Anal. Chem. 2009, 81, 9454–9461
NanoLC-MS/MS Analyses of Urinary Desmosine, Hydroxylysylpyridinoline and Lysylpyridinoline as Biomarkers for Chronic Graft-versus-Host Disease Michel Boutin,† Imran Ahmad,‡ Marjo Jauhiainen,† Nathalie Lachapelle, Claude Rondeau, Jean Roy,‡ and Pierre Thibault*,†,§ Institute for Research in Immunology and Cancer, and Department of Chemistry, Universite´ de Montre´al, P.O. Box 6128, Station Centre-ville, Montre´al, Canada H3C 3J7, and Blood and Marrow Transplant Program, Maisonneuve-Rosemont Hospital, 5415, Assomption Blvd., Montreal QC, Canada H1T 2M4 Chronic graft-versus-host disease (cGVHD) is a common and potentially lethal complication of allogeneic hematopoietic stem cell transplantation (HSCT). cGVHD as well as the transplant procedure itself (chemotherapy with or without radiotherapy) can lead to the degradation of connective tissue components such as elastin and collagen. The catabolism of these structural proteins releases desmosine (DES), lysylpyridinoline (LP), hydroxylysylpyridonoline (HP), and related pyridinium-based crosslinkers analogues that could represent potential biomarkers for cGVHD. This study reports the development of a sensitive liquid chromatography/tandem mass spectrometry method for the simultaneous analysis of N-propyl derivatives of DES, HP, and LP. The concentrations of free and total forms of urinary DES, HP, and LP were determined using synthetic deuterated internal standards. This method enabled accurate quantitation of these pyridinium-based cross-linkers from as little as 100 µL of urine with detection limits of 0.03-0.10 ng/mL. These compounds were analyzed in urine samples from three groups of patients: (1) Healthy volunteers, (2) Autologous HSCT recipients (who cannot develop cGVHD), and (3) Allogeneic HSCT recipients at onset of cGHVD. These analyses revealed that the urinary concentrations of DES, HP, and LP in the autologous recipients were greater or equal to the cGVHD group although both groups showed marked increase in the levels of these compounds compared to healthy individuals. These results suggest that the chemotherapy treatment has significant effects on the turnover of elastin and collagen, and that these biomarkers could be effective during prospective analyses to determine the onset of cGVHD. Hematopoietic stem cell transplantation (HSCT) is a life-saving treatment for children and adults with leukemia, inborn immune disorders, marrow failure, and several types of cancers. The number of these procedures has increased dramatically over the * To whom correspondence should be addressed. E-mail: pierre.thibault@ umontreal.ca. Phone: +1 (514) 343-6910. Fax: +1 (514) 343-7586. † Institute for Research in Immunology and Cancer. ‡ Blood and Marrow Transplant Program, Maisonneuve-Rosemont Hospital. § Department of Chemistry.
9454
Analytical Chemistry, Vol. 81, No. 22, November 15, 2009
last 20 years, and an estimated 50000 HSCT are performed annually worldwide.1 The harvest of hematopoietic stem cells can be performed either in the bone marrow or in the peripheral blood after drug-induced stem-cell mobilization from the marrow. While autologous HSCT uses the patient’s own cells and thus includes no immunological complications, patients undergoing allogeneic HSCT can develop graft-versus-host disease (GVHD), a potentially lethal complication where the immune system arising from the graft recognizes the recipient as foreign and mounts an immunological response against