Application of a Linear Ion Trap/OrbitrapMass Spectrometer in MetaboliteCharacterization Studies: Examinationof the Human Liver Microsomal Metabolismof the Non-Tricyclic Anti-DepressantNefazodone Using Data-DependentAccurate Mass Measurements

Scott M. Peterman and Nicholas Duczak, Jr.Thermo Electron Corporation, Somerset, New Jersey, USA

Amit S. Kalgutkar, Mary E. Lame, and John R. SogliaPharmacokinetics, Dynamics, and Metabolism Department, Pfizer Global Research and Development,Groton, Connecticut, USA

We report herein, facile metabolite identification workflow on the anti-depressant nefazodone,which is derived from accurate mass measurements based on a single run/experimentalanalysis. A hybrid LTQ/orbitrap mass spectrometer was used to obtain accurate mass full scanMS and MS/MS in a data-dependent fashion to eliminate the reliance on a parent mass list.Initial screening utilized a high mass tolerance (�10 ppm) to filter the full scan MS data forpreviously reported nefazodone metabolites. The tight mass tolerance reduces or eliminatesbackground chemical noise, dramatically increasing sensitivity for confirming or eliminatingthe presence of metabolites as well as isobaric forms. The full scan accurate mass analysis ofsuspected metabolites can be confirmed or refuted using three primary tools: (1) predictivechemical formula and corresponding mass error analysis, (2) rings-plus-double bonds, and (3)accurate mass product ion spectra of parent and suspected metabolites. Accurate masscharacterization of the parent ion structure provided the basis for assessing structuralassignment for metabolites. Metabolites were also characterized using parent product ion m/zvalues to filter all tandem mass spectra for identification of precursor ions yielding similarproduct ions. Identified metabolite parent masses were subjected to chemical formulacalculator based on accurate mass as well as bond saturation. Further analysis of potentialnefazodone metabolites was executed using accurate mass product ion spectra. Reported massmeasurement errors for all full scan MS and MS/MS spectra was �3 ppm, regardless ofrelative ion abundance, which enabled the use of predictive software in determining production structure. The ability to conduct biotransformation profiling via tandem mass spectrom-etry coupled with accurate mass measurements, all in a single experimental run, is clearly oneof the most attractive features of this methodology. (J Am Soc Mass Spectrom 2006, 17,363–375) © 2006 American Society for Mass Spectrometry

Metabolite identification constitutes an impor-tant paradigm at various stages of drug dis-covery and development. In the early discov-ery stage, it is crucial to identify metabolic soft spots inlead chemical series to provide feedback to medicinalchemists for further optimization of the lead com-

Published online January 25, 2006Address reprint requests to Dr. A. Kalgutkar, Pharmacokinetics, Dy-namics, and Metabolism Department, Pfizer Global Research and De-

velopment, Eastern Point Road, Groton, CT 06340, USA. E-mail:amit.kalgutkar@pfizer.com

© 2006 American Society for Mass Spectrometry. Published by Elsevie1044-0305/06/$32.00doi:10.1016/j.jasms.2005.11.014

pounds. Metabolite identification studies at this stageare usually designed to avoid metabolic instability thatmay lead to extensive first pass metabolism and poororal bioactivity [1]. In addition, early metabolite identi-fication studies are also extensively utilized to minimizepotential safety concerns that may arise from bioactiva-tion of lead compounds to electrophilic reactive inter-mediates [2]. For a drug candidate in preclinical devel-opment an early understanding of the in vivo metabolicfate is essential since metabolites can contribute to

pharmacologic action or possess toxicological conse-

r Inc. Received September 29, 2005Revised November 21, 2005

Accepted November 21, 2005

364 PETERMAN ET AL. J Am Soc Mass Spectrom 2006, 17, 363–375

quences independent of the parent. Owing to its selec-tivity, sensitivity, and speed of analysis with minimalsample processing, liquid chromatography tandemmass spectrometry (LC-MS/MS) has become themethod of choice for metabolite identification in the fastpaced environment of drug discovery/development[3]. Crude extracts from complex biological matricescan be subject to metabolite identification studies di-rectly on a LC-MS/MS. Complex metabolite profilescan be chromatographically separated on an HPLCcolumn and full scan MS and product ion scan MS/MSdata generated on-line. The molecular weight of metab-olites and localization of biotransformation sites can beelucidated based on interpretation of MS/MS dataobtained by collision-induced dissociation (CID) of theparent structure and the various metabolites.

Structural elucidation of metabolites is not alwaysstraightforward, particularly in cases of complex re-arrangement of chemical architecture leading tounique metabolites whose structure(s) cannot be elu-cidated by the usual LC-MS/MS methodology [3–5].Likewise, the presence of metabolites with isobaricmolecular ions and/or competing fragment ions inparent drug MS confound precise determination offragmentation pattern in the parent drug and itsmetabolites [3, 6]. In the latter case, competing frag-mentation pathways in parent and metabolite massspectra can afford product ions separated by as littleas 0.04 to 0.1 Da. Using low-resolution mass spec-trometry, these minute differences in observed frag-ment ion masses often preclude an unambiguouselucidation of fragment ion structure and conse-quently metabolite structure. In some instances, theensuing dogma may be resolved via synthesis ofstable isotope (e.g., 13C, 2H, 15N) derivatives of theparent molecule [7]. However, such strategies mayhinder progress in a fast paced drug discovery envi-ronment, since additional efforts/resources must bededicated towards the synthesis of stable isotopeanalogs, which can often require multistep synthesis.Supplementing low-resolution MS with accuratemass measurements in metabolite identification stud-ies can provide valuable information in distinguish-ing isobaric molecular ions and provide unequivocalproof of correct fragmentation patterns, resulting in asignificant increase in confidence towards predictionsof metabolite structure [8, 9]. While time-of-flight(TOF) instrumentation is often used to acquire accu-rate mass information [10, 11], the methodologyrequires numerous LC runs to acquire accurate masson multiple product ions and can be potentiallyviewed as time and cost prohibitive in a typical drugdiscovery scenario. Difficulties associated with usingTOF technology are attributed to a dependence oninternal calibrants or lock masses for obtaining highmass accuracy measurements or recalibration off-line. In addition, mass accuracy may be compromised

at extremely low relative ion intensities resulting in

missed diagnostic ions that can be used to furtheridentify metabolites.

