Anal. Chem. 2001, 73, 5358-5364
Articles
Determination of Folates in Human Plasma Using Hydrophilic Interaction Chromatography-Tandem Mass Spectrometry Spiros D. Garbis,† Alida Melse-Boonstra,‡ Clive E. West,‡,§ and Richard B. van Breemen*,†
Department of Medicinal Chemistry and Pharmacognosy, College of Pharmacy, University of Illinois at Chicago, 833 South Wood Street, Chicago, Illinois 60612, Department of Human Nutrition and Epidemiology, Wageningen University, P.O.B. 8129, 6700 EV Wageningen, The Netherlands, and Department of Gastroenterology, University Medical Centre, Nijmegen, The Netherlands
Folic acid is an essential nutrient, and folate deficiency is associated with a variety of disorders including neural tube defects (during pregnancy) and heart disease. A fast, sensitive, and robust HPLC-tandem mass spectrometry (LC-MS-MS) method was developed for the quantification of free folic acid, tetrahydrofolate, 5′-methyltetrahydrofolate, and 5′-formyltetrahydrofolate in human plasma. Sample preparation required only acetonitrile precipitation of proteins followed by filtration instead of solid-phase extraction or solvent-solvent extraction as in other methods. The rapid and streamlined sample handling procedure minimized degradation of the highly unstable folate species. Hydrophilic interaction chromatography was used for additional sample cleanup on-line, and baseline separation and detection of all four folate species was achieved in less than 30 min. The folate species were detected using negative ion electrospray-tandem mass spectrometry with multiple reaction monitoring of the diagnostic fragment ions of each deprotonated molecule. The predominately organic (hydrophobic) solvent system combined with the microbore flow rate (50 µL/min) used for the chromatography resulted in enhanced electrospray signal response compared to reversed-phase HPLC using a wider bore column. The recovery of all folate species (from spiked plasma) was >97% over a concentration range from 300 pg/L to 12 mg/L with intraday precision (RSD, n ) 5) of 3.7-6.5%. Stability studies were carried out for spiked samples in order to define storage and handling conditions. The folic acid limit of quantification (LOQ) in human plasma was 80 pmol/L ( 10%, and the limit of detection (LOD) was 37.5 pmol/L. The LOQ and LOD for tetrahydrofolate, 5′-methyltetrahydrofolate, and 5′-formyltetrahydrofolate were 1250, 400, and 360 pmol/L of plasma and 425, 165, and 140 pmol/L of plasma, respectively. Folate is an essential nutrient in the human diet exhibiting hematopoietic properties. Recent studies indicate that levels of * Corresponding author: (tel) (312) 996-9353; (fax) (312) 996-7107; (e-mail)
[email protected]. † University of Illinois at Chicago. ‡ Wageningen University. § University Medical Centre.
5358 Analytical Chemistry, Vol. 73, No. 22, November 15, 2001