host organs such as the liver, skin, gastrointestinal tract, oral mucosa, lacrymal and salivary glands, lungs, and possibly any other organ.2-4 Despite continuous improvement in supportive care over the past decades, allogeneic HSCT remains associated with significant toxicity and a treatmentrelated mortality rate at 2 years ranging from 14 to 41%.5 Clinically, GVHD is divided into acute and chronic forms. Acute graft-versushost disease (aGVHD) usually occurs within the first 100 days after transplantation in 30-50% of HSCT recipients, and is the principal cause of noninfectious early death. It is believed to result from graft T-cells recognizing histoincompatibility between donor and recipient with an allo-immune response leading to skin, liver, and intestinal damage.6 Chronic graft-versus-host disease (cGVHD) occurs more than 100 days after transplantation3 in as many as 60-70% of HSCT recipients and is the main cause of long-term complications, impaired quality of life and death in cured patients.2,7 cGVHD is the manifestation of end-stage organ damage resulting from chronic subclinical inflammation, and early clinical presentation is often insidious and non specific: anorexia, weight loss, pruritus, blood eosinophilia. Virtually every organ can be damaged, each to a various extent, including most frequently: skin, mouth, (1) Horowitz, M. M.; Confer, D. L. Hematol. Am. Soc. Hematol. Educ. Program 2005, 469–475. (2) Joseph, R. W.; Couriel, D. R.; Komanduri, K. V. J. Supportive Oncol. 2008, 6, 361–372. (3) Jacobsohn, D. A. Bone Marrow Transplant. 2008, 41, 215–221. (4) Higman, M. A.; Vogelsang, G. B. Br. J. Haematol. 2004, 125, 435–454. (5) Sorror, M. L.; Maris, M. B.; Storb, R.; Baron, F.; Sandmaier, B. M.; Maloney, D. G.; Storer, B. Blood 2005, 106, 2912–2919. (6) Sun, Y.; Tawara, I.; Toubai, T.; Reddy, P. Transl. Res. 2007, 150, 197–214. (7) Filipovich, A. H.; Weisdorf, D.; Pavletic, S.; Socie, G.; Wingard, J. R.; Lee, S. J.; Martin, P.; Chien, J.; Przepiorka, D.; Couriel, D.; Cowen, E. W.; Dinndorf, P.; Farrell, A.; Hartzman, R.; Henslee-Downey, J.; Jacobsohn, D.; McDonald, G.; Mittleman, B.; Rizzo, J. D.; Robinson, M.; Schubert, M.; Schultz, K.; Shulman, H.; Turner, M.; Vogelsang, G.; Flowers, M. E. Biol. Blood Marrow Transplant. 2005, 11, 945–956. 10.1021/ac9018796 CCC: $40.75 2009 American Chemical Society Published on Web 10/22/2009
eyes, gastrointestinal tract, liver, and lungs. cGVHD pathophysiology is complex and shares many common features with autoimmune connective tissue diseases such as systemic lupus erythematosus and scleroderma. Treatment, which consists of glucocorticoids and other immunosuppressant drugs (calcineurin inhibitors, mycophenolate mofetil, etc.) for several months, has significant side effects including lethal infections, and varying efficacy because of partially irreversible lesions, with >50% of patients on therapy beyond 2 years of onset.8 To date, the initial diagnosis of cGVHD is mainly based on clinical features and sometimes on tissue biopsies. Currently, there is no noninvasive, unbiased laboratory test for the diagnosis of cGVHD at initial presentation or at relapse. There is thus an urgent need to develop biomarkers that could identify cGVHD at an early stage, that is, before irreversible tissue damage occurs, to initiate appropriate immunosuppressive therapy. While a large number of proteomics investigations9-16 reported the identification of potential GVHD serum and urine biomarkers, the clinical