Recently, the combination of Orbitrap technologywith a linear ion trap has been shown to enable fast,sensitive and reliable detection and identification ofsmall molecules regardless of relative ion abundance[12, 13]. The orbitrap mass analyzer employs the trap-ping of pulsed ion beams in an electrostatic quadro-logarithmic field. This field is created between an axialcentral electrode and a coaxial outer electrode. Stableion trajectories combine rotation around the centralelectrode with harmonic oscillations along it. The fre-quencies of axial oscillations and hence mass-to-chargeratios of ions are obtained using fast Fourier transformof the image current detected on the two split halves ofthe outer electrode. The hybrid linear ion trap/orbitrapmass spectrometer is related to the Finnigan LTQ FT incombining fast scan speeds and high mass accuracymeasurements (within a few ppm) for precursor andproduct ions on an LC timescale which is critical formetabolic profiling. Furthermore, external calibrationcan be used to obtain high mass accuracy (�3 ppm) ofall measured spectra resulting in a simplified experi-mental protocol.

As a demonstration of the capability of this de Novotechnology, we re-examined the metabolic pathways ofthe non-tricyclic anti-depressant drug nefazodone inNADPH-supplemented human liver microsomes utiliz-ing accurate mass measurements of precursor andMS/MS product ions for the parent drug and its me-tabolites [14 –17]. Data-dependent MS/MS analyseswere conducted on a hybrid linear ion trap/orbitrapmass spectrometer that enabled high-resolution/highmass accuracy measurements on precursors and prod-uct ions regardless of relative ion abundance and on aLC-timescale [12, 13, 18]. Furthermore, increased massaccuracy measurements helped in assigning fragmention structure for all anticipated and novel metabolites ofthe drug.

Experimental

Materials

Nefazodone hydrochloride and NADPH were pur-chased from Sigma-Aldrich (St. Louis, MO). Humanliver microsomes pooled from 53 individual donorswere purchased from BD Gentest (Woburn, MA).

Microsomal Incubations

Stock solutions of nefazodone were prepared in meth-anol. The final concentration of methanol in the incu-bation media was 0.2% (vol/vol). Incubations werecarried out at 37 °C for 60 min in a shaking water bath.The incubation volume was 1 mL and consisted of thefollowing: 0.1 M potassium phosphate buffer (pH 7.4),human liver microsomes (P450 concentration � 0.5

�M), NADPH (1.2 mM), and nefazodone (10 �M). The

365J Am Soc Mass Spectrom 2006, 17, 363–375 ORBITRAP MS IN METABOLITE IDENTIFICATION STUDIES

reaction mixture was prewarmed at 37 °C for 2 minbefore adding NADPH and incubations were termi-nated by the addition of ice-cold acetonitrile (1 mL). Thesolutions were centrifuged (3000 g for 15 min) [16], andthe supernatants were dried under a steady nitrogenstream. The residue was reconstituted with mobilephase and analyzed for metabolite formation.

LC-MS/MS Assay for Identification of NefazodoneMetabolites

Metabolites were identified by a Finnigan LTQ orbitrapmass spectrometer (Thermo Electron, Bremen, Ger-many) and all ion acquisition was performed using theorbitrap mass spectrometer. The LC system consisted ofa Hypersil Gold (1.9 �M C18 50 � 2.1 mm) column, aFinnigan Surveyor LC system comprising of an HPLCwith a built-in degasser, autosampler, and a FinniganLTQ orbitrap mass spectrometer. A constant flow rateof 0.27 mL/min was used with a gradient from 3–95% Bin 35 min before column re-equilibration. Solvent A was10 mM ammonium formate (0.1% formic acid) andSolvent B was acetonitrile (100%). The scan event cycleused a full scan mass spectrum at a resolving power of15,000 (at m/z 400) and three corresponding data-depen-dent MS/MS events acquired at a resolving power of7500. The three most intense ions detected during fullscan MS triggered data dependent scanning. Data de-pendent scanning was performed without the use of aparent ion list. The microscan count for full and MS/MSscan events were set to unity and a repeat count fordynamic exclusion was set to three for verification.MS/MS activation parameters used an isolation widthof 2 Da, normalized collision energy of 35%, and anactivation time of 30 milliseconds. Mass accuracy wasdetermined from an external calibration that was per-formed the prior day.

Data Interpretation

All data were processed using Qual Browser (ThermoElectron, San Jose, CA), which provides accuratemass thresholds to filter data. In addition, chemicalformula calculator was used to provide chemicalformula and saturation values (ring-plus-double-bonds [RDB]) for product ions of nefazodone and itsmetabolites [19]. Fragmentation patterns were gener-ated using Mass Frontier 4.0 (Thermo Electron) [20,21]. Predictive fragmentation pattern was derivedfrom plausible protonation sites, subsequent isomer-ization and even electron species, and bond satura-tion (i.e., RDB). Comparisons between theoretical andexperimental product ion spectra further aided in theidentification of metabolite structures and site(s) of

biotransformation with the parent drug.

Results and Discussion

Scheme 1 depicts the observed molecular ions [M � H] �,retention times [Rt, min], and calculated RDB values fornefazodone and its metabolites. Figure 1 shows the totalion chromatogram of a NADPH-supplemented humanliver microsomal incubation of nefazodone (10 �M) that isbaseline subtracted from control experiments lackingNADPH co-factor. In addition to unmetabolized nefaz-odone, which eluted at 13.36 min, there are numerouslower intensity peaks, which can be attributed to metab-olites derived from the oxidation of nefazodone. Metabo-lite profiling studies were initiated following a thoroughinvestigation of the mass spectral fragmentation pattern ofthe nefazodone parent molecule considering that determi-nation of precise fragmentation pattern of the parentmolecule enhances confidence towards predicting bio-transformation site(s) on the corresponding metaboliteproduct ion.

Figure 2 shows a product ion spectrum of nefaz-odone resulting from one total scan comprised of asingle microscan. The total time for product ion spec-trum acquisition was 340 milliseconds. The product ionspectrum is dominated by the fragment ion m/z 274(measured ion intensity �2.1 � e07 counts and masserror of �1.4 ppm) corresponding to the phenoxyethyl-5-ethyltriazolone propyl motif. In addition, there areseveral low intensity product ions that are consistentwith fragmentation spectra predicted from Mass Fron-tier 4.0. The structures and corresponding mass errorsfor these low intensity fragment ions are presented inFigure 2. It is interesting to point out that despite largedifferences in measured ion intensity between productions, the measured mass accuracy are quite comparable.For example, there is ca. a 100-fold difference betweenthe measured ion intensities of the product ions at m/z140 (5.7 � e05) and at m/z 274 (2.1 � e07), yet thedifferences in the measured mass accuracy are fairlycomparable (�2.8 and �1.4 ppm for the product ions atm/z 140 and 274, respectively). In addition, the inset inFigure 2 shows the lowest intensity product ions de-tected from the single product ion spectrum. The prod-uct ion at m/z 197 has a measured ion intensity of 8.2 �e03 and a measured mass error of 4.5 ppm. This repre-sents almost a 3000-fold difference in measured ionintensities without a significant compromise in production assignment.

