|
Advertisement | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Journal of Lipid Research, Vol. 48, 252-259, January 2007
Rapid measurement of deuterium-labeled long-chain fatty acids in plasma by HPLC-ESI-MS
Merck Frosst Canada & Co., Medicinal Chemistry Department, Kirkland, Quebec, Canada H9H 3L1 Published, JLR Papers in Press, October 4, 2006.
1 To whom correspondence should be addressed. e-mail: sebastien_gagne{at}merck.com
Imbalanced fatty acid metabolism contributes significantly to the increased incidence of metabolic disorders. Isotope-labeled fatty acids (2H, 13C) provide efficient means to trace fatty acid metabolism in vivo. This study reports a new and rapid method for the quantification of deuterium-labeled fatty acids in plasma by HPLC-MS. The sample preparation protocol developed required only hydrolysis, neutralization, and quenching steps followed by high-performance liquid chromatography-electrospray ionization-mass spectrometry analysis in negative ion mode using single ion monitoring. Deuterium-labeled stearic acid (d7-C18:0) was synthesized to reduce matrix interference observed with d5 analog, which improved the limit of detection (LOD) significantly, depending on the products analyzed. Linearity > 0.999 between the LOD (100 nM) and 30 µM, accuracy > 90%, precision > 88%, and adequate recovery in the dynamic range were obtained for d7-C18:0 and d7-oleic acid (C18:1). Upon oral dosing of d7-C18:0 in rats, the parent compound and its desaturation and ß-oxidation products, d7-C18:1 and d7-C16:0, were circulating with a maximal concentration ranging from 0.6 to 2.2 µM, with significant levels of d7-fatty acids detected for up to 72 h.
Supplementary key words deuterium-labeled stearic acid deuterium-labeled oleic acid deuterium-labeled palmitic acid high-performance liquid chromatography-electrospray ionization-mass spectrometry electrospray ionization negative mode quantification of plasma levels Abbreviations: C16:0, palmitic acid; C16:1, palmitoleic acid; C18:0, stearic acid; C18:1, oleic acid; HPLC-ESI-MS, high-performance liquid chromatography-electrospray ionization-mass spectrometry; LOD, limit of detection; MRM, multiple reaction monitoring; SIM, single ion monitoring
Fatty acid biosynthesis and degradation are of increasing interest in a context in which obesity affects a growing proportion of the population of the United States and is associated with higher risks of medical problems (1, 2). Analysis of natural fatty acids such as palmitic (C16:0), palmitoleic (C16:1), stearic (C18:0), and oleic (C18:1) acids in biological samples is commonly performed by gas chromatography coupled to flame ionization detection or electron impact/chemical ionization mass spectrometry (3). However, the presence of an important in vivo endogenous pool disables short-term in vivo studies of fatty acid biosynthesis and metabolism. An alternative approach commonly used consists of dosing isotopically labeled fatty acids (2H or 13C) to generate a mass shift that allows selective detection of the tracer using either GC-MS or GC coupled to isotope ratio mass spectrometry, with minimal interference attributable to natural fatty acids (48). Tracers such as [13C]C18:0 and deuterium-labeled (d4)-linoleic acid were dosed orally, and the disappearance of parent compound and the formation of labeled metabolites were reported in rat, cat, and human over a short time period (46, 8). The use of GC for fatty acid analysis is generally associated with time-consuming sample preparation resulting from the liquid-liquid extraction and chemical derivatization steps (46, 8, 9). HPLC and HPLC-MS were previously considered as alternative approaches (8, 1020), but all methodologies reported still involved long derivation and/or labor-intensive liquid-liquid extraction steps. There are increasing needs in the pharmaceutical industry to develop a fast and simple approach for the analysis of fatty acids. The objective of this work was to find a novel and rapid analytical method using high-performance liquid chromatography-electrospray ionization-mass spectrometry (HPLC-ESI-MS) for the analysis of natural and deuterium-labeled fatty acids in plasma involving a simple sample preparation procedure that would significantly increase the sample preparation throughput without compromising repeatability, accuracy, and robustness.
Chemicals C16:0, C16:1, C18:0, C18:1, and tricosanoic acid were obtained from Sigma-Aldrich (Milwaukee, WI) and were used without any further purification (99% purity). d5-C18:0 was purchased from C/D/N Isotopes (Pointe-Claire, Canada), whereas d7-C18:0 and d7-C18:1 were synthesized in-house (S. N. Crane, K. Bateman, S. Gagne, and J-F. Levesque, unpublished data). Methanol and acetone (optima grade) were obtained from Fisher Scientific, and NH4OH was from American Chemical, Ltd. (Montreal, Canada). Water was deionized using Milli-Q Plus from Millipore. NaOH (10 N) was obtained from J. T. Baker (Phillipsburg), and formic acid (99%) was purchased from Acros Organics.
