Anal. Chem. 2010, 82, 3650–3658
Quantitative High-Performance Liquid Chromatography-Electrospray Ionization Tandem Mass Spectrometry Analysis of Bis-N7-Guanine DNA-DNA Cross-Links in White Blood Cells of Cancer Patients Receiving Cyclophosphamide Therapy Bhaskar Malayappan,† L’Aurelle Johnson,‡ Bei Nie,† Dolly Panchal,† Brock Matter,† Pamala Jacobson,‡ and Natalia Tretyakova*,† Departments of Medicinal Chemistry and Experimental and Clinical Pharmacology, College of Pharmacy, and the Masonic Cancer Center, University of Minnesota, Minneapolis, Minnesota 55455 Cyclophosphamide (CPA) is a DNA alkylating agent widely used in cancer chemotherapy. CPA undergoes metabolic activation to phosphoramide mustard and nornitrogen mustard (NOR) which alkylate the N-7 position of guanine in DNA to produce N-[2-(N7-guaninyl) ethyl]-N-[2-hydroxyethyl]-amine (G-NOR-OH) monoadducts and N,Nbis[2-(N7-guaninyl) ethyl] amine cross-links (G-NOR-G). G-NOR-G cross-links are strongly cytotoxic and are thought to be responsible for the biological activity of CPA. In the present work, an isotope dilution high-performance liquid chromatography-electrospray ionization (positive ion) tandem mass spectrometry (HPLC-ESI+-MS/MS) methodology was developed to accurately quantify G-NOR-G adducts in human blood. In our approach, DNA extracted from white blood cells (5-20 µg) is spiked with an internal standard of [15N10]-G-NOR-G and subjected to thermal hydrolysis to release G-NOR-G adducts from the DNA backbone. Following solid phase extraction, G-NOR-G conjugates are quantified by capillary HPLC-ESI-MS/MS in the selected reaction monitoring mode. The application of the new methodology is demonstrated for DNA extracted from blood of three cancer patients receiving 50-60 mg/ kg of intravenous CPA. The highest numbers of GNOR-G adduct (up to 18 adducts per 106 normal nucleotides) were observed 4-8 h following CPA administration, followed by a gradual decrease over time, probably due to adduct hydrolysis, DNA repair, and white blood cell death. This methodology will be useful for future investigations of the interindividual differences for CPA-induced DNA-DNA cross-linking. Cyclophosphamide (CPA) is a DNA alkylating agent commonly used in the treatment of lymphoma, leukemia, and solid tumors.1,2 It is also included in most conditioning regimens prior to * To whom correspondence should be addressed. † Department of Medicinal Chemistry. ‡ Department of Experimental and Clinical Pharmacology.
3650
Analytical Chemistry, Vol. 82, No. 9, May 1, 2010
hematopoietic cell transplantation.2-5 CPA is a prodrug requiring metabolic activation by cytochrome P450 monooxygenases (mainly the 2B6 isoform), yielding 4-hydroxy-CPA which then breaks down releasing phosphoramide mustard (PM) and acrolein (Scheme 1).6-8 PM in turn undergoes nonenzymatic degradation to produce nornitrogen mustard (NOR). Both PM and NOR alkylate guanine bases within DNA to form guanine monoadducts, N-[2-(N7guaninyl) ethyl]-N-[2-hydroxyethyl]-amine (G-NOR-OH), and DNA-DNA cross-links, N,N-bis[2-(N7-guaninyl) ethyl] amine (GNOR-G) (Scheme 1), although PM is considered the major DNAreactive metabolite.9-11 The majority of G-NOR-G cross-links have interstrand geometry and preferentially form between the distal guanine residues in 5′-GNC-3′ trinucleotides.12-14 Both N-7(1) Reece, D. E.; Connors, J. M.; Spinelli, J. J.; Barnett, M. J.; Fairey, R. N.; Klingemann, H. G.; Nantel, S. H.; O’Reilly, S.; Shepherd, J. D.; Sutherland, H. J. Blood 1994, 83, 1193–1199. (2) Demirer, T.; Buckner, C. D.; Appelbaum, F. R.; Bensinger, W. I.; Sanders, J.; Lambert, K.; Clift, R.; Fefer, A.; Storb, R.; Slattery, J. T. Bone Marrow Transplant. 