The structure of nefazodone presents a challenge forinterpretation of the product ion spectra due to isobaricproduct ions that can be predicted to arise, based on siteof protonation and subsequent bond scission. For ex-ample, the N-alkylsubstituted nitrogen atom in thepiperazine ring (Figure 2) provides a basic center forcharge retention that could potentially lead to multiplebond scissions before observed product ion formation.The competing product ions may have isobaric mass-to-charge values separated by 0.05 Da (ca. 200 ppm),and thus, one would need to rely on accurate mass

measurements for correct prediction of fragment ion

e an

366 PETERMAN ET AL. J Am Soc Mass Spectrom 2006, 17, 363–375

structure. Figure 3 depicts the competing pathwayspredicted by Mass Frontier 4.0 that would give rise tothe product ion at m/z 180. Assuming protonation isoccurring on the piperazine nitrogen shown in Figure 3,the product ion of the “left” side of the nefazodonestructure would possess a mass of 180.0575 Da (Figure3, Pathway A) while the product ion that arises from theright hand side of the molecule will possess a mass of180.1131 Da (Figure 3, Pathway B). The measured massof 180.1128 unambiguously proves that cleavage on theright hand side of the nefazodone structure is the mostlikely route to the m/z 180 product ion given that theresulting mass error is only 2 ppm compared to 307ppm for the product ion resulting from cleavage on theleft hand portion of nefazodone (Table 1). In addition tothe product ion at m/z 180, nefazodone fragmentationaffords other product ions that can be assigned multiplestructures that differ by less than 0.1 Da. Table 1 lists theproduct ions observed following nefazodone collisionalactivation. The experimentally observed mass-to-chargevalues are listed with the predicted product ions fromMass Frontier 4.0. Note that five out of the eightproduct ions have predicted product ions with charge

N

Cl

N

Nef

Rt R

HO

m/z 21

Rt = 2.81 RDB = 5.

N

Cl

N NNH

NO

O

O

m/z 458

Rt = 12.16 minRDB = 11.5

M7 N

Cl

N

O

HO

m/z 486

Rt = 11.42 minRDB = 11.5

M6

N

Cl

N NN

NO

O

OH

m/z 486

Rt = 12.25 minRDB = 11.5

M8

N

Cl

N NN

NO

O

O

m/z 484

Rt = 13.59 minRDB = 12.5

M9

M1

Scheme 1. Metabolic pathways of th

retention on each side of the nefazodone structure, but

with accurate mass measurements, fragmentation pat-tern is accurately determined in each of these cases.

Figure 4 depicts a series of extracted ion chromato-grams for various nefazodone metabolites. The ob-served molecular ions and fragment ion spectra of thesemetabolites are consistent with those previously re-ported in the in vivo and in vitro metabolism studies onnefazodone in humans [14 –17]. Individual metaboliteions were initially extracted from the total ion chro-matogram using m/z values for nefazodone metabolitesfrom previously published reports. A mass threshold of10 ppm was used to determine “real” metabolite signal.The benefits of using a high mass tolerance dramaticallyimproved signal-to-noise ratio and allowed an easydetermination of the presence of isomeric metabolites aswell as metabolites which were not predicted fromMass Frontier 4.0 and/or theoretical reasoning. Further-more, by reducing the background signal from eachextracted ion chromatographic trace, a difference insignal intensity of ca. 2800 times is observed betweenthe greatest integrated peak area for the nefazodoneparent ion and the minor peak for the metabolite withm/z at 484 (mass error � 1.7 ppm) that eluted at 12.9

NN

NO

O

one70

6 min11.5

NH

N

Cl

NH

m/z 197

Rt = 8.58 minRDB = 5.5

M2

NN

NO

O

HO

m/z 292

Rt = 10.65 minRDB = 7.5

M3

NN

NO

O

HO

m/z 306

Rt = 10.83 minRDB = 8.5

M4ON

N

NO

O

HO

m/z 306

Rt = 11.01 minRDB = 8.5

M5

O

O

ti-depressant nefazodone in humans.

azodm/z 4

= 13.3DB =

N

Cl

3

min5

NN

N

min. The addition of m/z 14 to the nefazodone parent

rials

367J Am Soc Mass Spectrom 2006, 17, 363–375 ORBITRAP MS IN METABOLITE IDENTIFICATION STUDIES

molecular weight is consistent with introduction of acarbonyl functionality within the molecule and thismetabolite appears to be an isomer of M9. In addition tothis set of isomeric metabolites of nefazodone, theextracted ion chromatographic traces also indicated the

0 1 2 3 4 5 60

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100

Rel

ativ

e A

bu

nd

ance

Figure 1. Base peak chromatogram of a NADPHof nefazodone (10 �M) as described under Mate

150 200 250 300m/z

0

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100

Rel

ativ

e A

bu

nd

ance

274.15463

246.12343

140.08145

180.11278

237.11478 3

Rel

ativ

e A

bu

nd

ance

-2.8 ppm

-1.1 ppm

-1.4 ppm

NHN

NO N

N

NO

O

NN

NO

O

Figure 2. Full scan accurate mass spectrum ofspectrometer using external mass calibration. Tintensity product ions. Product ion assignment

Frontier 4.0.

possible presence of isomeric metabolites for m/z 306(Figure 4e) and m/z 486 (Figure 4f).

Figure 5 shows comparative product ion spectra forthe isomeric monohydroxy metabolites of nefazodone(m/z 486), which elute at 11.42 (Figure 5a) and 12.25 min

Nefazodone[M + H]+ = 470

8 9 10 11 12 13 14 (min)

13.36

14.36

8.90

10.23

12.2511.427.91 9.17 10.65

8.589.26 13.00

M2

M6

M4

M5

M3

M8

M7

M9

plemented human liver microsomal incubationand Methods.