plasma folate and its methylated and hydrogenated derivatives reflect dietary folate intake.1 Therefore, proper dietary intake should result in adequate folate concentrations in the plasma. An early indicator of folate deficiency is a decrease in plasma folate concentrations, which can subsequently produce deficient erythrocyte folate concentrations and induce increases in homocysteine levels.2-7 Elevated homocysteine is associated with increased risk of neural tube defects,8-11 coronary heart disease,12-14 and colon cancer.15 In particular, folate sufficiency during pregnancy is necessary for the prevention of neural tube defects such as spina bifida.16-18 Like S-adenosyl-L-methionine, the tetrahydrofolate (1) Tucker, K. L.; Selhub, J.; Wilson, P. W.; Rosenberg, I. H. J. Nutr. 1996, 126, 3025-3031. (2) Pancharunit, i N.; Lewis, C. A.; Sauberlich, H. E.; Perkins, L. L.; Go, R. C. P.; Alvarez, J. O.; Macaluso, M.; Acton, R. T.; Copeland, R. B.; Cousins, A. L.; Gore, T. B.; Cornwell, P. E.; Roseman, J. M. Am. J. Clin. Nutr. 1994, 59, 940-948. (3) Selhub, J.; Jacques, P. F.; Wilson, P. W. F.; Rush, D.; Rosenburg, I. H. J. Am. Med. Assoc. 1993, 270, 2693-2698. (4) Cuskelly, G. J.; McNulty, H.; Scott, J. M. Lancet 1996, 347, 657-659. (5) de Bree, A.; Verschuren, W. M.; Blom, H. J.; Kromhout, D. Am. J. Clin. Nutr. 2001, 73, 1023-1033. (6) Wald, D. S.; Bishop, L.; Wald, N. J.; Law, M.; Hennessy, E.; Weir, D.; McPartlin, J.; Scott, J. Arch. Intern. Med. 2001, 161), 695-700. (7) Henning, B. F.; Tepel, M.; Riezler, R.; Naurath, H. J. Gerontology 2000, 47, 30-35. (8) Graham, I. M.; Daly, L. E.; Refsum, H. M.; Robinson, K.; Brattstro ¨m, L. E.; Ueland, P. M.; Palma-Reis, R. J.; Boers, G. H. J.; Sheahan, R. G.; Israelsson, B.; Uiterwaal, C. S.; Meleady, R.; McMaster, D.; Verhoef, P.; Witteman, J.; Rubba, P.; Bellet, H.; Wautrecht, J. C.; de Valk, H. W.; Sales Lu´is A. C.; Parrot-Pouland F. M.; Tan, K. S.; Higgins, I.; Garcon, D.; Medrano, M. J.; Candito, M.; Evans, A. E.; Andria, G. J. Am. Med. Assoc. 1997, 277, 17751781. (9) Hernandez-Diaz, S.; Werler, M. M.; Walker, A. M.; Mitchell, A. A. Am. J. Epidemiol. 2001, 153, 961-968. (10) van der Put, N. M.; van Straaten, H. W.; Trijbels, F. J.; Blom, H. J. Exp. Biol. Med. (Maywood) 2001, 226 (4), 243-270. (11) Richter, B.; Stegmann, K.; Rapper, B.; Baddeker, I.; Ngo, E. T.; Koch, M. C. J. Hum. Genet. 2001, 46 (3), 105-109. (12) Brouwer, I. A.; van Dusseldorp, M.; West, C. E., Meyboom, S.; Thomas, C. M. G.; Duran, M.; van het Hof, K. H.; Eskes, T. K. A. B.; Hautvast, J. G. A. J.; Steegers-Theunissen, R. P. M. J. Nutr. 1999, 129, 1135-1139. (13) Nygård, O.; Vollset, S. E.; Refsum, H.; Stensvold, I.; Tverdal, A.; Nordrehaug, J. E.; Ueland, M.; Kvale, G. J. Am. Med. Assoc. 1995, 274, 1526-1533. (14) Rimm, E. B.; Willet, W. C.; Hu, F. B.; Sampson, L.; Colditz, G. A.; Manson, J. E.; Hennekens, C.; Stampfer, M. J. J. Am. Med. Assoc. 1998, 279, 359364. (15) Giovanucci, E.; Rimm, E. B.; Ascherio, A.; Stampfer, M. J.; Colditz, G. A.; Willet, W. C. J. Natl. Cancer Inst. 1995, 87, 265-273. 10.1021/ac010741y CCC: $20.00
© 2001 American Chemical Society Published on Web 10/19/2001
derivative of folic acid functions as a cofactor for enzymes that transfer methyl groups. During this transfer process, a single carbon unit is attached to the N5 atom of tetrahydrofolate forming 5′-methyltetrahydrofolate. Analogous reactions lead to the formation of the metabolically important 5′-formyltetrahydrofolate. Classically, the determination of total folate content in various foods is carried out using a microbiological assay.19,20 In addition, various high-performance liquid chromatography (HPLC) methods have been developed that utilize reversed-phase, affinity, ion exchange, and ion pair chromatography with UV, fluorescence, or electrochemical detection.21-40 However, these chromatographic