Anal. Chem. 2007, 79, 8293-8300
Structural Characterization of Photodegradation Products of Enalapril and Its Metabolite Enalaprilat Obtained under Simulated Environmental Conditions by Hybrid Quadrupole-Linear Ion Trap-MS and Quadrupole-Time-of-Flight-MS Sandra Pe´rez, Peter Eichhorn, and Damia` Barcelo´*
Department of Environmental Chemistry, IIQAB-CSIC, Jordi Girona 18-26, Barcelona 08034, Spain
In the environment, organic micropollutants such as pharmaceuticals can be degraded via various biotic and abiotic transformation routes. In surface waters, for example, photodegradation may constitute a relevant natural attenuation process for drug residues that have been discharged from sewage treatment facilities. In the present work, the photochemical fate of the prodrug enalapril (376 Da, C20H28N2O5) and its active metabolite enalaprilat (348 Da, C18H24N2O5), a hypotensive cardioprotector previously reported to occur in contaminated rivers, was investigated in aqueous media under the influence of irradiation generated by a sunlight simulator. The experiments yielded three detectable photodegradates for enalapril (346 Da, 2 × 207 Da) whereas the photolysis of enalaprilat went hand in hand with the intermittent buildup of one photodegradate (304 Da). Fragmentation patterns of the parent compounds were established on a hybrid quadrupole-linear ion trap-mass spectrometer exploiting its MS3 capabilities. Accurate mass measurements recorded on a hybrid quadrupole-time-of-flight instrument in MS/MS mode allowed us to propose elemental compositions for the molecular ions of the degradates (346 Da, C19H26N2O4; 207 Da, C12H17NO2; 304 Da, C17H24N2O3) as well as of their fragment ions. Based on these complementary data sets from the two distinct mass spectrometric instruments, plausible structures were postulated for the four photodegradates. The compounds formed by enalapril corresponded to the loss of formaldehyde out of the proline residue (346 Da), cleavage of the central amide bond (207 Da) followed by migration of the ethylester side chain (207 Da) while decarboxylation of the free carboxylic acid was described for enalaprilat (304 Da). The study emphasized the potential of sunlight for breaking down an environmentally relevant drug and its metabolite. The environmental occurrence of pharmaceuticals has captured the attention of the scientific community because such * Corresponding author. E-mail:
[email protected]. Phone: ++34-93 400 6100 ext 435. Fax: ++34-93 204 5904. 10.1021/ac070891u CCC: $37.00 Published on Web 10/03/2007
© 2007 American Chemical Society
contaminants do not result primarily from manufacturing but from widespread and continual use in human and veterinary clinical practice. As these compounds exhibit an inherent biological activity, their release into the environment can adversely affect aquatic ecosystems and potentially impact drinking water supplies. In the industrialized countries, several tons of pharmacologically active substances reach wastewater treatment plants each year. In many instances, their moderate to high polarity in conjunction with poor biodegradability results in inefficient elimination and ultimately their discharge into receiving water bodies where drug residues at levels up to the microgram per liter range have been detected.1-3 Undoubtedly, the currently applied wastewater technologies relying on mechanical and biological treatment are inadequate for the extensive removal of many pharmaceuticals. Regarding the further whereabouts of drugs after discharge from sewage treatment facilities, knowledge on the fate and distribution in surface waters is still very fragmentary. It is worth taking into account that natural attenuation processes such as biodegradation and photodegradation can play an important role during the transport of pharmaceutical residues in rivers and streams.4,5 The breakdown of drugs into metabolites and degradates adds to the complexity of the spectrum of anthropogenic pollutants and thus will complicate the assessment of environmental hazards associated therewith.6 In particular, the identification of degradation products constitutes a major challenge with respect to improving the understanding of the environmental fate of pharmaceuticals.7 Recent advances in mass spectrometric instrumentation have provided the environmental scientist with highly valuable tools to gain deeper insight into biotic and abiotic (1) Hirsch, R.; Ternes, T.; Haberer, K.; Kratz, K. L. Sci. Total Environ. 1999, 225, 109-118. (2) Carballa, M.; Omil, F.; Lema, J. M.; Llompart, M.; Garcia-Jares, C.; Rodrı´guez, I.; Go´mez, M.; Ternes, T. Water Res. 2004, 38, 2918-2926. (3) Pe´rez, S.; Eichhorn, P.; Aga, D. S. Environ. Toxicol. Chem. 2005, 24, 13611367. (4) Doll, T. E; Frimmel, F. H. Chemosphere 2003, 52, 1757-1769. (5) Booren, A. L.; Arnold, W. A.; McNeill, K. Aquat. Sci. 2003, 65, 320-341. (6) Boxall, A.; Sinclair, C. J.; Fenner, K.; Kolpin, D. W.; Maund, S. J. Environ. Sci. Technol. 2004, 38, 369A-375A. (7) Yu-Chen Lin, A.; Reinhard, M. Environ. Toxicol. Chem. 2005, 24, 13031309.