350 400 450

55

0 160 170 180 190 200 210 220 230 240m/z

180.11278

54.09712

237.11478

218.09167

197.08311

152.081164.5 ppm

4.5 ppm

3.4 ppm

2.2 ppm

NH

NN

NO

O

NH

NN

O

N

NN

O

N

NN

O

2.5 ppm

1.8 ppm

N 2H

N

Cl

N

N

N

O

O

N

N

Cl

odone (m/z 470) acquired in the Orbitrap massxpanded mass range shown is shown for lowlted from predictive fragmentation using Mass

7Time

-sup

17.198

150.050.100.150.200.250.300.350.400.45

0.500.550.600.650.700.750.800.850.900.951.00

1

4.3 ppm

nefazhe eresu

tier 4.0.

63 (2.1e7) 274.15500 �1.35

368 PETERMAN ET AL. J Am Soc Mass Spectrom 2006, 17, 363–375

(Figure 5b), respectively. The mass accuracy associatedwith each fragment ion resulted from comparison ofexperimental values with predicted structures fromMass Frontier 4.0. The spectrum shows the 20-foldmagnified region covering the mass range of 135 to 265Da. The product ion spectra (Figure 5a) of the monohy-droxynefazodone metabolite at a Rt of 11.42 min revealsa single fragment ion at m/z 253 representing an addi-tion of mass 16 to the fragment ion at m/z 237 in thenefazodone parent ion spectra (see Figure 2) implicat-ing the 3-chlorophenylpiperazine ring as the site ofhydroxylation. In contrast, the product ion spectrum(Figure 5b) of the monohydroxylated nefazodone me-tabolite at an Rt of 12.22 min depicts diagnostic frag-ment ions at m/z 152, 170, 209, and 262. The mass errorfor the product ion at m/z 209, which corresponds to the3-chlorophenylpiperazine ring region within nefaz-odone, is �0.3 ppm. The fragment ions at m/z 170 (mass

0 2 4 6 8 10 12 140

50

1000

50

1000

50

1000

50

1000

50

100

Rel

ativ

e A

bu

nd

ance

0

50

1000

50

1000

50

100 13.36

2.20 2.81

8.58

10.65

10.83

11.01

12.2511.42

12.16

13.59

M4 m/z 306 NL: 1.79e6

M6 m/z 486 NL: 6.26e613.68

(a)

(b)

(c)

(d)

(e)

(f)

(g)

(h)

Figure 4. Extracted ion chromatograms in the p

Table 1. List of experimentally observed mass measurements ofpredicted mass-to-charge values for two plausible product ions o

Correspondingmass error (ppm)

Theoretical mass(left side fragment)

Experim(measure

394.93 140.02615 140.081395.12 154.04180 154.097307.29 180.05745 180.112�4.52 197.08400 197.083N/A N/A 218.091

�2.19 237.11530 237.114N/A N/A 246.123N/A N/A 274.154

its primary metabolites. A 10 ppm mass tolerance w

N

Cl

N NN

NO

O

Nefazodone

+ H

N

Cl

N NN

NO

O

H

m/z 470.232

[A]

[B]

NN

NO

O

m/z 274.155

NN

NO

m/z 180.1131

N

Cl

HN NN

NO

O

m/z 470.232

N

Cl

m/z 180.0575

Figure 3. Competitive reaction schemes for the formation of them/z 180 product ion resulting from collisional activation of nefa-zodone. The reaction mechanisms are predicted using Mass Fron-

16 18 20 22 24 26 28 30 32 34Time (min)

Nefazodone NL: 5.01e7

M1 m/z 213 NL: 2.23e5

M2 m/z 197 NL: 1.57e6

M5 m/z 306 NL: 6.25e5

M3 m/z 292 NL: 4.11e6

M8 m/z 486 NL: 6.94e6

M7 m/z 458 NL: 2.19e6

M9 m/z 484 NL: 1.98e5

ositive ionization mode for nefazodone and eight of

the nefazodone product ions following CID and the theoreticallybtained from Mass Frontier 4.0

ental massd intensity)

Theoretical mass(right side fragment)

Correspondingmass error (ppm)

45 (5.7e5) 140.08184 �2.7812 (1.1e5) 154.09749 �2.4078 (1.8e5) 180.11314 �2.0011 (8.2e3) 197.13969 �287.0067 (3.6e4) 218.09240 �3.3578 (5.4e4) 237.17099 �237.0043 (2.6e6) 246.12370 �1.10

as used to filter the full scan MS data.

369J Am Soc Mass Spectrom 2006, 17, 363–375 ORBITRAP MS IN METABOLITE IDENTIFICATION STUDIES

error � 3.5 ppm) and 262 (1.1 ppm), which represent theadditions of mass 16 to the fragment ions at m/z 154 andm/z 246, respectively, in the product ion spectrum ofnefazodone implicates the ethyl group in the 5-ethyl-triazolone motif as the site of monohydroxylation.Overall, the structures of these metabolites M6 and M8(see Scheme 1) are consistent with those reported inprevious studies [14 –17].

Accurate mass filtering of the full scan mass spectraldata also revealed the possibility of a second m/z 306metabolite (M5 in Scheme 1), which eluted at 11.01 minand appears to be an isomer of the previously charac-terized metabolite M4 (see Scheme 1) that elutes at 10.83min. Both metabolites are derived from downstreammetabolism of the N-dealkylated metabolite M3 thathas lost the 3-chlorophenylpiperazine ring. Data-depen-dent MS/MS spectra for each of the m/z 306 metabolitesare presented in Figure 6. The resulting product ionspectra for M4 (Figure 6a) contain the product ions atm/z 140 (mass error � 2.6 ppm) and 246 (mass error �2.0 ppm), which are also present in the mass spectrumof nefazodone. These ions along with the fragment ionat m/z 288 (representative of the acylium ion derivedfrom carboxylic acid fragmentation) are in excellentagreement with the previously reported mass spectrumof M4 [14, 16]. The previously undisclosed m/z 306metabolite eluting at 11.01 min shows additional diag-nostic product ions that can be utilized to gain insightinto its structure (Figure 6b). Specifically the fragment

140 160 180 2000

10

20

30

40

50

60

70

80

90

1000

10

20

30

40

50

60

70

80

90

100

Rel

ativ

e A

bu

nd

ance 140.08147

180.11282

154.09700

209.084

152.08156

178.09694170.09180

x20

2.6 ppm

3.2 ppm

3.1 ppm3.5 ppm

1.8 ppm

(a)

(b)

1.8 ppm

0.3 pp

NHN

NO

N

NN

O

NH

NN

O

N

NH

N

O

NH

NN

OOH N

NN

O

N

Cl

Rel

ativ

e A

bu

nd

ance

Figure 5. Comparison of the MS/MS spectra o� 11.42 min) and M8 (Rt � 12.25 min).

ions at m/z 168 and 194 provide insight into the two

distinct sites of oxidative metabolism as shown in theproposed structure for M5 in Scheme 1. The mass errorassociated with each of the unique fragment ions is 1.1and 1.9 ppm. The degree of bond saturation based onthe Mass Frontier derived chemical formula is calcu-lated to be 4.5 and 5.5, which agrees with the predictedproduct ion structure. We propose that M5 arises fromthe P4503A4-mediated N-dealkylation of M9.