methods lack specificity and sensitivity, require elaborate sample preparation, and are not ideal for electrospray mass spectrometric detection. Gas chromatography/mass spectrometry (GC/MS) has been used for the analysis of folates and found to have greater sensitivity than previous chromatographic methods.41-44 However, GC/MS requires complex sample preparation including chemical derivatization that introduces new sources of experimental error (such as incomplete derivatization) and increases the probability of folate degradation. As an alternative approach, a HPLC-mass spectrometric (LC-MS) method has been developed for the determination of folates in vitamin tablets, cereal foodstuffs, and other fortified foods.45,46 However, the selectivity of this LC-MS (16) MRC Vitamin Study Research Group. Lancet 1991, 338, 131-137. (17) Czeizel, A. E.; Duda´s, I. N. Engl. J. Med. 1992, 327, 1832-1835. (18) Evans, M. I.; Wapner, R. J.; O’Brien, J. E.; Dvorin, E.; Huang, X.; Harrison, H. Obstet. Gynecol. 2001, 97 (4 Suppl. 1), S42. (19) Horne, D. W.; Patterson, D. Clin. Chem. 1988, 34, 2357-2359. (20) Rader, J. I.; Weaver, C. M.; Angyal, G. Food Chem. 1998, 62, 451-465. (21) Seyoum, E.; Selhub, J. J. Nutr. Biochem. 1993, 4, 488-494. (22) Iwase, H. J. Chromatogr. 1992, 609, 399-401. (23) Schleyer, E.; Reinhardt, J.; Unterhalt, M.; Hiddemann, W. J. Chromatogr., B 1995, 669, 319-330. (24) Lucock, M. D.; Daskalakis, I.; Schorah, C. J.; Levene, M. I.; Hartley, R. Biochem. Mol. Med. 1996, 58, 93-112. (25) Kelly, P.; McPartlin, J.; Scott, J. Anal. Biochem. 1996, 238, 179-183. (26) Etienne, M.-C.; Speziale, N.; Milano, G. Clin. Chem. 1993, 39 (1), 82-86. (27) Belz, S.; Frickel, C.; Wolfrom, C.; Nau, H.; Henze, G. J. Chromatogr., B 1994, 661, 109-118. (28) Vahteristo, L.; Lehikoinen, K.; Ollilainen, V.; Varo, P. Food Chem. 1997, 59, 589-597. (29) Vahteristo, L.; Ollilainen, V.; Koivistoinen, P.; Varo, P. J. Agric. Food Chem. 1996, 44, 477-482. (30) Pfeiffer, C. M.; Rogers, L. M.; Gregory, J. F. J. Agric. Food Chem. 1997, 45, 407-413. (31) Shimoda, M.; Shin, H.-C.; Kokue, E. J. Vet. Med. Sci. 1994, 56, 701-705. (32) Lim, H.-S.; Mackey, A. D.; Tamura, T.; Wong, S. C.; Picciano, M. F. Food Chem. 1999, 63, 401-407. (33) Tamura, T. J. Nutr. Biochem. 1998, 9 (5), 285-293. (34) Wigertz, K.; Ja¨gerstad, M. Food Chem. 1995, 54 (4), 429-436. (35) Ichinose, N.; Tsuneyoshi, T.; Kato, M.; Suzuki, T.; Ikeda, S. Fresenius J. Anal. Chem. 1993, 346, 841-846. (36) Bagley, P. J.; Selhub, J. Methods Enzymol. 1997, 281, 16-25. (37) Akhtar, M. J.; Khan, M. A.; Ahmad, I. J. Pharm. Biomed. Anal. 1997, 16, 95-99. (38) Mu ¨ ller, H. Z. Lebensm.-Unters. Forsch. 1993, 197, 573-577. (39) Papadoyannis, I. N.; Tsioni, G. K.; Samanidou, V. F. J. Liq. Chromatogr. Relat. Technol. 1997, 20, 3203-3231. (40) Santhosh-Kumar, C. R.; Deutsch, J. C.; Hassell, K. L.; Kolhouse, N. M.; Kolhouse, J. F. Anal. Biochem. 1995, 225, 1-9. (41) Santhosh-Kumar, C. R.; Kolhouse, N. M. Methods Enzymol. 1997, 281, 2638. (42) Toth, J. P.; Gregory, J. F. Biomed. Environ. Mass Spectrom. 1988, 17, 7379. (43) Pfeiffer, C. M.; Rogers, L. M.; Bailey, L. B.; Toth, J. P.; Cerda, J. J. Am. J. Clin. Nutr. 1997, 66, 1388-1397. (44) Gregory, J. F.; Bhandari, S. D.; Bailey, L. B.; Toth, J. P.; Cerda, J. J. Am. J. Clin. Nutr. 1992, 55, 1147-1153. (45) Stokes, P.; Webb, K. J. Chromatogr., A 1999, 864, 59-67.