manifestation of GVHD resulting in the targeted degradation of connective tissues can provide an opportunity for the discovery of disease relevant biomarkers. Indeed, GVHD affects many elastin and collagen-rich organs such as skin, lungs, and intestines. Other groups14,16 have already observed an increased excretion of Type III collagen in patients with GVHD. The integrity of elastin fibers is maintained via two cross-linking amino acids desmosine (DES) (Figure 1A) and isodesmosine (IDES) (Figure 1B). DES and IDES are pyridinium-based structural isomers arising from the condensation of four lysine residues by lysyl-oxidase.17-19 For simplicity, DESs will refer to both DES and IDES. We previously developed a high sensitivity nanoLCMS/MS method for the analysis of DESs in urine.20 DESs have already been used as biomarkers for other diseases affecting (8) Arora, M. Best Pract. Res. Clin. Haematol. 2008, 21, 271–279. (9) Fujii, H.; Cuvelier, G.; She, K.; Aslanian, S.; Shimizu, H.; Kariminia, A.; Krailo, M.; Chen, Z.; McMaster, R.; Bergman, A.; Goldman, F.; Grupp, S. A.; Wall, D. A.; Gilman, A. L.; Schultz, K. R. Blood 2008, 111, 3276–3285. (10) Kaiser, T.; Kamal, H.; Rank, A.; Kolb, H. J.; Holler, E.; Ganser, A.; Hertenstein, B.; Mischak, H.; Weissinger, E. M. Blood 2004, 104, 340– 349. (11) McGuirk, J.; Hao, G.; Hou, W.; Abhyankar, S.; Williams, C.; Yan, W.; Yuan, J.; Guan, X.; Belt, R.; Dejarnette, S.; Wieman, J.; Yan, Y. J. Hematol. Oncol. 2009, 2, 17. (12) Paczesny, S.; Krijanovski, O. I.; Braun, T. M.; Choi, S. W.; Clouthier, S. G.; Kuick, R.; Misek, D. E.; Cooke, K. R.; Kitko, C. L.; Weyand, A.; Bickley, D.; Jones, D.; Whitfield, J.; Reddy, P.; Levine, J. E.; Hanash, S. M.; Ferrara, J. L. Blood 2009, 113, 273–278. (13) Paczesny, S.; Levine, J. E.; Braun, T. M.; Ferrara, J. L. Biol. Blood Marrow Transplant. 2008, 15, 33–38. (14) Pihusch, M.; Wegner, H.; Goehring, P.; Salat, C.; Pihusch, V.; Andreesen, R.; Kolb, H. J.; Holler, E.; Pihusch, R. Bone Marrow Transplant. 2005, 36, 631–637. (15) Srinivasan, R.; Daniels, J.; Fusaro, V.; Lundqvist, A.; Killian, J. K.; Geho, D.; Quezado, M.; Kleiner, D.; Rucker, S.; Espina, V.; Whiteley, G.; Liotta, L.; Petricoin, E.; Pittaluga, S.; Hitt, B.; Barrett, A. J.; Rosenblatt, K.; Childs, R. W. Exp. Hematol. 2006, 34, 796–801. (16) Weissinger, E. M.; Schiffer, E.; Hertenstein, B.; Ferrara, J. L.; Holler, E.; Stadler, M.; Kolb, H. J.; Zander, A.; Zurbig, P.; Kellmann, M.; Ganser, A. Blood 2007, 109, 5511–5519. (17) Mithieux, S. M.; Wise, S. G.; Raftery, M. J.; Starcher, B.; Weiss, A. S. J. Struct. Biol. 2005, 149, 282–289. (18) Rucker, R. B.; Kosonen, T.; Clegg, M. S.; Mitchell, A. E.; Rucker, B. R.; Uriu-Hare, J. Y.; Keen, C. L. Am. J. Clin. Nutr. 1998, 67, 996S–1002S. (19) Ryvkin, F.; Greenaway, F. T. J. Inorg. Biochem. 2004, 98, 1427–1435. (20) Boutin, M.; Berthelette, C.; Gervais, F. G.; Scholand, M. B.; Hoidal, J.; Leppert, M. F.; Bateman, K. P.; Thibault, P. Anal. Chem. 2009, 81, 1881– 1887.
Figure 1. Structure of (A) Desmosine (DES) (X,R ) H); Deuterated desmosine (D4-DES) (X ) D, R ) H); Propionylated desmosine (DES(prop)) (X ) H, R ) COCH2CH3), (B) Isodesmosine (IDES) (R ) H); Propionylated isodesmosine (IDES(prop)) (R ) COCH2CH3), (C) Hydroxylysylpyridinoline (HP) (X,R ) H); Deuterated hydroxylysylpyridinoline (D4-HP) (X ) D, R ) H); Propionylated hydroxylysylpyridinoline (HP(prop)) (X ) H, R ) COCH2CH3), and (D) Lysylpyridinoline (LP) (R ) H); Propionylated lysylpyridinoline (LP(prop)) (R ) COCH2CH3).