Further examination of the LC-MS/MS analysis ofnefazodone microsomal incubation was conducted toprobe for the formation of additional and perhapsunanticipated metabolites. The fragment ions identifiedin the mass spectrum of nefazodone were also utilizedin extracting metabolite ions from all data-dependentMS/MS spectra, thus providing an alternate tool fordata analysis. Figure 7 shows extracted ion chromato-grams for the product ions at m/z values of 140.0818(Figure 7a) and 274.15,500 (Figure 7b) from MS/MSspectra obtained from nefazodone microsomal incuba-tions. The mass tolerance used for the extracted ionchromatograms was 10 ppm and the precursor ionyielding the selected product ions are listed at the top ofeach peak. It is worth commenting on the selectivitydisplayed in this analysis when using a tight masstolerance. For example, Figure 7a shows predominantpeaks at retention times 8.91, 10.67, 11.43, and 13.38min. The baseline in between each of the major peaks ismagnified 20-fold. Inspection of the magnified regionreflects the degree of selectivity considering that addi-

20 240 260 280

274.15469246.12343

.09195 253.10872

290.14951

09222

274.15506

246.12355

262.11832

235.09946

1.1 ppm

5.9 ppm

1.1 ppm

ppm

0.6 ppm1.1 ppm

1.4 ppm

ppm

NN

NO

O

N

N

ClOH

NN

NO

O

N

N

N

O

O

NN

NO

OH

ONN

NO

OH

O

0.2 ppm

nohydroxylated nefazodone metabolites M6 (Rt

2m/z

218

218.

06

0.8

3.3

m

N

f mo

tional metabolite peaks identified in Figure 7b (metab-

370 PETERMAN ET AL. J Am Soc Mass Spectrom 2006, 17, 363–375

80 100 120 140 160 180 200 220 240 260 280 300 320m/z

0

10

20

30

40

50

60

70

80

90

100

Rel

ativ

e A

bu

nd

ance

288.13375

246.12321140.08148212.10191

x15

80 100 120 140 160 180 200 220 240 260 280 300 320m/z

0

10

20

30

40

50

60

70

80

90

100

Rel

ativ

e A

bu

nd

ance

288.13372

212.10255

194.09203168.07655

x15 x15

234.12323

2.6 ppm 2.0 ppm

1.8 ppm

5.0 ppm

2.1 ppm

1.1 ppm

1.9 ppm

2.0 ppm1.9 ppm

(a)

(b)

NHN

NO

N

N

O

NOH

O

N2HN

NO

O ONN

NO

O

N

N NO

OH

N N

NOO

Figure 6. Comparison of the MS/MS spectra for the isomeric m/z 306 metabolites M4 (R � 10.83

tmin) and M5 (Rt � 11.01 min).

7.5 8.0 8.5 9.0 9.5 10.0 10.5 11.0 11.5 12.0 12.5 13.0 13.5 14.0 14.5Time (min)

0

10

20

30

40

50

60

70

80

90

100

Rel

ativ

e A

bu

nd

ance

13.38

13.34

11.43 13.428.91 10.678.93 10.84 13.46 13.6911.48

x20 x20 x20

MS/MS Product IonThreshold140.0801- 140.0829

m/z 360m/z 292

m/z 306

m/z 486

m/z 470

(a)

7.5 8.0 8.5 9.0 9.5 10.0 10.5 11.0 11.5 12.0 12.5 13.0 13.5 14.0 14.5Time (min)

0

10

20

30

40

50

60

70

80

90

100

Rel

ativ

e A

bu

nd

ance

13.38

13.34

8.91

10.67

11.4313.42

10.71 13.698.96 11.48

12.227.92 10.24 12.5010.20 14.089.24 10.297.97

x20 x20 x20

MS/MS Product IonThreshold274.1519 - 274.1573

m/z 470

m/z 486

m/z 486

m/z 292

m/z 502

m/z 274

m/z 376

(b)

Figure 7. Extracted ion chromatograms for MS/MS product ions from the nefazodone parent drug.Masses labeled indicate the precursor mass isolated for collisional activation. The mass threshold used

for chromatographic reconstruction is 10 ppm.

d iso

371J Am Soc Mass Spectrom 2006, 17, 363–375 ORBITRAP MS IN METABOLITE IDENTIFICATION STUDIES

olites at m/z 376 and 502 with a retention time of 7.92and 10.24 min, respectively) are not found in Figure 7a.The use of the linear trap/orbitrap hybrid arrangementthus enables detection of all product ions within a widemass spectral window as opposed to SRM analysisemployed by triple quadrupole mass spectrometers.Therefore, a more thorough interrogation in a post-acquisition method can be done by probing the result-ing data for evidence of additional metabolites viadifferent product ions from either the parent drug orpreviously characterized metabolites of the parentdrug. In addition, the high mass accuracy data acquisi-tion enables a selective and sensitive means of filteringthe resulting product ion spectra. Figure 7 underscoresthe utility of this approach showing identification ofpotential metabolites using only two nefazodone prod-uct ions. Many more product ions could be used in anexhaustive searching algorithm to thoroughly processthe data for identification of as many metabolites aswould be deemed necessary.