1996, 17, 491–495. (3) Demirer, T.; Buckner, C. D.; Appelbaum, F. R.; Clift, R.; Storb, R.; Myerson, D.; Lilleby, K.; Rowley, S.; Bensinger, W. I. Bone Marrow Transplant. 1996, 17, 769–774. (4) Demirer, T.; Buckner, C. D.; Appelbaum, F. R.; Lambert, K.; Bensinger, W. I.; Clift, R.; Storb, R.; Slattery, J. T. Bone Marrow Transplant. 1996, 17, 341–346. (5) Demirer, T.; Weaver, C. H.; Buckner, C. D.; Petersen, F. B.; Bensinger, W. I.; Sanders, J.; Clift, R. A.; Lilleby, K.; Anasetti, C.; Martin, P. J. Clin. Oncol. 1995, 13, 596–602. (6) Sadagopan, N.; Cohen, L.; Roberts, B.; Collard, W.; Omer, C. J. Chromatogr., B: Anal. Technol. Biomed. Life Sci. 2001, 759, 277–284. (7) Kalhorn, T. F.; Howald, W. N.; Cole, S.; Phillips, B.; Wang, J.; Slattery, J. T.; McCune, J. S. J. Chromatogr., B: Anal. Technol. Biomed. Life Sci. 2006, 835, 105–113. (8) Connors, T. A.; Foster, A. B.; Gilsenan, A. M.; Jarman, M.; Tisdale, M. J. Biochem. Pharmacol. 1972, 21, 1373–1376. (9) Benson, A. J.; Martin, C. N.; Garner, R. C. Biochem. Pharmacol. 1988, 37, 2979–2985. (10) Hemminki, K. Chem. Biol. Interact. 1987, 61, 75–788. (11) Hemminki, K. Cancer Res. 1985, 45, 4237–4243. (12) Millard, J. T.; Luedtke, N. W.; Spencer, R. J. Anti-Cancer Drug Des. 1996, 11, 485–492. (13) Rajski, S. R.; Williams, R. M. Chem. Rev. 1998, 98, 2723–2796. (14) Dong, Q.; Barsky, D.; Colvin, M. E.; Melius, C. F.; Ludeman, S. M.; Moravek, J. F.; Colvin, O. M.; Bigner, D. D.; Modrich, P.; Friedman, H. S. Proc. Natl. Acad. Sci. U.S.A. 1995, 92, 12170–12174. 10.1021/ac902923s 2010 American Chemical Society Published on Web 04/02/2010
Scheme 1. Metabolism of Cyclophosphamide and the Formation of Phosphoramide Mustard- and Nornitrogen Mustard-Induced DNA Adducts
guanine lesions are hydrolytically labile and spontaneously depurinate to produce nucleobase adducts and abasic sites in DNA.10 Previous studies reported that ∼90% of CPA-induced DNA lesions were guanine monoadducts11,15 and that the emergence of such adducts correlated with the onset of cytotoxicity.9 However, 1,3-interstrand G-NOR-G lesions are thought to play a central role in the cytotoxic response because they prevent DNA strand separation which is required for DNA replication and transcription.13,15,16 Therefore, G-NOR-G lesions are considered the primary cytotoxic adducts responsible for the antineoplastic activity of CPA in vivo.9-11 The therapeutic efficacy of CPA and other DNA cross-linking agents in cancer patients relies on a fine balance of interstrand DNA-DNA cross-link (ICL) formation and repair. Although positive therapeutic response to these agents is dependent on efficient DNA adduct formation,15 an overproduction of DNA-DNA cross-links and associated chromosomal damage can lead to systemic toxicity, especially in individuals with deficient DNA repair. For example, Fanconi Anemia patients exhibit varying degrees of hypersensitivity to CPA and other DNA cross-linking agents.17 Knowledge of the amounts of G-NOR-G adducts produced in the tissues of patients undergoing CPA therapy could guide patient treatment, thereby improving drug efficacy and reducing side effects. However, previous studies of CPA induced DNA damage have been limited. In vitro detection of G-NOR-G in NOR-treated calf thymus DNA has been demonstrated by high-performance liquid chromatography (HPLC)-UV analysis of samples subjected to neutral thermal hydrolysis.10,18 The