method was too low to determine specific folates in plasma. Also, this approach utilized heat pretreatment of the sample followed by C18 solid-phase extraction, which might contribute to sample degradation. Therefore, the development of a more selective, fast, robust, and more sensitive method to analyze folic acid and its derivatives in human plasma is essential for the determination of the bioavailability and biotransformation of food folate and dietary supplements. In this paper, the development of a LC-MS-MS method is described for the simultaneous measurement of folic acid, tetrahydrofolate, 5′-methyltetrahydrofolate, and 5-formyltetrahydrofolate in human plasma. This method is sensitive enough to measure endogenous levels of 5′-methyltetrahydrofolate levels in human plasma (2.2-25.6 µg/L).26 MATERIALS AND METHODS Reagents. HPLC-grade methanol, acetonitrile, ammonium acetate, aqueous ammonia, water, and triethylamine were purchased from Fisher Scientific (Pittsburgh, PA). The antioxidants ascorbic acid, citric acid, and 2-mercaptoethanol; the folates folic acid, folinic acid (calcium salt), tetrahydrofolate, and 5′-methyltetrahydrofolate (disodium salt); and the internal standard methotrexate and human plasma were obtained from Sigma Chemical (St. Louis, MO). Standards. The folate standards (corrected for purity) were separately weighed into 35-mL amber vials. Each standard (10.0 mg) was dissolved in 10 mL of methanol containing 0.1 mg/L each of ascorbic acid, citric acid, and 2-mercaptoethanol (to inhibit oxidation). Aliquots of 5.0 mL from each of the four standard solutions were combined into a 50-mL amber volumetric flask, which was diluted to volume with acetonitrile/0.1% ammonium acetate (3:1, v/v) to obtain a stock solution containing 100 mg/L of each compound. Starting with the 100 mg/L stock solution, additional dilutions with the acetonitrile ammonium acetate solution were carried out to obtain spiking solutions ranging from 0.6 to 50 mg/L each containing 50 mg/L methotrexate internal standard. Additional dilutions with acetonitrile/0.1% ammonium acetate (3:1, v/v) were performed to obtain composite calibration solutions ranging from 3 µg/L to 10 mg/L each containing 2.5 mg/L methotrexate as internal standard. Preparation of Spiked Plasma Samples. Plasma samples (1.9 mL) were spiked with 100-µL aliquots of composite folate standard solutions so that folate concentrations ranged from 3 µg/L to 10 mg/L with each containing 2.5 mg/L methotrexate as internal standard. A reagent blank containing no analytes or plasma and a plasma control were analyzed with each sample set. Analyses were typically carried out within 24 h of sample preparation. However, the standard and sample solutions remained stable for up to 3 days when stored at -20 °C under a blanket of argon. Subdued lighting was also used during sample preparation and handling to minimize analyte degradation. Sample Pretreatment. To each spiked plasma sample, 2 mL of acetonitrile was added to precipitate proteins. The resulting solution was vortexed and centrifuged for 5 min at ∼10500g. The supernatant was removed and passed through a 0.45-µm PTFE syringe filter. A 1-mL aliquot of acetonitrile was used to rinse the precipitate and then was passed through the syringe filter to ensure full recovery of the folates. The resulting filtered extract (46) Pawlosky, R. J.; Flanagan, V. P. J. Agric. Food Chem. 2001, 49, 1282-1286.