elastin such as COPD,20 cystic fibrosis,21 acute lung injury,22,23 pseudoxanthoma elasticum,24 and aneurysm.25 Similarly, collagen fibrils are cross-linked via hydroxylysylpyridinoline (HP) (Figure 1C) and lysylpyridinoline (LP) (Figure 1D). HP is widely distributed in Type I collagen from bone and in Type II collagen from cartilage and other connective tissues except skin. LP is found in Type I collagen of bone and dentin. HP and LP were used in other studies as biomarkers for rheumatoid arthritis,27 osteomyelitis,26 osteoarthritis,27,28 osteoporosis,29,30 and oral cancer.31 DESs, HP, and LP can be excreted as free pyridinium analogues (free form) (21) Bode, D. C.; Pagani, E. D.; Cumiskey, W. R.; von Roemeling, R.; Hamel, L.; Silver, P. J. Pulm. Pharmacol. Ther. 2000, 13, 175–180. (22) Fill, J. A.; Brandt, J. T.; Wiedemann, H. P.; Rinehart, B. L.; Lindemann, C. F.; Komara, J. J.; Bowsher, R. R.; Spence, M. C.; Zeiher, B. G. Biomarkers 2006, 11, 85–96. (23) McClintock, D. E.; Starcher, B.; Eisner, M. D.; Thompson, B. T.; Hayden, D. L.; Church, G. D.; Matthay, M. A. Am. J. Physiol. Lung Cell Mol. Physiol. 2006, 291, L566–571. (24) Annovazzi, L.; Viglio, S.; Gheduzzi, D.; Pasquali-Ronchetti, I.; Zanone, C.; Cetta, G.; Iadarola, P. Eur. J. Clin. Invest. 2004, 34, 156–164. (25) Osakabe, T.; Usami, E.; Sato, A.; Sasaki, S.; Watanabe, T.; Seyama, Y. Biol. Pharm. Bull. 1999, 22, 854–857. (26) Springer, I. N.; Wiltfang, J.; Dunsche, A.; Lier, G. C.; Bartsch, M.; Warnke, P. H.; Barth, E. L.; Terheyden, H.; Russo, P. A.; Czech, N.; Acil, Y. Int. J. Oral Maxillofac. Surg. 2007, 36, 527–532. (27) Muller, A.; Jakob, K.; Hein, G. E. Ann. Rheum. Dis. 2003, 62, 65–67. (28) Kaufmann, J.; Mueller, A.; Voigt, A.; Carl, H. D.; Gursche, A.; Zacher, J.; Stein, G.; Hein, G. Rheumatology (Oxford) 2003, 42, 314–320. (29) Yu, S. L.; Ho, L. M.; Lim, B. C.; Sim, M. L. Ann. Acad. Med. Singapore 1998, 27, 527–529.
Analytical Chemistry, Vol. 81, No. 22, November 15, 2009
9455
or covalently attached to peptides (bounded form). To identify all of these cross-linkers, (Total ) free + bounded forms) the samples are digested in acid to release the bounded form.21,22,24,26,27,31-41 Results from other research groups27,37 suggest that the free crosslinkers and the peptides-bounded cross-linkers excreted in urine are generated by different degradation pathways. For example, Ma et al.37 measured an increased urinary excretion of free DESs for COPD patients compared to controls while total DESs remained unaffected. They suggested that an increase of the free form of DESs reflects an increased activity of elastase. Mu¨ller et al.27 observed that the peptides-bounded forms of HP and LP represent meaningful biomarkers for rheumatoid arthritis compared to their corresponding free forms. Some of the free crosslinkers would be produced in the kidney. In the present study, we sought to examine urinary crosslinking amino acids, characteristics of the degradation of elastin and collagen as potential cGVHD biomarkers. In addition, we describe a chromatographic method that enables simultaneous analysis of DESs, HP, and LP from a single run. This high sensitivity method is based on nanoflow liquid chromatography tandem mass spectrometry (nanoLC-MS/MS) and allows the analysis of DESs, HP, and LP in urine samples with sub ng/mL detection limits. The approach involves the formation of stable N-propionyl derivatives and the use of custom synthetized D4DES and D4-HP as internal standards to compensate for sample losses and instrumental variability. We demonstrate the application of this approach in a cohort of 39 patients from three groups: (1) healthy volunteers, (2) autologous transplant recipients (who never develop cGVHD), and (3) allogeneic transplant recipients with new onset of extensive cGVHD. This study also reports on the concentration levels of the free and total forms of DESs, HP, and LP as potential urinary biomarkers for cGVHD. EXPERIMENTAL SECTION Chemicals. DES (99.5%) was purchased from Elastin Products Company Inc. (Owensville, MO), and HP was kindly provided by (30) Kushida, K.; Takahashi, M.; Kawana, K.; Inoue, T. J. Clin. Endocrinol. Metab. 