automated molecular formula determination by tandem mass spectrometry (ms/ms

PAPER www.rsc.org/analyst | Analyst

Automated molecular formula determination by tandem mass spectrometry(MS/MS)†

Suwatchai Jarussophon,ab Stephane Acoca,ac Jin-Ming Gao,ad Christophe Deprez,a Taira Kiyota,‡a

Cristina Draghici,‡a Enrico Purisimaa and Yasuo Konishi*a

Received 20th October 2008, Accepted 12th January 2009

First published as an Advance Article on the web 5th February 2009

DOI: 10.1039/b818398h

Automated software was developed to analyze the molecular formula of organic molecules and

peptides based on high-resolution MS/MS spectroscopic data. The software was validated with 96

compounds including a few small peptides in the mass range of 138–1569 Da containing the elements

carbon, hydrogen, nitrogen and oxygen. A Micromass Waters Q-TOF Ultima Global mass

spectrometer was used to measure the molecular masses of precursor and fragment ions. Our software

assigned correct molecular formulas for 91 compounds, incorrect molecular formulas for 3 compounds,

and no molecular formula for 2 compounds. The obtained 95% success rate indicates high reliability of

the software. The mass accuracy of the precursor ion and the fragment ions, which is critical for the

success of the analysis, was high, i.e. the accuracy and the precision of 850 data were 0.0012 Da and

0.0016 Da, respectively. For the precursor and fragment ions below 500 Da, 60% and 90% of the data

showed accuracy within #0.001 Da and #0.002 Da, respectively. The precursor and fragment ions

above 500 Da showed slightly lower accuracy, i.e. 40% and 70% of them showed accuracy within

#0.001 Da and #0.002 Da, respectively. The molecular formulas of the precursor and the fragments

were further used to analyze possible mass spectrometric fragmentation pathways, which would be

a powerful tool in structural analysis and identification of small molecules. The method is valuable in

the rapid screening and identification of small molecules such as the dereplication of natural products,

characterization of drug metabolites, and identification of small peptide fragments in proteomics. The

analysis was also extended to compounds that contain a chlorine or bromine atom.

Introduction

Over the past decades, the major sources of marketed drugs have

been natural products, or their semi-synthetic derivatives.

Natural products provide larger structural diversity than

combinatorial chemistry products and offer significant oppor-

tunities for finding novel lead compounds.1–5 Each year, a large

number of new natural products are discovered and fully char-

acterized; however, during the courses of isolation, character-

ization and identification, natural products chemists have faced

increasing problems of replication, i.e. re-discovery of known

natural products.6,7

aBiotechnology Research Institute, National Research Council Canada,6100 Royalmount Avenue, Montr�eal, Qu�ebec, Canada H4P 2R2. E-mail:[email protected]; Fax: +1 514 496 5243; Tel: +1 514 4966339bNational Nanotechnology Center, National Science and TechnologyDevelopment Agency, 130 Thailand Science Park, Paholyothin Rd.,Klong Luang, Pathumthani 12120, Thailand. E-mail: [email protected] of Biochemistry, McIntyreMedical Sciences Building,McGillUniversity, 3655 Promenade Sir William Osler, Montr�eal, Qu�ebec, CanadaH3G 1Y6dNatural Products Resource Research Centre, College of Science,Northwest A & F University, Yangling, Shaanxi 712100, PR China

† Electronic supplementary information (ESI) available: Table showingMFA of 5-leucine enkephalin. See DOI: 10.1039/b818398h

‡ Current address: Wyeth Research, 1025 Marcel Laurin Bd.,Saint-Laurent, Qu�ebec, Canada H4R 1J6.

690 | Analyst, 2009, 134, 690–700

A method that identifies and eliminates known compounds in

the early stages of the natural product discovery process is

generally known as dereplication. It plays a key role in phyto-

chemistry and in an effective natural product discovery

program.8–10 Dereplication typically uses a combination of

analytical techniques and database searching11–14 to identify

active compounds. The databases that are extensively used are

Chemical Abstracts (CA), Beilstein, Bioactive Natural Product

Database, Chapman & Hall’s Dictionary of Natural Products and

Natural Products Alert (NAPRALERT). There have been

considerable developments of analytical separation techniques

such as GC, LC and CE, and of spectroscopic characterization

techniques such as PDA, IR, NMR, and MS. Hyphenated

techniques couple the separation techniques with online spec-

troscopic characterization, e.g. LC-MS, GC-MS, CE-MS,

LC-UV and LC-NMR. They are expected to resolve the

complexity of the natural product extracts. Recent advances of

hyphenated techniques and their applications have been reported

by several research laboratories.15–18 In addition, multiple

combinations of the characterization techniques have been

developed, e.g. LC-PDA-MS,19 LC-NMR-MS,20,21 LC-SPE-

NMR,22 LC-PDA-IR-NMR-MS,23 and 2D-LC(IEC-RP)-MS.24

These improve the sample separation, structural characterization

and elucidation, and also the detection of valuable minor

components in natural sources.