Analytical Chemistry, Vol. 79, No. 21, November 1, 2007 8293
transformation processes of drugs.8-10 Studies about the photofate of pharmaceuticals in surface waters are reported.11,12 For instance, the analgesic drug diclofenac, which is only poorly amenable to biodegradation during activated sludge treatment of sewage, was reported to undergo photodegradation in rivers.13 That the study of the photolytic degradation of organic micropollutants in surface waters deserves further research efforts has been clearly illustrated by the case study on the antimicrobial compound triclosan, which upon direct photolysis was described to form the toxic substance 2,8-dichlorodibenzo-p-dioxin.14 In the present work, the photodegradation of the prodrug enalapril, an angiotensin converting enzyme inhibitor used in the treatment of hypertension and some types of chronic heart failure, and its active metabolite, enalaprilatsrecently detected in samples collected from sewage-impacted surface waters15,16swas studied in HPLC-grade water and in reconstructed freshwater. This is the first report describing the photofate of a drug along with its metabolite. For the simulation of environmental conditions, the experiments were carried out in a solar simulator that generated an emission wavelength spectrum similar to natural sunlight. For the structure elucidation of the degradates formed, a quadrupoletime-of-flight (QqToF) and a hybrid quadrupole-linear ion trap (QqLIT)-mass spectrometer were used. The combination of the high-resolution mass spectrometer providing accurate mass measurements on MS/MS data with a hybrid QqLIT instrument with MS3 capability proved to be very powerful in the identification of the unknown compounds. EXPERIMENTAL SECTION Chemical Standards. Enalapril maleate (CAS No. 76095-164) was purchased from Sigma Aldrich (St Louis, MO) and enalaprilat (CAS No. 76420-72-9) from USP (Rockville, MD). All organic solvents were Chromasol LC grade. Water was purchased from Sigma Aldrich; acetonitrile and methanol were from Riedel de Haen (Steinheim, Germany). Formic acid Suprapur (>98%) was obtained from Merck (Darmstadt, Germany). Photodegradation Experiments. Photodegradation experiments were conducted in a Suntest CPS simulator (Heraeus, Hanau, Germany) equipped with a xenon lamp. Using the appropriate window glass filters, the device emitted radiation across a wavelength spectrum similar to that of natural solar radiation. The test solutions were prepared in HPLC water and reconstructed standard freshwater containing 96 mg/L NaHCO3, 60 mg/L CaSO4‚2H2O, 60 mg/L MgSO4, and 4 mg/L KCl in distilled water, which resembled the composition of a moderately (8) Detomaso, A.; Mascolo, G.; Lopez, A. Rapid Commun. Mass Spectrom. 2005, 19, 2193-2202. (9) Eichhorn, P.; Aga, D. S. Anal. Chem. 2004, 78, 6002-6011. (10) Thurman, E. M.; Ferrer, I.; Furlong, E. T. A.C.S. Symp. Ser. 2003, 850, 128-144. (11) Chiron, S.; Minero, C.; Vione, D. Environ. Sci. Technol. 2006, 40, 59775983. (12) Latch, D. E.; Packer, J. L.; Stender, B. L.; VanOverbeke, J.; Arnold, U. A.; McNeill, K. Environ. Toxicol. Chem. 2005, 24, 517-525. (13) Agu ¨ era, A.; Pe´rez-Estrada, L. A.; Ferrer, I.; Thurman, E. M.; Malato, S.; Ferna´ndez-Alba, A. R. J. Mass Spectrom. 2005, 40, 908-915. (14) Latch, D. E.; Packer, L.; Arnold, W. A.; McNeill, K. J. Photochem. Photobiol. A: Chem. 2003, 158, 63-66. (15) Calamari, D.; Zuccato, E.; Castiglioni, S.; Bagnati, R.; Fanelli, R. Environ. Sci. Technol. 2003, 37, 1241-1248. (16) Vanderford, B. J.; Snyder, S. A. Environ. Sci. Technol. 2006, 40, 73127320.