Instruments
Analytical conditions
Rat dosage and plasma collection Male Sprague-Dawley rats and C57BL mice were supplied by Charles River Canada (St. Constant, Canada), squirrel monkeys by PAHO and Osage Research Primates LLC, and beagle dogs by Marshall Farms. Blood samples were collected from healthy animals in heparinized tubes, and plasma was prepared by centrifugation at 11,000 g for 3 min at 4°C. For the rat pharmacokinetic study of d7-C18:0, Sprague-Dawley rats were fasted overnight, and 0 h blood sample was obtained by tail bleeding. d7-C18:0 was then administered orally at 10 or 30 mg/kg in 1% Methocel at 5 ml/kg, and plasma samples were collected at 1, 2, 4, 6, 24, 48, and 72 h after dose. Plasma samples were frozen right after collection. The rats were allowed to receive food 4 h after dosing, and water was provided at all times during the study.
Sample and standard curve preparation The standard curve samples were prepared using the same protocol except that 2 µl of a stock solution of d7-C18:0 and d7-C18:1 was added to the initial 50 µl of plasma for each desired concentration between the limit of detection (LOD) and 30 µM.
Analytical performance evaluation
Mass spectrometric approach Typical methods reported in the literature for fatty acid analysis necessitate long sample preparation, including multiple liquid-liquid extractions, hydrolysis, and chemical derivatization steps, followed by GC-MS analysis (46). This paper describes a rapid and new method developed for the measurement of natural and deuterium-labeled fatty acids in plasma by HPLC-ESI-MS. The proposed approach consists of the hydrolysis of fatty acids incorporated into the triglyceride and other fat pools, followed by the acidification of the hydrolyzate (46), and a rapid and simple protein precipitation step using organic solvent. Then, samples are centrifuged and supernatants are analyzed by HPLC-ESI-MS. This protocol significantly reduces the sample preparation time required to analyze total fatty acids by avoiding numerous evaporation steps involved in derivation and liquid-liquid extraction procedures (46). As a first attempt, fatty acids were analyzed using multiple reaction monitoring (MRM) as an MS/MS approach to minimize interference from the plasma matrix, using C18:0 as a model compound for the optimization (21). As shown in Fig. 1 , good ionization of C18:0 was obtained in negative electrospray ion mode, but its fragmentation did not allow an optimal sensitivity compared with SIM detection. The maximal fragment represented 15% of the parent ion, as already reported (19) with a [M-18µ] ion observed, which can be correlated with a loss of water from the carboxylic acid group. This poor fragmentation efficiency does not allow good sensitivity. The SIM mode was found to give better sensitivity compared with the MRM mode, despite the fact that more analytical interference may be expected in SIM mode than in MRM mode.
The simple sample preparation procedure combined with the HPLC-ESI-MS method in SIM mode allowed good detection of the natural fatty acids C14:0, C14:1, C16:0, C16:1, C18:0, and C18:1, as shown in Fig. 2 . As expected, plasma concentrations of these endogenous fatty acids were estimated to be in the high micromolar or low millimolar range in rat, mouse, dog, and monkey plasma (data not shown) (6, 22). Such high levels of natural fatty acids justify the need to use isotopically labeled fatty acids to monitor in vivo biosynthesis and metabolism of fatty acids over a short period with minimal interference from endogenous material. 2H and 13C tracers have been widely used for such studies, in which 2H- and 13C-labeled essential fatty acid biotransformations were followed in rat, cat, and human (46, 8). C18:0 was selected as a prototypical fatty acid to evaluate the potential use of this new analytical methodology for the detection of isotopically labeled tracers.
Commercially available sources for labeled C18:0 materials are d2, d3, d4, d5, d35, 13C1, and 13C18. In theory, the presence of a 2H or 13C label on fatty acids should minimize the contamination of the SIM trace by endogenous fatty acid. However, the mass contribution from the added isotope must be superior to the natural isotopic effect of the endogenous form, because natural and isotopically labeled materials coelute by chromatography (46). In addition, the presence of isotopes on C18:0 were minimized to avoid potential isotopic effects on the activity of enzymes involved in fatty acid metabolism (6, 23). Because at least five deuterium atoms were needed to minimize signal contamination attributable to the isotopic composition of endogenous fatty acids, the commercially available d5-C18:0 was first considered as a potential tracer for in vivo studies in rats. To evaluate the presence of interference from the plasma matrix, blank plasma from Sprague-Dawley rat, mouse, squirrel monkey, and beagle dog was extracted and analyzed with SIM descriptors corresponding to putative C16:0, C16:1, C18:0, and C18:1 labeled with five deuterium atoms. SIM traces obtained in blank rat plasma are presented in Fig. 3 . These traces are representative of results obtained across species. In summary, interference was detected in the SIM trace of a d5-C18:1 and more importantly for a d5-C16:0, whereas clean traces were observed for d5-C16:1 and d5-C18:0 at the chromatographic retention times expected based on the endogenous form. This interference issue can be addressed by the use of SIM traces calculated for d7-C16:0, d7-C16:1, d7-C18:0, and d7-C18:1. For these traces, the chromatogram is exempt from interference at the expected chromatographic retention times of all compounds in both rodent and nonrodent species. The use of d7-labeled fatty acids would allow easy detection of the fatty acids of interest, which would be difficult using a commercially available d5-labeled fatty acid because of background noise. In dog plasma, for example, the detection of d5-C18:1 was compromised as a result of high levels of interference, whereas the d7-C18:1 trace was cleaner (Fig. 4 ), proving once more the advantage of d7-labeled fatty acids as tracer.