Further structural analysis of the metabolites identi-fied via the accurate mass product ion filtering was alsoexamined using Qual Browser capability of this instru-mentation. Using the elemental composition and rangesof the nefazodone parent structure, each metabolite wasmatched with a predicted chemical formula and RDBvalue. Predicted isotopic distribution was also com-pared to the experimentally determined ratios. Anexample of the described strategy is shown in Figure 8.Figure 8a shows an extracted ion chromatogram for them/z 502 ion filtered using a mass tolerance of 10 ppmand reveals only one elution peak. The observed m/zvalue of 502.22,138 was fitted with potential elemental

Figure 8. Extracted ion chromatogram for tcorresponding full scan MS. The screen capturformula and theoretical mass-to-charge value an

compositions showed in the bottom left range from the

Qual Browser screen-capture. The 5 ppm mass toler-ance was used for prediction of a singly charged speciesand the RDB constraint was widened to accommodatevalues ranging from �100 to 100. This range is fordemonstration purposes and can be reduced further tobetter fit the parent and/or metabolite structure. Asshown in the screen capture image in Figure 8, theformulas predict four possible elemental compositionsfor the observed m/z value. The first two formulae areassociated with mass errors less than 1 ppm while thelatter two formulas have mass errors around 3 ppm.Examination of the isotopic distribution indicates thepresence of one chlorine atom, which eliminates thesecond and fourth formulae. The correct molecularformula could also be determined based on other RDBvalue information. In the current scenario, the resultingions formed from electrospray ionization are even-electron species, and if they contain a charge, then theirbond saturation would result in a non-integer. Asshown in Figure 8, only the top choice shows a valuerepresenting a protonated species, thus all other possi-ble formulae could be eliminated. Finally, a comparisonof experimental and theoretical isotopic distributionscan also be used in structure elucidation. Figure 8shows excellent agreement between the experimentaland theoretical distribution for the m/z 502 metabolite.

Figure 9 displays the low mass range for the data-dependent product ion spectrum from the metabolitewith molecular ion at m/z 502, which is consistent withdihydroxylation of the parent nefazodone molecule.The base peak, which constitutes the ion at m/z 290(mass error � 2.1 ppm), is consistent with addition of 16Da to the ion at m/z 274 identical to that observed for the

ihydroxynefazodone metabolite M10 and theQual Browser shows the predicted chemical

topic distribution.

he de from

monohydroxy nefazodone metabolite M8 (m/z 486, Rt �

ure o

372 PETERMAN ET AL. J Am Soc Mass Spectrom 2006, 17, 363–375

12.25 min). Based on this observation and the appear-ance of the fragment ions at m/z 170, 188, 218, 246, and262 (average mass error associated with these fragmentions is 2.7 ppm), we propose that one site of hydroxy-lation is the ethyl group on the triazolone ring as isnoted with M8. Finally, the structures of the product

140 160 180 2000

2

4

6

8

10

12

14

16

18

20

22

24

26

28

30

32

34

36

Rel

ativ

e A

bu

nd

ance

188.10690

152.08147

178.09712

170.09230 2

145.06425

H+

5.7 ppm

3.7 ppm3.3 ppm

0.4 ppm

0.4 ppm

M10

N

Cl

N

NN

NO

O

OH

OH

NH

NN

O

OH

Figure 9. Full scan accurate mass product ionmetabolite M10 at m/z 502. The proposed struct

100 120 140 1600

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100

Rel

ativ

e A

bu

nd

ance

140.08156

180154.09703

x15

2.0 ppm

3.0 ppm1.0

H+

M11

NH

N

NN

NO

O

NHN

NO

NH

NN

O

O

Figure 10. Full scan accurate mass productmetabolite M11. The proposed structure is pres

mechanism proposed by Kalgutkar and coworkers [1

ions at m/z 225 (mass error � 0.4 ppm) and 253 (masserror � 0.04 ppm) led us to propose 4-hydroxylation ofthe 3-chlorophenyl ring as a second site of hydroxyla-tion considering that 4-hydroxynefazodone is the onlymetabolite detected from aromatic hydroxylation ofnefazodone [14 –17]. A similar conclusion was reached

220 240 260 280z

290.14941

218.09210

274.15488246.12334

262.11816

225.07886

96

253.11023

0.4 ppm

0.04 ppm

4.7 ppm

1.2 ppm

1.5 ppm

2.3 ppm

2.1 ppm

N

N

N

O

O

N

N

Cl

OH

N

N

Cl

OH

NN

NO

OH

O

ctrum for the suspected dihydroxynefazodonef M10 is presented.

200 220 240 260 280z

274.15488

246.123750.2 ppm

0.4 ppm

NN

NO

O

NN

NO

O

spectrum for the N-dephenylated piperazine. The formation of M11 is consistent with the

m/

10.054

x5

spe

180m/

.11296 ppm

N

NN

ionented

6].

373J Am Soc Mass Spectrom 2006, 17, 363–375 ORBITRAP MS IN METABOLITE IDENTIFICATION STUDIES

Jemal and coworkers [17] who identified similar meta-bolic pathways for nefazodone.

Of much interest in the present context was the faciledetection of the N-dephenylated piperazine metabolites ofnefazodone namely M11, M12, and M13, respectively, thathave been previously reported [16]. Consistent with theloss of the 3-chlorophenyl moiety, there was a loss of thediagnostic chlorine isotopic pattern in the full scan spectra.In the present study, accurate mass full scan and MS/MSspectra were used to predict and verify the structures.Figure 10 shows the data-dependent MS/MS spectrum for

100 120 140 160 1800

2

4

6

8

10

12

14

16

18

20

22

24

26

28

30

32

Rel

ativ

e A

bu

nd

ance

168.07649

140.04514

121.06438

x15

3.3 ppm

2.9 ppm

1.8 ppm

H+

M12

NH

N

NN

NO

O

O

O

O

NNH

NHO

NH

NN

OO

Figure 11. Full scan accurate mass product ion sfurther oxidation of M11.

100 120 140 160 1800

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100

Rel

ativ

e A

bu

nd

ance

18178.09702

152.08159

127.12237

156.07602

H+

0.5 ppm

1.3 ppm

5.1 ppm

2.8 ppm1

M13

NH

N

NN

NO

OH

O

N

NH

N

NH

N

O

NHN

NO

OH

Figure 12. Full scan accurate mass product ion

of M11.

the M11 metabolite. The predicted product ion structuresand corresponding mass errors are also provided. All ofthe product ions of metabolite M11 as discerned from theCID spectrum are identical to those observed in the massspectrum of parent nefazodone. Overall, the low masserrors associated with each of these product ions and thelow mass error (1.1 ppm) observed in the full scan MSanalysis of M11 increased confidence in the assignedmetabolite structure.