8294
Analytical Chemistry, Vol. 79, No. 21, November 1, 2007
hard freshwater.13 The water samples spiked with enalapril or enalaprilat were irradiated in capped quartz (not airtight). Control solutions were kept in the dark and sampled at the same time points as those from the irradiation experiments. UPLC/ESI-QqToF-MS Analysis. Accurate mass MS and MS/MS analyses of enalapril, enalaprilat, and their photodegradation products were performed using a Waters/Micromass QqToF-Micro system coupled to a Waters Acquity UPLC system (Micromass, Manchester, UK). Samples from the photodegradation experiments were separated on a Waters Acquity BEH C18 column (50 × 2.1 mm, 1.7-µm particle size). The mobile phases were (A) water acidified with 0.3% formic acid and (B) acetonitrile. The flow rate was 400 µL/min, with a gradient starting at 95% A. After 1 min, the percentage of A was decreased linearly to 5% A within 6 min. This condition was held for 2 min. The initial mobilephase composition was restored within 0.1 min and maintained for column regeneration for another 1.9 min. The injection volume was 10 µL. The MS analysis was performed with an electrospray ionization (ESI) interface in the positive or negative ion mode with a capillary voltage of (3000 V. The cone gas flow was set to 50 L/h and the desolvation gas flow to 600 L/h. The instrument was operated at a resolution of 5000 (fwhm), and ESI mass spectra were acquired at 1-s intervals. High-resolution product ion spectra of the degradates were acquired using nitrogen as nebulizer and drying gas. Accurate mass measurements of the product ions were carried out in MS/MS mode by fragmenting the precursor ions with either low or high collision energies using argon as a collision gas at a pressure of ∼20 psi. All MS data handling was performed using the software package MassLynx V4.0. External mass calibration for positive and negative ESI modes was conducted prior to analysis for the mass range m/z 80-500 infusing a solution of acetonitrile/0.1 M NaOH/10% HCOOH (98:1:1) at a flow rate of 10 µL/min. Tyrosine-valine-tyrosine served as internal lock mass with [M + H]+ ) m/z 380.2185. HPLC/ESI-QqLIT-MS Analysis. The analysis of enalapril, enalaprilat, and their photodegradation products was carried out on an Agilent Series 1100 liquid chromatograph coupled to an API 4000 QTRAP mass spectrometer (Applied Biosystems/MSD Sciex, Foster City, CA). The chromatographic separations were achieved on a Waters XBridge C18 (100 × 2.1 mm, 3.5-µm particle size) equipped with a 10 × 2.1 mm guard column of the same packing material. The mobile phases were (A) water acidified with 0.3% formic acid and (B) acetonitrile. The flow rate was 200 µL/ min. The gradient was as follows: 0 min, 95% A; 0.5 min, 95% A; 4.5 min, 5% A; 5.5 min, 5% A; 5.6 min, 95% A; 10.5 min, 95% A. The injection volume was 10 µL. The Turbo Ion Spray source was operated in positive or negative ion mode for all compounds using the following settings for the ion source and mass spectrometer: curtain gas 30 psi, spraying gas 50 psi, drying gas 50 psi, drying gas temperature of 500 °C, and ion spray voltage of 4500 V. The declustering potential was 90, excitation energy was set at 60 V, and the collision energy was optimized for each compound. RESULTS AND DISCUSSION Photodegradation Profiles of Enalapril and Enalaprilat. With the objective of exploring the photodegradation of enalapril and enalaprilat in aqueous media using a sunlight simulator, HPLC-grade water and artificial freshwater were spiked with either test compound at 40 mg/L. In the ULPC/(+)-ESI-QqToF-MS
Figure 1. (A) UPLC/(+)-ESI-QqToF-MS chromatogram corresponding to an irradiated solution of reconstructed freshwater spiked with enalapril at 40 mg/L: (A) extracted ion chromatogram of m/z 377; (B) extracted ion chromatogram of m/z 208; (C) extracted ion chromatogram m/z 347.