1995, 80, 2447–2450. (31) Springer, I. N.; Terheyden, H.; Dunsche, A.; Czech, N.; Suhr, M. A.; Tiemann, M.; Hedderich, J.; Acil, Y. Br. J. Cancer 2003, 88, 1105–1110. (32) Annovazzi, L.; Viglio, S.; Perani, E.; Luisetti, M.; Baraniuk, J.; Casado, B.; Cetta, G.; Iadarola, P. Electrophoresis 2004, 25, 683–691. (33) Boschetto, P.; Quintavalle, S.; Zeni, E.; Leprotti, S.; Potena, A.; Ballerin, L.; Papi, A.; Palladini, G.; Luisetti, M.; Annovazzi, L.; Iadarola, P.; De Rosa, E.; Fabbri, L. M.; Mapp, C. E. Thorax 2006, 61, 1037–1042. (34) Cocci, F.; Miniati, M.; Monti, S.; Cavarra, E.; Gambelli, F.; Battolla, L.; Lucattelli, M.; Lungarella, G. Int. J. Biochem. Cell Biol. 2002, 34, 594–604. (35) Cumiskey, W. R.; Pagani, E. D.; Bode, D. C. J. Chromatogr. B: Biomed. Appl. 1995, 668, 199–207. (36) Kaga, N.; Soma, S.; Fujimura, T.; Seyama, K.; Fukuchi, Y.; Murayama, K. Anal. Biochem. 2003, 318, 25–29. (37) Ma, S.; Lin, Y. Y.; Turino, G. M. Chest 2007, 131, 1363–1371. (38) Ostanek, L.; Pawlik, A.; Brzosko, I.; Brzosko, M.; Sterna, R.; Drozdzik, M.; Gawronska-Szklarz, B. Clin. Rheumatol. 2004, 23, 214–217. (39) Stolk, J.; Veldhuisen, B.; Annovazzi, L.; Zanone, C.; Versteeg, E. M.; van Kuppevelt, T. H.; Berden, J. H.; Nieuwenhuizen, W.; Iadarola, P.; Luisetti, M. Respir. Res. 2005, 6, 47. (40) Stolk, J.; Veldhuisen, B.; Annovazzi, L.; Zanone, C.; Versteeg, E. M.; van Kuppevelt, T. H.; Nieuwenhuizen, W.; Iadarola, P.; Berden, J. H.; Luisetti, M. Respir. Res. 2006, 7, 20. (41) Viglio, S.; Iadarola, P.; Lupi, A.; Trisolini, R.; Tinelli, C.; Balbi, B.; Grassi, V.; Worlitzsch, D.; Doring, G.; Meloni, F.; Meyer, K. C.; Dowson, L.; Hill, S. L.; Stockley, R. A.; Luisetti, M. Eur. Respir. J. 2000, 15, 1039–1045.
9456
Analytical Chemistry, Vol. 81, No. 22, November 15, 2009
Table 1. Demographics and Mean Post Transplant Collection Times of Patient Groups
groups healthy autologous cGVHD
mean post trans-plant men (n) women (n) mean age (yrs) collection (days) 3 9 7
7 4 9
34(26-51) 45(25-69) 56(35-68)
n.a. 162(90-377) 536(107-2966)
Dr. Kevin Bateman (Merck Frosst Center for Therapeutic Research, Kirkland, Canada). HPLC grade acetonitrile (ACN), methanol (MeOH), and toluene, and molecular biology grade isopropanol (i-Pr) were from Fisher Scientific Company (Ottawa, Canada). ACS grade hydrochloric acid (HCl) and 98% formic acid were from EMD Chemicals Inc. (Darmstadt, Germany). ACS grade ammonium hydroxide, 99+% propionic anhydride, 99.0% ammonium bicarbonate (NH4HCO3), 99.0% trifluoroacetic acid (TFA), 98% 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU), 99.96% deuterium oxide (D2O) were from Sigma-Aldrich Canada Ltd. (Oakville, Canada). Deuterated Standard Synthesis and Purification. The protocol used to synthesize and purify D4-DES (Figure 1A) was described elsewhere.20 This procedure was adapted for the synthesis of D4-HP. Briefly, 240 µg of HP were dissolved in 0.5 mL of D2O, and 25 µL of DBU was added. The solution was stirred at 110 °C for 7 days in a closed vessel. The isotopic profile of HP was analyzed at day 0, day 3, and day 7. At the end of the reaction, the BDU was removed from the reaction mixture using 5 consecutive extractions with 500 µL of CHCl3. The reaction mixture was then evaporated to dryness in a SPDseries SpeedVac concentrator (Thermo Fisher Scientific Inc., Waltham, MA.). The reaction product was redissolved in 1 mL of H2O and evaporated to dryness three times to remove unlabeled deuterium atoms. The purified D4-HP was dissolved and stored in 1 mL of 0.2% formic acid to inactivate the remaining traces of DBU and thus prevent the loss of the deuterium atoms. D4-HP was subsequently analyzed by infusion and nano-LC-MS on a Q-trap 4000 mass spectrometer (ABI/ MDS Analytical Technologies, Foster City, CA) to confirm its purity. To determine the D4-HP concentration, an aliquot of the D4-HP solution was propionylated and analyzed as described in Sections 2.4 and 2.5. Urine Samples. Urine samples from 10 healthy individuals, 13 recipients of autologous HSCT, and 16 recipients of allogeneic HSCT at onset of cGVHD were collected prior to initiation of immunosuppressive drugs. Demographic data and the mean posttransplant collection times of each