Analytical Chemistry, Vol. 73, No. 22, November 15, 2001
5359
Table 1. Evaluation of HPLC Columns and Mobile Phases column type Waters XterraMS, hybrid 3.5 µm, 2.1 × 150 mm, 125-Å pore size
electrospray ionization polarity
solvent system
negative ion
a. 20-25 mM ammonium acetate pH 7/acetonitrile (90:10)
positive ion
b. 20-25 mM ammonium formate pH 3.5/acetonitrile (90:10)
negative ion
a. 20-25 mM ammonium acetate pH 7/acetonitrile (90:10)
positive ion
b. 20-25 mM ammonium formate pH 3.5/acetonitrile (90:10)
positive ion
a. 20-25 mM ammonium acetate pH 7.0/acetonitrile (90:10)
positive ion
b. 20-25 mM ammonium formate pH 3.5/acetonitrile (90:10)
negative ion
a. 20-25 mM ammonium acetate pH 7/acetonitrile (90:10)
positive ion
b. 20-25 mM ammonium formate pH 3.5/acetonitrile (90:10
5. J.T. Baker (Phillipsburg, NJ) cyanopropyl, 5 µm, 4.6 × 150 mm, 100-Å pore, 0.2 mL/min split
positive or negative ion
methanol/acetonitrile (60:40) with or without 50-200 mM ammonium acetate or ammonium formate; or 100% toluene
6. J. T. Baker aminopropyl, 5 µm, 4.6 × 150 mm, 100-Å pore size
positive or negative ion 0.2 mL/min split
7. The Nest Group HILIC, 5 µm, 1.0 × 150 mm, 100-Å pore size
positive or negative ion
methanol/acetonitrile (60:40) with or without 50-200 mM ammonium acetate or ammonium formate; or toluene 100% isocratic acetonitrile/water (75:25) with 5 mM ammonium acetate, pH 6.9
2. Supelco (Bellefonte, PA) Discovery C18, 5-µm silica, 2.1 × 150 mm, 180-Å pore size
3. Alltech Associates (Deerfield, IL) Alltima C18, 5-µm silica, 2.1 × 150 mm, 180-Å pore size
4. EIChrom Technologies (Darien, IL) NPS C18, 1.5-µm silica, 4.6 × 53 mm, 120-Å pore size
was evaporated to dryness under vacuum. The sample residue was then reconstituted in 100 µL of acetonitrile/0.1% (w/v) ammonium acetate (3:1, v/v). Optimization of Chromatography. Seven different HPLC columns and 11 mobile phases were evaluated for the separation of the folate species under conditions compatible with electrospray ionization mass spectrometry (Table 1). The stationary phases included C18 on porous and nonporous silica, C18 on a copolymer support, aminopropyl silica, cyanopropyl silica, and polyhydroxyethyl aspartamide. The mobile phases included organic solvents and volatile aqueous buffers at various pH values and ionic strengths. For compatability with electrospray mass spectrometry, only the volatile buffers ammonium acetate (pH 5.5-8.0) and ammonium formate (pH 3.0-5.5) were investigated over the concentration range of 5-200 mM. LC-MS-MS. HPLC was carried out using a Waters (Milford, MA) Alliance 2690 binary pump equipped with an autosampler, 100-µL sample loop, and in-line membrane degasser. The tubing within the pump and injector were retrofitted with low-dispersion tubing (0.13-mm i.d.) in order to reduce extracolumn volume. Fused-silica capillary tubing (0.25-mm i.d.) was used as the transfer 5360 Analytical Chemistry, Vol. 73, No. 22, November 15, 2001
comments a. RT < 3.5min SPE required to reduce interference from plasma components coeluting species suppressed ionization b. RT 9-14 min tailing folate peaks SPE required to reduce interference from plasma components a. RT < 3.5 min SPE required to reduce interference from plasma components coeluting species suppressed ionization b. RT 7-10 min tailing folate peaks SPE required to reduce interference from plasma components a. RT < 3.5 min coeluting species suppressed ionization SPE required to reduce interference from plasma components b. RT 6-9 min tailing folate peaks SPE required to reduce interference from plasma components a. RT < 1.7 min tailing folate peaks coeluting species suppresse dionization SPE required to reduce interference from plasma components b. RT 5-8 min tailing folate peaks SPE required to reduce interference from plasma components RT > 80 min broad and tailing peaks SPE required to reduce interference from plasma components addition of ammonium acetate or ammonium formate suppressed ionization no peaks observed
baseline resolution of folates in