chromatogram, obtained on a C18 reversed-phase column, a very broad peak with a relative (2.74 min) and an absolute maximum (3.28 min) was observed for enalapril (Figure 1A). This chromatographic phenomenon was attributable to the equilibrium between the cis and the trans conformer that arose from the hindered rotation around the amide bond17 having partial double bond character. Samples taken at various time points during irradiation were analyzed by UPLC/(+)-ESI-QqToF-MS monitoring a mass range from m/z 80 to 500. In the freshwater matrix, two major components with protonated molecules, [M + H]+, of m/z 208 emerged at 5.20 and 5.98 min, respectively, while a further peak with low intensity showed up at 4.40 min (m/z 347) (Figure 1B and C). As none of these signals was observed in the chromatograms corresponding to the control samples, they originated from photolysis of enalapril. The three photodegradates, which were also detectable in the experiment carried out in HPLCgrade water (data not shown), were referred to as D207-A, D207B, and D346, respectively. The semilogarithmic plot of the normalized peak area versus irradiation time is depicted in Figure 2A. After 40 h of irradiation, more than 90% of the parent compound had disappeared whereas the relative peak intensities of D207-A and D207-B reached ∼20% at their highest concentration. The degradate D346 in turn did not account for more than 0.1% of the enalapril signal intensity at t0. Screening of the irradiated samples in the negative ion mode over a mass range from m/z 80 to 500 did not provide any evidence for further photodegradates. As to the degradation profile of enalaprilat in the reconstructed freshwater, the decline in concentration followed first-order kinetics as obtained by UPLC/(+)-ESI-QqToF-MS analysis (Figure 2B). After an irradiation time of 63 h, ∼99% of the initially present (17) Trabelsi, H.; Bouabdallah, S.; Sabbah, S.; Raouafi, F.; Bouzouita, K. J. Chromatogr., A 2000, 871, 189-199.
enalaprilat had disappeared. Screening of the timed samples in positive and negative ion modes allowed to detect a single photodegradate (also found in the irradiated HPLC-grade water but not present in any of the control samples) with [M+H]+ of m/z 305 corresponding to a mass difference of -44 Da relative to the parent compound. Its broad peak with a relative (2.49 min) and an absolute maximum (2.72 min) was very similar in shape to the one of enalaprilat indicating that this photodegradate, denoted as D304, also existed as two conformational isomers (chromatogram not shown). The maximum concentration of D304 was determined in the sample taken at 24 h whereas by the end of the irradiation time, its level had dropped about tenfold. Apparently, D304 was subject to further breakdown. Fragmentation Pathways of Enalapril and Enalaprilat. The first step in elucidating the structures of the four photodegradates consisted of determining the fragmentation patterns of both parent compounds. The mass spectrometric characterization of the ethyl ester and its free carboxylic acid was carried out on the QqLIT system as its MS3 capability aided in establishing the fragmentation pathways. Upon comparing the (+)-ESI-MS2 spectra of enalapril (Figure 3A1) and enalaprilat (Figure 3B1) recorded in enhanced product ion (EPI) mode, two series of fragment ions could be distinguished (see also Figure 4, scheme A): One series comprised ions of identical m/z values for enalapril and enalaprilat, namely m/z 303 (7), 160 (5), 134 (3), and 117 (6). In the other series, pairs of analogous ions differing by m/z 28 (m/z 234/206 (2), 206/178 (4), and 130/102 (1)) could be discerned; these sets belonged to fragment ions containing the ester (enalapril, R ) CH3-CH2) or the acid functionality (enalaprilat, R ) H) (Figure 4A, scheme A). The latter series could be rationalized by cleavage of the bond between the carbon atom of the amide group and the R-carbon (m/z 234/206), followed by further loss of CH2dCH2 to yield m/z 206/178 or subsequent elimination of styrene to produce m/z 130/102 (Figure 4, scheme A). The intense fragment ion m/z 303 (7) that was observed for both enalapril and enalaprilat could be attributed to the loss of ethyl formiate (74 Da) and formic acid (46 Da), respectively, upon formation of a double bond in the backbone. That the loss of 46 Da from the protonated enalaprilat molecule occurred exclusively from the side chain, being attached to the R-carbon relative to the secondary amine, but not from the carboxylic acid of the proline residue being in R-position to the amide nitrogen, could be corroborated by recording the product ion spectra of the fragment ion m/z 303 in the QqLIT, i.e., the (+)-ESI-MS3 spectra of the sequence m/z 377 f 303 for enalapril and m/z 349 f 303 for enalaprilat (Figure 3A2 and B2). As these were identical in qualitative terms and also showed good agreement regarding relative peak intensities, the ion m/z 303 was confirmed to be devoid of the side chain bearing the R residues. The MS3 spectra of m/z 303 revealed the presence of several abundant fragment ions that were barely detectable or even absent in the MS2 spectra of the protonated parent compounds, namely m/z 257 (11) and 187 (8). Most unexpected, however, was the detection of the base peak ion at m/z 206 as its structure had to be distinct from the ion m/z 206 (9) which was detected previously in the MS2 spectra of enalapril (Figure 3A1) and enalaprilat (Figure 3B1). According to the fragmentation pattern shown in Figure 4, scheme A, this ion was proposed to bear the R residue (enalaprilat, 4; enalapril, Analytical Chemistry, Vol. 79, No. 21, November 1, 2007
8295
Figure 2. Degradation profiles of (A) enalapril and (B) enalaprilat in reconstructed freshwater as determined by UPLC/(+)-ESI-QqToF-MS.