group are presented in Table 1. All patients and healthy controls enrolled in this study were exempt of chronic diseases. This study was approved by the Research Ethics Committee of Maisonneuve-Rosemont Hospital, and consent forms were obtained from all healthy volunteers and patients enrolled in this study. Patients who received an autologous HSCT were selected as negative controls because they received high-dose conditioning regimens but were unable to develop cGVHD. The creatinine level was measured for each urine sample using a Creatinine Assay Kit from BioVision Inc. (Mountain View, CA) to account for the variability in urine dilution. Sample Preparation. For the total form of DESs, HP, and LP, an aliquot of 100 µL of each urine sample was digested in 500
µL of 6 N HCl (24 h, 110 °C), under N2 atmosphere, using an AccuBlock Digital Dry Bath from Labnet International Inc. (Woodbridge, NJ) to release DESs, HP, and LP covalently bounded to proteins. The samples were evaporated to dryness in the SpeedVac, and 0.75 ng of D4-DES and D4-HP were added as internal standards to compensate for any loss during the subsequent steps of sample preparation and to increase the precision of the mass spectrometry analysis. The sample residues were then resuspended in 500 µL of 0.05 N HCl and purified using 1 mL (30 mg, 60 µm) mixed-mode strong cation exchange (MCX) cartridges (Oasis, Waters Corporation, Milford, MA).37,42 The samples were loaded on the MCX cartridges previously conditioned with 1 mL of MeOH and 1 mL of deionized water. The cartridges were washed with 1 mL of deionized water, 1 mL MeOH, and 1 mL of HCl 0.1 N prior to samples’ elution with 1.5 mL of 2 N HCl. The eluates were dried in the SpeedVac and were propionylated (Figure 1) to facilitate the analysis of analytes on C18 reverse phase chromatography. For sample derivatization, 50 µL of NH4HCO3 0.1 N was added to each sample followed by 50 µL of a freshly prepared solution containing propionic anhydride (23% v/v) and NH4OH (9% v/v) in i-Pr. After 30 min of reaction, the samples containing the propionylated cross-linkers were evaporated to dryness in the SpeedVac, and the residues were resuspended in 50 µL of 0.2% formic acid. For the analysis of free DESs, HP, and LP, the method presented above was used except that the acid digestion step was omitted, and the urine samples, spiked with the internal standards, were diluted with 1 mL of HCl 0.05 N and directly loaded on the MCX cartridges. NanoLC-MS/MS Analysis. The samples were separated with an Agilent (Santa Clara, CA) nano-LC (1100 Series) system using a custom-made pre-column (0.3 i.d. × 4 mm) and analytical column (150 µm i.d. × 10 cm) packed with a C18 (3 µm, 300 A) Jupiter (Phenomenex Inc., Torrance, CA) stationary phase. The injection volume was 5 µL and sample loading was performed at a flow rate of 10 µL/min for 4 min. Sample elution on the analytical column was achieved under isocratic conditions (13% ACN, 0.2% formic acid) at a flow rate of 0.6 µL/min. Under these chromatographic conditions, DES and IDES co-elute. According to the literature,37 the ratio between these two biomarkers appears to be constant across different sample sets suggesting that there is no inherent advantage in resolving them at the expense of sensitivity and sample throughput. Three replicates analyses were performed for each sample. For the calibration curve, DES and HP standards ranging from 0.005 to 20 ng were added to 0.75 ng of D4-DES and D4-HP, propionylated (Section 2.4), dried with a SpeedVac, and resuspended in 0.2% aqueous formic acid. NanoLC-MS/MS analyses were performed on the Q-trap 4000 mass spectrometer using the following conditions: source temperature, 160 °C; curtain gas, 10; nebulizer gas, 2; source voltage, 4200 V; and declustering potential, 120 V. The singly charged molecular ions of DESs(prop) (m/z 750.4), D4-DES(prop) (m/z 754.4), HP(prop) (m/z 597.3), D4-HP(prop) (m/z 601.3), and LP(prop) (m/z 581,3) were analyzed simultaneously using five alternative MS/MS experiments (0.2 s each). Transitions (42) Ma, S.; Lieberman, S.; Turino, G. M.; Lin, Y. Y. Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 12941–12943.