formation of ICL lesions has been detected in human leukemia cells following in vivo treatment with CPA, though no structural information was provided.15,19 In a murine system, clinical doses of CPA induced approximately 3 nucleobase adducts per 106 normal guanines, among those, over 90% were guanine monoadducts of, i.e., G-NOR-OH in Scheme 1, which were detectable up to 72 h following exposure.9,10 Souliotis et al.15 quantified DNA-DNA (15) Souliotis, V. L.; Dimopoulos, M. A.; Sfikakis, P. P. Clin. Cancer Res. 2003, 9, 4465–4474. (16) Scharer, O. D. ChemBioChem 2005, 6, 27–32. (17) Yabe, M.; Yabe, H.; Hamanoue, S.; Inoue, H.; Matsumoto, M.; Koike, T.; Ishiguro, H.; Morimoto, T.; Arakawa, S.; Ohshima, T.; Masukawa, A.; Miyachi, H.; Yamashita, T.; Katob, S. Int. J. Hematol. 2007, 85, 354–361. (18) Thulin, H.; Zorcec, V.; Segerback, D.; Sundwall, A.; Tornqvist, M. Chem. Biol. Interact. 1996, 99, 263–275. (19) DeNeve, W.; Valeriote, F.; Edelstein, M.; Everett, C.; Bischoff, M. Cancer Res. 1989, 49, 3452–3456.
cross-links in the blood of a cancer patient treated with intravenous CPA. Maximum DNA-DNA cross-link formation (3 adducts per 106 nucleotides) was reached 24 h postexposure; however, the study was limited to one patient and employed nonspecific and indirect gel electrophoresis methods to estimate DNA adduct numbers.15 To our knowledge, no accurate quantitative methods for the analysis of G-NOR-G in humans have been described. Isotope dilution mass spectrometry has been previously used to quantify DNA-DNA cross-links induced by carcinogens and drugs. For example, Van den Driessche et al.20 developed a quantitative high-performance liquid chromatography-electrospray ionization tandem mass spectrometry (HPLC-ESI-MS/MS) methodology for guanine-guanine adducts of melphalan in vivo. Baskerville-Abraham et al. reported a sensitive method for UPLC-ESI-MS/MS analysis of intrastrand guanine-guanine cross-links induced by cisplatin.21 Our laboratory employed nanoflow HPLC-nanoESI-MS/MS to quantify bis-N7-guanine and guanine-adenine adducts of 1,2,3,4-diepoxybutane in tissues of laboratory animals exposed to its metabolic precursor, 1,3butadiene, by inhalation.22-24 We now describe an accurate, sensitive, and specific isotope dilution HPLC-ESI-MS/MS methodology for quantifying CPA induced G-NOR-G lesions in human blood. The new methodology was employed to investigate the formation and persistence of G-NOR-G cross-links in blood of three cancer patients receiving CPA therapy. To our knowledge, this is the first report of a mass spectrometry based quantitative analysis of DNA-DNA cross-links of CPA in vivo. MATERIALS AND METHODS Caution. Nornitrogen mustard is a known human carcinogen and must be handled with adequate safety precautions. Materials. 2′-Deoxyguanosine (dG), bis-2-chloroethylamine hydrochloride (nornitrogen mustard), and calf thymus (CT) DNA (20) Van den Driessche, B.; Lemiere, F.; Witters, E.; Van Dongen, W.; Esmans, E. L. Rapid Commun. Mass Spectrom. 2005, 19, 449–454. (21) Baskerville-Abraham, I. M.; Boysen, G.; Troutman, J. M.; Mutlu, E.; Collins, L.; Dekrafft, K. E.; Lin, W.; King, C.; Chaney, S. G.; Swenberg, J. A. Chem. Res. Toxicol. 2009, 22, 905–912. (22) Goggin, M.; Loeber, R.; Park, S.; Walker, V.; Wickliffe, J.; Tretyakova, N. Chem. Res. Toxicol. 2007, 20, 839–847. (23) Goggin, M.; Anderson, C.; Park, S.; Swenberg, J.; Walker, V.; Tretyakova, N. Chem. Res. Toxicol. 2008, 21, 1163–1170. (24) Goggin, M.; Swenberg, J. A.; Walker, V. E.; Tretyakova, N. Cancer Res. 2009, 69, 2479–2486.