Figure 3. (+)-ESI-QqLIT mass spectra of enalapril (A) and enalaprilat (B): (A1) MS2 [enhanced product ion], [M + H]+ m/z 377, (A2) MS3 m/z 377 f 303, (A3) MS3 m/z 377 f 234; (B1) MS2 [enhanced product ion], [M + H]+ m/z 349, (B2) MS3 m/z 349 f 303, (B3) MS3 m/z 349 f 206.
2). Assuming the formation of fragment ions with paired electrons, the even m/z value indicated the presence of a single nitrogen atom in this positively charged species. Yet no simple cleavage of the precursor ion’s backbone could be proposed that would have generated an ion with m/z 206 alongside a neutral moiety of 97 Da likewise having one nitrogen atom. Only an intermolecular rearrangement could explain this fragmentation behavior. A possible mechanism that would be consistent with the observations is depicted in scheme B of Figure 4. After an initial neutral loss of HCOORseither preceding the rearrangement or occurring in a concerted fashion with itsa transamidation takes place consisting of (a) nucleophilic attack of the secondary-amine nitrogen on the carbonyl of the proline carboxylic group, (b) migration of the hydroxyl group to the amide carbonyl, and (c) 8296 Analytical Chemistry, Vol. 79, No. 21, November 1, 2007
subsequent hydrolysis of the amide bond with concomitant liberation of an amine in the proline moiety. Departing from this structure for m/z 303 (7*), the cleavage of the amide bond would now afford the fragment ion m/z 206 (9) sought for. As the elemental composition of this ion (C12H15NO2) is identical to those of the two proposed fragment ions in the MS2 spectra (Figure 4, scheme A), even high-resolution mass spectrometry would not yield more than a single signal at m/z 206 in the product ion spectra of both enalapril and enalaprilat. With respect to the ion m/z 257 (11), an elimination of 46 Da from rearranged ion m/z 303 was rationalized as the loss of HCOOH. This resembled the sequence m/z 349 f 303 (Figure 3A2) in that the R-carbon atom of the carboxylic acid was attached to a nitrogen atom (Figure 4, scheme B). The fragment ion m/z 160 (5) in turn was proposed
Figure 4. Scheme A, proposed fragmentation pathway of enalapril (R ) CH3CH2) and enalaprilat (R ) H) under (+)-ESI conditions as derived from MS2 and MS3 experiments in the QqLIT mass spectrometer. In structures containing the residue R, the first m/z value corresponds to the ion derived from enalapril, and the second to the one of enalaprilat. The ion m/z 303 in scheme A is proposed to be subject to rearrangement (transamidation) according to the mechanism depicted in scheme B, which shows possible structures of the products ions observed in the MS3 spectrum of enalapril (m/z 377 f 303) and enalaprilat (m/z 377 f 349).
to have the same structure as m/z 160 in scheme A (Figure 4) and was attributed to the breakage of the amide bond in m/z 257 whereas cleaving off the phenyl-bearing residue from the amidenitrogen yielded either m/z 117 (6) or 187 (8). The (+)-ESI-MS2 data for enalapril and enalaprilat recorded on the QqToF system showed the formation of the same set of fragment ions as observed in the QqLIT instrument with the exception of the low-intensity ion m/z 232 and the pair m/z 206/ 178, which were not detectable in the product ion profile obtained on the ToF instrument (mass spectra are provided in Figure A of the Supporting Information). The good agreement in mass spectral quality between the two instruments is due to the comparable CID fragmentation processes taking place in their respective collision cells (q). Minor differences are likely related to events occurring in the region between the exit of the collision cell and the detector. In this respect, the two instruments differ in their
MS/MS modes in the way the second mass analyzer is operated: In the QqToF, all product ions leaving the collision cell are, after an orthogonal acceleration, separated as a function of their massspecific flight times, whereas the Q3 of the QqLIT serves as a trapping device for intermediate storage of the product ions, with the aim of affording increased sensitivity in this so-called EPI mode, prior to scanning them out. Seemingly, the rearrangement as postulated in scheme B of Figure 4 took place to only a minor extent in Q3 during the EPI scan but became a dominant process in the MS3 scans of m/z 377 f 303 and 349 f 303, which involved isolation, activation, and fragmentation of the precursor ion m/z 303. This hypothesis was in line with the absence of the ions m/z 232 and 206 (enalapril) and m/z 232 and 178 (enalaprilat) in their MS2 spectra when using the conventional product ion mode of the QqLIT. Under these conditions, fragment ions generated in the collision cell were not intermediately stored in Q3 to gain Analytical Chemistry, Vol. 79, No. 21, November 1, 2007
8297
Figure 5. (+)-ESI-QqToF-MS2 spectra of photodegradation products of enalapril (A) and enalaprilat (B): (A1) protonated D207-A, [M + H]+ m/z 208, (A2) protonated D207-B, [M + H]+ m/z 208, (A3) protonated D346, [M + H]+ m/z 347; (B1) protonated D304, [M + H]+ m/z 305. The mass spectra of D207-A, D207-B, and D347 were acquired at a collision energy of 20 eV; collision energy for degradate D304 was 30 eV.
