.; Han, X. Anal. Chem. 2011, 83, 4243-4250. (31) Guo, K.; Li, L. Anal. Chem. 2009, 81, 3919-3932. (32) Xu, F.; Zou, L.; Liu, Y.; Zhang, Z.; Ong, C. N. Mass Spectrom. Rev. 2011, 30, 1143-1172. (33) Niwa, M.; Watanabe, M.; Watanabe, N.; Bioanalysis, 2015, 7, 2443-2449. (34) Zhao, S.; Dawe, M.; Guo, K.; Li, L. Anal. Chem. 2017, 89, 67586765. (35) Yang, W. C.; Adamec, J.; Regnier, F. E. Anal. Chem. 2007, 79, 5150-5157. (36) Cai, T.; Ting, H.; Zhang, X.; Zhou, J.; Zhang, J. Analyst 2014, 139, 6154-6159. (37) Leng, J.; Wang, H.; Zhang, L.; Zhang, J.; Wang, H.; Guo, Y. Anal. Chim. Acta 2013, 758, 114-121. (38) Han, J.; Lin, K.; Sequeira, C.; Borchers, C. H. Anal. Chim. Acta 2015, 854, 86-94. (39) Narayana, V. K.; Tomatis, V. M.; Wang, T.; Kvaskoff, D.; Meunier, F. A. Chem. Biol. 2015, 22, 1552-1561. (40) Zhang, P.; Chen, J.; Wang, Y.; Huang, Y.; Tian, Y.; Zhang, Z.; Xu, F. Chem. Res. Toxicol. 2016, 29, 776-783. (41) Wang, J.; Fan, H.; Wang, Y.; Wang, X.; Zhang, P.; Chen, J.; Tian, Y.; Zhang, W.; Xu, F.; Zhang, Z. RSC Adv. 2015, 5, 26420-26429. (42) Zhu, Q.; Zhang, Z.; Liu, P.; Zheng, S.; Peng, K.; Deng, Q.; Zheng, F.; Yuan, B.; Feng, Y. J. Chromatogr. A 2016, 1460, 100-109. (43) FDA. Bioanalytical Method Validation. Available at the following: http://www.fda.gov. FDA Guidance for Industry Bioanalytical Method Validation. (44) Carter, P. Clin. Chem. 1978, 24, 362-364. (45) Chen, R. F. J. Biol. Chem. 1967, 242, 173-181. (46) Bansal, S.; DeStefano, A. AAPS J. 2007, 9, 109-114. (47) Xu, Y.; Zhang, L.; Lu, H.; Yang, P. Anal. Chem. 2008, 80, 83248328. (48) Avery, M. J. Rapid Commun. Mass Spectrom. 2003, 17, 197-201.
Page 8 of 9
(49) Fu, I.; Woolf, E.; Matuszewski, B. J. Pharm. Biomed. Anal. 1998, 18, 347-357. (50) Cajka, T.; Fiehn, O. Anal. Chem. 2015, 88, 524-545. (51) Nicholson, J. K.; Lindon, J. C.; Holmes, E. Xenobiotica 1999, 29, 1181-1189. (52) Nicholson, J. K.; Lindon, J. C. Nature 2008, 455, 1054-1056. (53) Griffiths, W. J.; Koal, T.; Wang, Y.; Kohl, M.; Enot, D. P.; Deigner, H. P. Angew. Chem., Int. Ed. 2010, 49, 5426-5445. (54) Klepacki, J.; Klawitter, J.; Klawitter, J.; Karimpour-fard, A.; Thurman, J.; Ingle, G.; Patel, D.; Christians, U. Clin. Biochem. 2016, 49, 955-961.
8
ACS Paragon Plus Environment
Page 9 of 9
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry
for TOC only
9
ACS Paragon Plus Environment