Analytical Chemistry, Vol. 82, No. 9, May 1, 2010
3651
Scheme 2. Synthesis of G-NOR-G and [15N10]-G-NOR-G
were obtained from Sigma-Aldrich. 15N5-dG was purchased from Cambridge Isotopes Inc. (Andover, MA), and HPLC-MS grade ammonium acetate was from Fisher Scientific (Pittsburgh, PA). All other chemicals and solvents were obtained from SigmaAldrich (Millwaukee, WI, St. Louis, MO), unless otherwise specified. G-NOR-G and 15N10-G-NOR-G were synthesized in our laboratory and described below. N,N-Bis(2-[N7-guaninyl]ethyl)amine (G-NOR-G) Synthesis. Bis2-chloroethylamine hydrochloride (HCl salt of nornitrogen mustard, 11 mg) was mixed with 2′-deoxyguanosine (4.6 mg) in 0.25 M sodium acetate solution (1 mL). The reaction mixture was incubated at 37 °C for 5 h. At the end of incubation, NaCl (56 mg) was added, and the resulting solution was heated at 95 °C for 30 min to cleave the glycosidic bonds (Scheme 2). G-NOR-G was isolated by HPLC using a Supelcosil LC-18 DB (4.6 mm × 250 mm, 5 µm) column eluted with a gradient of 15 mM ammonium acetate, pH 6.8 (A) and 100% methanol (B). The flow rate was 1 mL/min, and the solvent composition was kept at 100% (A) for 0-6 min, followed by a linear increase to 4.5% (B) for 2 min and further increased to 35% (B) over 32 min. Under these conditions, G-NOR-G eluted between 24 and 25 min. 15N10-GNOR-G was prepared analogously starting with 15N5-dG. Both synthetic standards were characterized by UV and mass spectrometry, and their purity was established by capillary HPLC-UV-ESI-MS/MS. G-NOR-G: UVmax ) 251 and 280 nm; ESI+-MS, m/z 372 ([M + H] +); MS/MS, m/z 372 f 221 ([M + H - Gua]+), 178 ([GuasCHdCH2 + H]+). 15N10-G-NOR-G: UVmax ) 251 and 280 nm; ESI+-MS, m/z 382 ([M + H] +); MS/MS, m/z 382 f 226 ([M + H - 15N5-Gua]+), 183 ([15N5GuasCHdCH2 + H]+). The concentrations of G-NOR-G stock solutions were determined by UV spectrophotometry (ε251 ) 15 700), while 15N10-G-NOR-G concentrations were determined by HPLC-ESI-MS/MS analyses of solutions spiked with known amounts of unlabeled G-NOR-G. HPLC-ESI-MS/MS confirmed that the residue of unlabeled G-NOR-G in [15N10]labeled standard was less than