sensitivity (EPI mode) but were immediately subjected to mass separation in the Q3 analyzer (data not shown). Concerning the fragmentation pathways of enalapril and enalaprilat as presented in scheme A of Figure 4, the proposed structures of the fragment ions, i.e., their elemental compositions, were consistent with the results from the accurate mass measurements performed on the QqToF-MS in product ion scan mode. The relative errors of precursor and product ions as determined by triplicate analysis were typically below 5 ppm for most of the ions (Table 1). Identification of the Photodegradates of Enalapril. The two major photodegradation products of enalapril, which were detected in the positive ion mode at retention times of 5.20 and 5.98 min, yielded the (+)-ESI-QqToF-mass spectra given in Figure 5A1 and A2. Both product ion profiles contained fragment ions with nominal masses m/z 117 and 134; these were identical in elemental composition to the ions m/z 117 (C9H12N) and 134 (C9H9) observed in the QqToF-MS2 spectrum of enalapril (Tables 1 and 2). This indicated that the left part of the molecule comprising the aromatic ring was intact in both degradates. Based on the proposed elemental composition of the protonated D207-A (Table 2), C12H18NO2 (-4.3 ppm), a plausible structure corresponded to the cleavage between the nitrogen of the secondary amine and the R-carbon of the amide group. The presence of a primary amine in D207-A facilitated the neutral loss of ammonia resulting in m/z 191 (C12H15O2, -2.0 ppm) while the ion m/z 180 (C10H14NO2, +1.5 ppm) was identified as corresponding to the elimination of ethylene (Table 2) in the ester side chain, accomplished through a McLafferty rearrangement within this moiety; the loss of both ammonia and ethylene resulted in the formation of m/z 163 (C10H11O2, +1.5 ppm). With respect to the photodegradate D207-B, accurate mass measurements on the QqToF instrument (Table 2) suggested that its elemental composition was identical to the one of D207-A 8298 Analytical Chemistry, Vol. 79, No. 21, November 1, 2007
Table 1. Accurate Mass Measurements of Enalapril and Enalaprilat As Determined by UPLC/ (+)ESI-QqToF-MS in MS/MS Mode mass (m/z) (fragment) elemental ion composition
calcd
meada
rel error (SD) (ppm)
DBEb
[M + H]+ m/z 303 m/z 234 m/z 160 m/z 134 m/z 130 m/z 117 m/z 102 m/z 91
C20H29N2O5 C17H23N2O3 C14H20NO2 C11H14N C9H12N C6H12NO2 C9H9 C4H8NO2 C7H7
Enalapril 377.2076 377.2070 -2.7 (1.4) 303.1709 303.1707 -0.7 (2) 234.1494 234.1498 +1.6 (1.1) 160.1126 160.1120 -3.9 (3.2) 134.0970 134.0967 -2.0 (2.9) 130.0868 130.0854 -10.8 (2.0) 117.0704 117.0697 -6.3 (3.5) 102.0555 102.0546 -8.8 (4.2) 91.0548 91.0549 -0.7 (17.8)
7.5 7.5 5.5 5.5 4.5 1.5 5.5 1.5 4.5
[M + H]+ m/z 303 m/z 206 m/z 160 m/z 134 m/z 117 m/z 102 m/z 91
C18H25N2O5 C17H23N2O3 C12H16NO2 C11H14N C9H12N C9H9 C4H8NO2 C7H7
Enalaprilat 349.1763 349.1765 -0.2 (2.1) 303.1709 303.1708 +1.5 (3.8) 206.1181 206.1184 +1.5 (0) 160.1126 160.1130 +2.4 (1.6) 134.0970 134.0960 -7.6 (0.6) 117.0704 117.0695 -7.6 (0.8) 102.0555 102.0549 -6.4 (0.8) 91.0548 91.0538 -10.4 (3.24)
7.5 7.5 5.5 5.5 4.5 5.5 1.5 4.5
a
N ) 3.
b
Double bond equivalents.