Due to high productivity and sensitivity, mass spectrometry

has become the most powerful and essential technique providing

This journal is ª The Royal Society of Chemistry 2009

critical information in many phases of drug discovery and

development. Examples are structural characterization and

identification,25,26 quantitation,27 high-throughput screening,28,29

proteomics,30,31 metabonomics,32–34 metabolomics,35–41 and der-

eplication of natural products. In the area of dereplication, MS

delivers the molecular mass information that can be used as

a search query in almost all databases of small molecules.42,43

Accurate mass obtained from high-resolution MS is commonly

used to search for candidates in literature databases.44 Unfor-

tunately, in general, several molecular formulas (MFs) are

compatible with the observed molecular mass within the exper-

imental accuracy, resulting in a large set of compounds, most of

which are false positives. MS/MS measurements often provide

more information of some sub-structures, or fingerprints of sub-

structures.45,46 There are few databases that provide MS/MS data

such as NIST/EPA/NIH Mass Spectral Library which contains

14 802 MS/MS spectra. They may become practical and useful

when more MS/MS data are accumulated and systematically

analyzed.

The dereplication process starts by using the information

stored in natural product databases. The molecular formula

(MF) is one of the most valuable pieces of information as it is

available in any natural product database and is independent of

the source, of the sample preparation, and of the conditions of

the measurements. Conventional elemental composition analysis

is typically achieved by high-resolution MS47,48 such as magnetic

sector MS, time-of-flight MS (TOF-MS)49 and Fourier transform

MS (FTMS).50 These instruments provide a set of MFs that can

be further narrowed down in combination with NMR data.

However, NMR is much less sensitive than MS, requiring a large

quantity of purified sample. Thus, the combination of MS and

NMR practically brings little contribution to dereplication. It

should be mentioned that the number of possible MFs expo-

nentially increases with the size of the molecules, whereas there

has been no drastic progress to improve accurate mass determi-

nation.51,52

In our previous paper,53 we analyzed MF and fragmentation

pathways of small molecules based on their accurate MS, MS/

MS, and MS/MS/MS data. The method provided detailed

information on the structure and sub-structures and their frag-

mentation pathways. The method is useful when a specific

molecule is targeted for detailed structural analysis. However,

the measurements and the analysis were not automated. Thus, we

have now developed simple, automated, and productive soft-

ware, which determines MF of small molecules and their frag-

ments based on the accurate MS and MS/MS data.

Experimental

Materials

All compounds used in our study were randomly selected from

an organo-chemical drug library (Negwer’s database) and those

that are commercially available. [Glu1]-Fibrinopeptide B was

purchased from Sigma (Oakville, ON, Canada) and used as

a reference compound for the calibration of the mass spec-

trometer. All reagents were used without further purification.

Water and acetonitrile are of HPLC grade and were purchased

from Anachemia (Lachine, QC, Canada) and J. T. Baker


(Phillipsburg, NJ), respectively. Formic acid was purchased from

Fluka (Oakville, ON, Canada) and was used to aid the positive

ion electrospray ionization process. All solvents were degassed

for at least 30 minutes before use.

Instrumentation

All MS and MS/MS measurements were performed in positive

ion electrospray mode (+ESI) on a Micromass Waters Q-TOF

Ultima Global mass spectrometer equipped with a Z-spray ion

source and NanoLockSpray (Waters, Mississauga, ON, Canada)

source. The m/z was acquired within the mass range of 50–990 m/z

for small organic molecules and 100–1990 m/z for organic

molecules with >900 Da molecular mass and small peptides. The

acquisition time per spectrum was set to 1 s, inter-scan delay was

set to 0.1 s, with the lock spray frequency set to 4 s.54–56 The mass

spectrometer was set up in V mode with instrument resolution

between 9000 and 10 000 based on FWHM. The source and

desolvation temperature were set to 80 and 150 �C, respectively.

The TOF was operated at an acceleration voltage of 9.1 kV,

a cone voltage of 100 V, an RF lens of 45 V, and a capillary

voltage of 3.8 kV. Operating parameters of the ESI interface were

optimized by infusing standard solutions of [Glu1]-fibrinopeptide

B, 100 nM in a solution of water–acetonitrile 50 : 50 (v/v) with

0.1% formic acid at a flow rate of 1.0 mL/min. The instrument was

carefully calibrated such that the error of the MS/MS fragments

of [Glu1]-fibrinopeptide B was less than 4 ppm. All measurements

were performed at room temperature. The MassLynx 4.0 chro-

matographic software was used for instrument control and data

analysis.