For the purposes of this study, d7-C18:0 and d7-C18:1 were synthesized in-house and fully characterized by NMR (S. N. Crane, K. Bateman, S. Gagne, and J-F. Levesque, unpublished data) and by mass accuracy using an ultra-performance liquid chromatography-quadrupole-time of flight apparatus. All exact masses measured were in good agreement with the theoretical expected mass within 10 ppm.
Analytical performance
Table 2
summarizes all of the analytical parameters evaluated. The LOD for both d7-C18:0 and d7-C18:1 was 100 nM, and minimal interfering background was detected, compared with the d5-fatty acid analogs (Figs. 35
). The linearity was evaluated between the LOD and a maximal concentration fixed at 30 µM, and for both standards, R2
In addition, the recovery was measured for d7-C18:0 and d7-C18:1, and the results obtained are presented in Table 3 . Over the entire dynamic range reported, the recovery allowed an accurate measurement for each d7-fatty acid ± 15% at concentrations of 1 µM and within an acceptable range at 0.3 µM for d7-C18:0 and d7-C18:1. Similar analytical performances were presented for fatty acid analysis by LC-MS in other matrices than plasma (18, 19). This validated analytical methodology for the extraction and analysis of d7-fatty acids was then used to analyze plasma samples from an in vivo experiment.
In vivo evaluation in rat To evaluate the potential of this new d7 tracer to study fatty acid metabolism, d7-C18:0 was administered orally at two doses and the plasma samples collected were treated using the optimized analytical method. Typical transformations expected (24) for d7-C18:0 should include CoA formation, unsaturation, and ß-oxidation of its alkyl chain as well as esterification to triglycerides or incorporation into fat pools. However, the plasma hydrolysis step in the sample preparation disables the detection of d7-fatty acid esters. Rat plasma samples collected were analyzed, and a typical HPLC-ESI-MS chromatogram is shown in Fig. 6 . Excellent signal-to-noise ratios were observed for d7-C18:0 and d7-C18:1 at all time points, whereas d7-C16:0 was detected only up to 24 h. For both dosages, only the d7-C16:1 stayed under the LOD at all time points. Figure 7 depicts the pharmacokinetic profile of d7-fatty acids upon a 30 mg/kg dose of d7-C18:0. Data obtained for the 10 mg/kg dose were similar, except that lower concentrations were observed for each d7-fatty acid (data not shown). As expected, slow rates of elimination were observed for the d7-fatty acids analyzed. After 72 h, important levels of d7-C18:0 and d7-C18:1 was still detected, and elimination half-lives were estimated to be 17 and 45 h, respectively. These results clearly demonstrate that unsaturation and ß-oxidation are the main biotransformations involved in d7-C18:0 metabolism in vivo in rats, as reported in the literature for the de novo fatty acid biosynthesis (24, 25). Interestingly, d7-C16:0 has a shorter elimination half-life than d7-C18:0 and d7-C18:1 (7 h), as shown previously in rat (26).
In summary, an improved methodology was developed for the analysis of natural fatty acids (C14:0, C16:0, and C18:0 and their monounsaturated versions) and deuterium-labeled analogs in plasma samples, involving only hydrolysis, neutralization, and quenching steps followed by HPLC-ESI-MS analysis in negative ion mode using SIM. This method is rapid and robust and is useful for the detection of fatty acids from 100 nM to the high micromolar range. A new d7-fatty acid tracer, d7-C18:0, was synthesized to reduce matrix interference observed with its d5 analog, which significantly improved the LOD depending on the products analyzed. Upon the oral dosing of d7-C18:0 in rats, the parent compound and its desaturation and ß-oxidation products, d7-C18:1 and d7-C16:0, were circulating with a maximal concentration ranging from 0.6 to 2.2 µM, with significant levels of d7-fatty acids detected for up to 72 h. This methodology could be extended to other d7-fatty acids, as needed, based on results obtained with endogenous fatty acids. In the course of these studies, only d7-C18:0 and d7-C18:1 were synthesized, because monodesaturation was expected to be the main biotransformation involved in d7-C18:0 metabolism, and the synthesis of labeled material can be time- and resource-consuming.
The authors thank the members of the Department of Comparative Medicine at Merck Frosst Canada for their assistance with the animal work. The instrumentation used in this project was supported by Merck Frosst Canada. Manuscript received September 11, 2006 and in revised form October 3, 2006.
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Advertisement | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||