Figures 11 and 12 show the product ion spectra for M12and M13, respectively. M12 and M13 are derived from

200 220 240 260 280

288.13391

218.09213

9222

246.12376

x15

ppm

0.9 ppm

1.4 ppm

0.4 ppm

N

N

N

N

N

O

O

O

NN

NO

O

um for the amide metabolite M12 obtained after

200 220 240 260 280

290.14917

274.15460

218.09200

246.12318262.11795

2.0 ppm2.3 ppm

2.1 ppm

1.8 ppm

1.5 ppm

2.4 ppm

NN

NO

OH

O

NN

NO

OH

O

trum for M13 obtained via monohydroxylation

m/z

194.0

1.0

O

N

O

pectr

m/z

8.10677

x15

.2 ppm

spec

4

374 PETERMAN ET AL. J Am Soc Mass Spectrom 2006, 17, 363–375

further oxidative metabolism of M11. Comparison of theresulting product ion spectra with that shown in Figure 10was used to determine the site of modification. For exam-ple, Figure 11 shows the primary fragment ion at m/z 288(14 mass units greater than the m/z 274 product ionobserved in Figure 10) is consistent with the formation ofthe acylium ion (mass error � 1.4 ppm) and suggestive oflack of modification to the piperazine ring. The observedproduct ions at m/z 140, 168, and 194 is consistent with thesite of oxidation residing on the propyl chain as opposedto either the 5-ethyltriazolone motif or the phenoxyethylgroup. The observed acylium ion is consistent with P450-catalyzed oxidation on the methylene group � to thetertiary piperazine nitrogen in M11 affording amide M12.Figure 12 shows the product ion spectrum for M13. Theprimary fragment ions used to determine the site oftransformation were m/z 127, 152, 156, 262, and 290. Thepresence of the m/z 127 product ion indicates the pipera-zine ring was not the site of hydroxylation, rather thedetection of the m/z 156, 262, and 290 (average mass error� 2.1 ppm) show consistency with predicted structureinvolving the site of monohydroxylation on the 5-ethyl-triazolone moiety.

Combining the results from full scan MS and MS/MSspectra provide a comprehensive list of nefazodone me-tabolites listed in Table 2 obtained from one LC-MS/MS/MS experiment. The chemical formulas listed werepredicted using Qual Browser and used for mass errordetermination. The average mass error for all 16 identifiedcompounds was 1.8 ppm despite a 20-fold difference inintegrated peak areas. In addition, the RDB values calcu-lated shows all values are related to the predicted struc-tures increasing confidence in the predicted structures.

Conclusions

The present study outlines a facile method for identify-

Table 2. List of nefazodone and identified nefazodone metabolichemical formulas used were obtained from Qual Browser basedwere compared to the predicted chemical formula

No. Chemical formula Rt (min) Integrated peak area E

1 C10H13O1N2Cl1 2.70 2.23e52 C19H30O3N5 7.19 5.75e63 C10H13N2Cl1 8.51 1.57e64 C19H30O2N5 8.90 1.60e75 C15H20O2N3 8.90 5.21e56 C19H28O3N5 9.17 5.28e67 C25H33O4N5Cl1 10.23 1.22e78 C15H21O3N3 10.65 4.11e69 C15H19O4N3 10.83 1.79e6

10 C15H19O4N3 11.01 6.25e511 C25H32O3N5Cl1 11.42 6.26e612 C23H28O3N5Cl1 12.16 2.19e613 C25H32O3N5Cl1 12.25 6.94e614 C25H32O2N5Cl1 13.35 5.01e715 C25H31O3N5Cl1 13.59 1.98e516 C25H32O3N5Cl1 13.68 6.25e5

ing obvious and unanticipated metabolites in a single

chromatographic run using accurate mass analysis offull scan MS and MS/MS spectra without the need foran inclusion list. The success of the methodology isbased on fast scan speeds, high mass accuracy measure-ments with large intra-scan dynamic range, and the useof external calibration. Resulting mass spectral data canthen be processed manually by experts in the biotrans-formation field or by predictive metabolite identifica-tion software to generate possible parent and production structures that are easily accepted or refuted usingaccurate mass analysis, bond saturation, and chemicalformula ranges based on comparisons with mass spec-tral data on the parent drug. Combination of thesefeatures will undoubtedly improve confidence in theassignments of chemical formulae and biotransforma-tion sites on metabolites. Furthermore, the ability toaccurately measure product ion mass values withoutthe need for internal calibrants or multiple experimentalruns suggest that such analyses may be useful in earlystages of drug discovery to screen the metabolic patternof multiple compounds within a structural class or evenstructurally diverse chemical series in a rapid anditerative fashion and the knowledge of biotransforma-tion pathways gained from such analysis could allow afacile optimization of ADME attributes. In vitro metab-olite identification studies are currently underway onnovel compounds with unknown biotransformationpathways and at concentrations close to typical thera-peutic range (0.1 to 0.5 �M) to fully demonstrate themerits of this instrumentation.

References1. Watt, A. P.; Mortishire-Smith, R. J.; Gerhard, U.; Thomas, S. R. Metab-

olite Identification in Drug Discovery. Curr. Opin. Drug Discov. Dev.2003, 6(1), 57–65.

2. Kalgutkar, A. S.; Soglia, J. R. Minimizing the Potential for Metabolic

om a 10 �M incubation in human liver microsomes. Proposedefazodone elemental composition and ranges. Mass accuracies

imental mass Theoretical mass RDB Mass error (ppm)

13.07921 213.07892 5.5 1.476.23505 376.23432 7.5 1.997.08452 197.08400 5.5 2.660.23981 360.23940 7.5 1.174.15567 274.15500 7.5 2.474.21829 374.21867 8.5 �1.002.22086 502.22253 11.5 �3.392.16608 292.16557 7.5 1.706.14557 306.14483 8.5 2.406.14539 306.14483 8.5 1.886.22742 486.22664 11.5 1.658.19559 458.19534 11.5 1.686.22748 486.22664 11.5 1.770.23224 470.23173 11.5 1.184.22167 484.21099 12.5 1.486.22754 486.22664 11.5 1.9

tes fron n

xper

231323523344444

Activation in Drug Discovery. Exp. Opin. Drug Metab. Toxicol. 2005, 1 1),91–141.