(C12H18NO2 for the protonated compound, -4.3 ppm); i.e. it had to be a constitutional isomer. The absence of a fragment ion at m/z 191 in the MS2 spectrum of D207-B (Figure 5A2) indicated that the neutral loss of ammonia from the protonated precursor ion was not feasible. Thus, the presence of a primary amine in this molecule could be ruled out. On the other hand, the compound had to include a structural element that could be easily cleaved off to yield the fragment ion m/z 134. This ion had been detected and identified previously in the MS2 spectrum of both
Table 2. Accurate Mass Measurements of the Photodegradation Products of Enalapril (D207-A, D207-B, D346) and of Enalaprilat (D304) As Determined by UPLC-(+)ESI-QqToF-MS in Product Ion Scan Mode mass (m/z) proposed (fragment) elemental ion composition [M+H]+ m/z 273 m/z 234 m/z 160 m/z 134 m/z 130 m/z 117 [M+H]+ m/z 191 m/z 180 m/z 163 m/z 145 m/z 134 m/z 117 m/z 107 m/z 102 m/z 91 [M+H]+ m/z 134 m/z 117 m/z 91 [M+H]+ m/z 162 m/z 117 m/z 91 a
calcd
measda
Degradate D346 347.1971 347.1985 273.1603 273.1594 234.1494 234.1504 160.1126 160.1120 134.0970 134.0974 130.0868 130.0868 117.0704 117.0702 Degradate D207-A C12H18NO2 208.1338 208.1330 C12H15O2 191.7072 191.1068 C10H14NO2 180.1025 180.1027 C10H11O2 163.0759 163.0762 C10H9O 145.0653 145.0651 C9H12N 134.0970 134.0973 117.0704 117.0707 C9H9 C8H11 107.0861 107.0853 C4H8NO2 102.0555 102.0559 C7H7 91.0548 91.05443 Degradate D207-B C12H18NO2 208.1338 208.1330 C9H12N 134.0970 134.0971 117.0704 117.0707 C9H9 C7H7 91.0548 91.0545 Degradate D304 C17H25N2O3 305.1865 305.1879 C11H16N 162.1283 162.1277 117.0704 117.0709 C9H9 C7H7 91.0548 91.0544 C19H27N2O4 C16H21N2O2 C14H20NO2 C11H14N C9H12N C6H12NO2 C9H9
rel error (SD) (ppm) DBEb +4.0 (1.9) -3.4 (1.8) +4.3 (1.7) -3.9 (2.1) +3.2 (2.1) -0.4 (3.8) -1.9 (1.8)
7.5 7.5 5.5 5.5 4.5 1.5 5.5
-4.2 (0.33) -2.0 (2.4) +1.5 (1.0) +1.5 (5.6) -2.0 (1.4) +2.7 (1.4) +5.4 (1.2) -7.3 (2.3) +7.6 (0.5) -8.7 (2.1)
4.5 5.5 4.5 5.5 6.5 4.5 5.5 3.5 1.5 4.5
-4.3 (0.55) +1.2 (3.3) +6.6 (1.6) -8.0 (3.4)
4.5 4.5 5.5 4.5
+4.3 (0.9) -3.5 (4.3) +7.1 (0.8) -9.8 (1.7)
6.5 4.5 5.5 4.5
N ) 3. b Double bond equivalents.
enalapril and enalaprilat (for structure, see scheme A in Figure 4). A migration of the ester side chain in D207-A to the nitrogen atom was put forward to explain the formation of the carbamate structure of the degradate D207-B (Figure 5A2). As the weakest bond in this molecule was thought to be the nitrogen-carbon bond of the carbamate, the formation of m/z 134 was favored over other fragmentation routes. This resulted in a rather simple fragmentation pattern of D207-B in the positive ion mode. As far as the photodegradate D346 is concerned, the findings of the accurate mass measurements on the QqToF machine (Table 2) were indicative of a loss of CH2O from the parent compound (for the protonated species: C20H29N2O5 f C19H27N2O4). Comparison of the MS2 spectra of enalapril and D346 (Figure 5A3) revealed that all fragment ions with m/z e234 were identical for both compounds, allowing us to confirm that the structural modification had to be beyond the amide group. The most plausible explanation was attributed to the loss of formaldehyde out of the carboxylic group of the proline residue. This photolytic reaction has been described for other aliphatic carboxylates such as, for example, the non-steroidal anti-inflammatory drug naproxen.18 In enalapril the conversion of the acid functionality into a ketone was also in agreement with a stronger retention of D346 as compared to enalapril on the RP column. (18) Isidori, M.; Lavorgna, M.; Nardelli, A.; Parella, A.; Previtera, L.; Rubino, M. Sci. Total Environ. 2005, 348, 93-101.