corresponding to the neutral loss of one propionyl group were selected for the precursor ions of DESs(prop) (m/z 694.4) and D4-DES(prop) (m/z 698.4), and transitions corresponding to the neutral loss of propylamide followed by the neutral loss of water were selected for the precursor ions of HP(prop) (m/z 506.3), D4-HP(prop) (m/z 509.3), and LP(prop) (m/z 490.3). Optimal sensitivity was obtained using the following conditions: Ecoll ) 58 eV, Coll gas ) 12, and declustering potential ) 120 eV for DESs(prop) and D4-DES(prop), and Ecoll ) 47 eV, Coll gas ) 12, and declustering potential ) 80 eV for HP(prop), D4-HP(prop), and LP(prop). Reproducibility and Recovery from Solid Phase Extraction. The protocols previously used to evaluate the reproducibility of the analytical method, the limit of detection, and the recovery after the solid phase extraction for DESs were applied to HP.20 RESULTS AND DISCUSSION DESs, HP, and LP Sample Preparation, Optimization, and Recovery. Pyridinium-based cross-linkers such as DES, HP, and LP present hydrophilic moieties, and separation of the native compounds by reverse phase chromatography is very challenging. Previous reports described the separation of these compounds on reverse phase chromatography using high concentration of ion pairing modifiers37,42 although the use of these modifiers impairs the electrospray ionization sensitivity. In an effort to alleviate these limitations, we previously evaluated the formation of N-propionyl derivatives for the analysis of urinary DESs in COPD patients.20 We used the same reaction conditions as part of this study to simultaneously derivatize DESs, D4-DES, HP, D4HP, and LP. Under these conditions the HP(prop) yield was greater than 98%. The only side reaction observed was the amidation of one carboxylic group of HP by the NH4OH used as catalyst. This reaction yielded a complete conversion of all free amines to their corresponding N-propyl derivative. No esterification of the carboxylic groups by iPr were observed under the present conditions. It is noteworthy that propionylation conducted using methanol led to the formation of methyl ester products representing up to 17% of the reaction products.20 The recovery of DESs and HP after the MCX solid phase extraction was evaluated using urine aliquots spiked with D4-DES and D4-HP before the extraction, which were then compared to urine samples spiked with the same amount of D4-DES and D4-HP after the extraction. Recoveries of 96 ± 3% were observed for each of the two cross-linkers. It is noteworthy that the loss of DESs or HP during the extraction is compensated for by an equivalent loss of the deuterated internal standards, while the measured DESs and HP concentrations remained unaffected. The internal standards D4-DES and D4-HP were added to the samples to compensate for sample loss during sample preparation and to increase the reproducibility of the nanoLC-MS/ MS analyses. The electronic resonance of the phenolate group from the pyridinium ring of HP decreases the acidity of the aromatic hydrogen and thus considerably reduces the exchange rate of deuterium compared to DES. For this reason, 7 days of reaction at 110 °C were required to increase the deuterium incorporation to D4-HP compared to 36 h at 70 °C for D4-DES.20 Figure 2 shows the isotopic profile of D4-HP after 0, 3, and 7 days of reaction. After 3 days of reaction, the Analytical Chemistry, Vol. 81, No. 22, November 15, 2009
9457
Figure 2. Deuterium incorporation in HP following reaction with 1,8diazabicyclo[5.4.0]undec-7-ene (DBU). The isotopic profile is shown for the native protonated HP and for the reaction products after 3 and 7 days of deuterium exchange.
deuterium exchange was not sufficient, and the proportion of D0-HP remaining in the standard corresponded to 4.7%. After 7 days of reaction, no detectable D0-HP remained in the standard, and the deuterium isotopomers corresponded to the following distribution D1-HP: 1.3%, D2-HP: 11.8%, D3-HP: 38.2%, and D4-HP: 48.7%. The stability of the deuterated standard was evaluated over a period extending 7 days, and no statistically significant change (