375J Am Soc Mass Spectrom 2006, 17, 363–375 ORBITRAP MS IN METABOLITE IDENTIFICATION STUDIES

3. Liu, D. Q.; Hop, C. E. C. A. Strategies for Characterization of DrugMetabolites Using Liquid Chromatography-Tandem Mass Spectrome-try in Conjunction with Chemical Derivatization and On-Line H/DExchange Approaches. J. Pharm. Biomed. Anal. 2005, 37(1), 1–18

4. Hop, C. E. C. A.; Yu, X.; Xu, X.; Singh, R.; Wong, B. K. Elucidation ofFragmentation Mechanisms Involving Transfer of Three HydrogenAtoms Using a Quadrupole Time-of-Flight Mass Spectrometer. J. MassSpectrom. 2001, 36(5), 575–579

5. Yin, W.; Doss, G. A.; Stearns, R. A.; Chaudhary, A. G.; Hop, C. E.;Frankin, R. B.; Kumar, S. A Novel P450-Catalyzed Transformation of the2,2,6,6-Tetramethyl Piperidine Moiety to a 2,2-Dimethyl Pyrrolidine inHuman Liver Microsomes: Characterization by High Resolution Qua-drupole-Time-of-Flight Mass Spectrometry and 1H-NMR. Drug Metab.Dispos. 2003, 31(2), 215–223

6. Gangl, E.; Utkin, I.; Gerber, N.; Vouros, P. Structural Elucidation ofMetabolites of Ritonavir and Indinavir by Liquid Chromatography-Mass Spectrometry. J. Chromatogr. A. 2002, 974(1/2), 91–101

7. Mutlib, A. E.; Shockcor, J. P. 2003; Application of LC/MS, LC/NMR,NMR and Stable Isotopes in Identifying and Characterizing Metabo-lites. Lee, J. S.; Obach, R. S.; Fisher, M. B., Eds.; In Drug MetabolizingEnzymes: Cytochrome P450 and Other Enzymes in Drug Discovery andDevelopment; p 33Marcel Dekker: New York, NY, Fontis Media:Lausanne, Switzerland.

8. Li, L.; Chiarelli, M. P.; Branco, P. S.; Marques, M. M.; Goncalves, L. L.;Beland, F. A. Differentiation of Isomeric C8-Substituted AlkylanilineAdducts of Guanine by Electrospray Ionization and Tandem Quadru-pole Ion Trap Mass Spectrometry. J. Am. Soc. Mass Spectrom. 2003,14(12), 1488–1492

9. Hopfgartner, G.; Chernushevich, I. V.; Covey, T.; Plomley, J. B.; Bonner,R. Exact Mass Measurement of Product Ions for the Structural Elucida-tion of Drug Metabolites with a Tandem Quadrupole Orthogonal-Acceleration Time-of-Flight Mass Spectrometer. J. Am. Soc. Mass Spec-trom. 1999, 10(12), 1305–1314

10. Deroussent, A.; Re, M.; Hoellinger, H.; Vanquelef, E.; Duval, O.;Sonnier, M.; Cresteil, T. In Vitro Metabolism of Ethoxidine by HumanCYP1A1 and Rat Microsomes: Identification of Metabolites by High-Performance Liquid Chromatography Combined with ElectrosprayTandem Mass Spectrometry and Accurate Mass Measurements byTime-of-Flight Mass Spectrometry. Rapid Commun. Mass Spectrom. 2004,18(4), 474–482

11. Castro-Perez, J.; Plumb, R.; Granger, J. H.; Beattie, I.; Joncour, K.;

Wright, A. Increasing Throughput and Information Content for in VitroDrug Metabolism Experiments Using Ultra-Performance Liquid Chro-

matography Coupled to a Quadrupole Time-of-Flight MassSpectrometer. Rapid Commun. Mass Spectrom. 2005, 19(6), 843–848

12. Hardman, M.; Makarov, A. A. Interfacing the Orbitrap Mass Analyzerto an Electrospray Ion Source. Anal. Chem. 2003, 75(7), 1699–1705

13. Thevis, M.; Makarov, A. A.; Horning, S.; Schanzer, W. Mass Spectrom-etry of Stanozolol and Its Analogs Using Electrospray Ionization andCollision-Induced Dissociation with Quadrupole-Linear Ion Trap andLinear Ion Trap-Orbitrap Hybrid Mass Snalyzers. Rapid Commun. MassSpectrom. 2005, 19(22), 3369–3378

14. Mayol, R. F.; Cole, C. A.; Luke, G. M.; Colson, K. L.; Kerns, E. H.Characterization of the Metabolites of the Antidepressant Drug Nefaz-odone in Human Urine and Plasma. Drug Metab. Dispos. 1994, 22(2),304–311

15. Von Moltke, L. L.; Greenblatt, D. J.; Granda, B. W.; Grassi, J. M.;Schmider, J.; Harmatz, J. S.; Shader, R. I. Nefazodone, Meta-Chlorophe-nylpiperazine, and their Metabolites in Vitro: Cytochromes MediatingTransformation, and P450-3A4 Inhibitory Actions. Psychopharmacology1999, 145(1), 113–122.

16. Kalgutkar, A. S.; Vaz, A. D. N.; Lame, M. E.; Henne, K. R.; Soglia, J.;Zhao, S. X.; Abramov, Y. A.; Lombardo, F.; Collin, C.; Hendsch, Z. S.;Hop, C. E. C. A. Bioactivation of the Nontricyclic AntidepressantNefazodone to a Reactive Quinone-Imine Species in Human LiverMicrosomes and Recombinant Cytochrome P450 3A4. Drug Metab.Dispos. 2005, 33(2), 243–253

17. Jemal, M.; Ouyang, Z.; Zhao, W.; Zhu, M.; Wu, W. W. A Strategy forMetabolite Identification Using Triple-Quadrupole Mass Spectrometrywith Enhanced Resolution and Accurate Mass Capability. Rapid Com-mun. Mass Spectrom. 2003, 17(24), 2732–2740

18. Hu, Q.; Noll, R. J.; Li, H.; Makarov, A.; Hardman, M.; Cooks, R. G. TheOrbitrap: A new Mass Spectrometer. J. Mass Spectrom. 2005, 40(4),430–443

19. McLafferty, F. W.; Turecek, F. Interpretation of Mass Spectra, 4th ed.;University Science Books: Mill Valley, CA, 1993, p 27.

20. Pittenauer, E.; Zehl, M.; Mistrik, R.; Allmaier, G. Evaluation of ESI-MSnIon Trap and MALDI-MSn Ion Trap/Reflectron TOF Mass Spectra(n�1–3) for Incorporation into a Searchable Database. Proceedings of the52nd ASMS Conference on Mass Spectrometry and Allied Topics; Nashville,TN 2004.

21. Mistrik, R. Comprehensive Database of Mechanisms of Peptide Frag-mentation: A Key Step Towards Confident Identification of Proteins

Proceedings of the 52nd ASMS Conference on Mass Spectrometry and AlliedTopics; Nashville, TN 2004.

Application of a Linear Ion Trap/Orbitrap Mass Spectrometer in Metabolite Characterization Studies: Examination of the...ExperimentalMaterialsMicrosomal IncubationsLC-MS/MS Assay for Identification of Nefazodone MetabolitesData Interpretation

Results and DiscussionConclusionsReferences