Identification of the Photodegradate of Enalaprilat. Accurate mass measurements of the protonated D304 and of its fragment ions were performed on the QqToF instrument in the (+)-ESI-MS/MS mode (Table 2). In the search for an elemental composition, the best hit (+4.3 ppm) corresponded to the loss of CO2 from enalaprilat (C18H25N2O5 f C17H25N2O3 for the protonated species). Whether this leaving group originated from the carboxylic acid of the proline residue or from the one in the aliphatic chain could be unequivocally determined based on the product ion spectrum. In comparison to the (+)-ESI-MS2 spectrum of the parent compound (see Figure 3B1), D304 displayed a rather simple fragmentation pattern characterized by an intense ion at m/z 162 (Figure 5B1). That this neutral loss of 143 Da corresponded to the cleavage of the bond between the amide carbon and the R-carbon was corroborated through the accurate mass of m/z 162, which showed a relative error of -3.5 ppm for the composition C11H16N (Table 2). Moreover, the product ion spectrum of D304 was consistent with the structure proposed in Figure 5B1 because it did not exhibit fragment ions at m/z 160 and 134. These had been observed in the MS2 profiles of both enalapril and enalaprilat (see Figure 4, scheme A); their formation involved the elimination of ethyl formiate and formic acid, respectively, upon formation of a double bond in the backbone of the molecule. The absence of the carboxylic acid functionality in D304 excluded the formation of m/z 160 and 134. Implications. Enalapril and its pharmacologically active metabolite enalaprilat were demonstrated to be amenable to direct photolysis by sunlight-like radiation. These findings constitute important information aiding in predicting the environmental fate of these two drugs. In general terms, investigations into the photofate of not only intact pharmaceuticals but also their metabolites are crucial for the understanding of their environmental impact. As to the analytical approach employed, the use of HPLC/ QqLIT-MS and UPLC/QqToF-MS proved to be a powerful combination for elucidating the unknown structures of the phototransformation products. The method involved comparison of the fragmentation patterns of the parent compounds and their degradation products obtained in MS2 and MS3 experiments in the QqLIT system followed by confirmation of the proposed elemental compositions of the fragment ion through accurate mass measurements with the QqToF mass spectrometer. In comparison to a simple ToF instrument, the former configuration offers higher selectivity and thus accuracy in the analysis of fragment ions, which in the latter case can only be generated by in-source dissociation. This advantage can turn out to be critical in analyzing fragment ions of chromatographically unresolved compounds as frequently encountered in samples of natural origin. On the other hand, the gain in selectivity afforded by the quadrupole mass filter in the QqToF is in part achieved at the expense of sensitivity. With respect to the hybrid QqLIT, it is capable of performing MS3 experiments thereby generating highly valuable information required for structural elucidation of unknowns. Although it does not proceed further in the multiplestage fragmentation processsthis feature is still confined to conventional (linear) ion trapssits triple-quadrupole architecture makes it at the same time a very sensitive and selective instrument for the quantitative determination of target analytes. Analytical Chemistry, Vol. 79, No. 21, November 1, 2007
8299
ACKNOWLEDGMENT The work presented in this article was supported by the EU Project EMCO-INCO-CT-2004-509188) and by the Spanish Ministerio de Educacio´n y Ciencia, Project EVITA (CTM2004-06255CO3-01). This work reflects only the author’s views and the European Community is not liable for any use that may be made of the information contained therein. S.P. acknowledges a postdoctoral contract from I3P Program (Itinerario Integrado de Insercio´n Profesional); co-financed by CSIC and European Social
8300
Analytical Chemistry, Vol. 79, No. 21, November 1, 2007
Funds. S.P. acknowledges Roser Charler and Dori Fanjul, for their support with the MS instrumentation. SUPPORTING INFORMATION AVAILABLE Additional information as noted in text. This material is available free of charge via the Internet at http://pubs.acs.org. Received for review May 1, 2007. Accepted August 2, 2007. AC070891U