MS/MS experiments

A precursor ion of interest was selected at the quadrupole (Q1). It

was fragmented in the hexapole collision cell with argon collision

gas. Fragment ions were measured to obtain their m/z and peak

area. The collision energy was adjusted for each compound

typically from 5 to 40 eV in order to maintain the precursor ion

peak in the range of 35–100 counts/scan, while maximizing the

peak areas of the product ions. Sometimes, high collision energy

was used just to enhance the peak areas of the fragment ions at

low m/z. Fragment ions of 20–400 counts/scan were used for

most of the analyses. The acquired mass spectra were accumu-

lated for at least 2 min. The mass measurements are most accu-

rate when analyte/lock mass intensity ratio is between 0.5 and

2.0.57,58 In a few cases, fragment ions lower than 20 counts/scan

were used for the analysis after accumulating many scans.

Typically, analytes were dissolved in water–acetonitrile 50 : 50 (v/v)

with 0.1% formic acid and directly infused to the mass spec-

trometer using a Harvard syringe pump or autosampler direct

injection. For the reference channel, freshly prepared [Glu1]-

fibrinopeptide B (�1 mM) in water–acetonitrile 50 : 50 (v/v) with

0.1% formic acid was continuously infused to maintain the

constant concentration of reference solution. Both analyte and

reference channels were controlled by NanoLockSpray. The

TOF mass correction (accurate mass measurement) parameters

were performed with the following parameter sets: no back-

ground subtraction;59 Savitsky–Golay’s smooth type; smooth

window 3 channels; number of smooths 1;60 minimum peak

Analyst, 2009, 134, 690–700 | 691

width at half-height 4 channels; centroid top 60%; the dead-time

correction was turned on. Spectral intensity cut-off threshold

setting of 0.1–1.0% was used to simplify and reduce the number

of peaks analyzed which have not sufficient intensity to get good

accuracy.61,62 The TOF transform was used to exclude all isotope

peaks.

The algorithm of molecular formula analysis (MFA)

The forward and reverse MFA algorithms have been described

previously.53 Briefly, the MS/MS experiment of a precursor ion A

generates several fragment ions, A1, A2, A3, A4, A5, A6, A7, A8,

and A9 (Table 1). All possible neutral fragments Ni (i¼ 1–9) and

Nij (i ¼ 1–8; j ¼ 2–9; j > i) are listed in Table 1, i.e. Ni (i ¼ 1–9)

are generated from A and Nij (i¼ 1–8; j¼ 2–9; j > i) are possibly

generated from Ai (i ¼ 1–8). H+ (¼ 1.0078 Da) is added artifi-

cially as the smallest product ion to obtain the corresponding

neutral product N, N1H,., N9H in Table 1.

(1) The first step of the analysis is designated as ‘Forward

Analysis’. Briefly, accurate mass measurement uniquely deter-

mines the MF of some small fragments. The MFs of these small

fragments are then added up sequentially to determine the MF of

the precursor ion, A. More specifically, the MFA was carried out

for the neutral losses NiH (i ¼ 1–9), N, and Ni9 (i ¼ 1–8). Two

restrictions were applied in the search. One is the molecular size

of the neutral loss, which is typically limited to 200–400 Da,

preferably 200 Da, and the other is the error cut-off, which is

0.002–0.003 Da, preferably 0.002 Da. These restrictions reduce

CPU time and enhance the identification of a unique MF,

respectively. The neutral molecules that have been uniquely

identified are highlighted in bold (N9H, N8H, N89, N6H, .,

etc.). It should be emphasized that all Ni9 (i ¼ 1–8) are simply

listed to fill Table 1 and some of them may not exist mathe-

matically or physically. The MF of fragment ion A9 is assigned

from the added MFs (N9H + H+). The observed m/z value of A9

is then replaced with the one calculated from the assigned MF.

The molecular masses of Ni9 (i ¼ 1–8) are also replaced

accordingly. Similarly, the MF of fragment ion A8 is assigned

from (N8H + H+) and (N89 + A9). The process continues up to

the assignment of the MF of A from (N1 + A1), (N2 + A2), (N3 +

A3), (N4 + A4), and (N5 + A5). In most of the analysis, the MF

of A is uniquely assigned; however, sometimes, two or more MFs

are assigned for A. All of the MFs assigned are further examined

in the next step, which is designated as ‘Reverse Analysis’, where

Table 1 Potential neutral losses in the MS/MS experiment in forward MFA

Precursor Product ions and neutral products

A A1 A2 A3 A4A N1 N2 N3 N4A1 N12 N13 N14A2 N23 N24A3 N34A4A5A6A7A8A9

692 | Analyst, 2009, 134, 690–700

each of the MFs of the fragment ions and neutral losses assigned

in ‘Forward Analysis’ is re-examined.

(2) The second step is designated as ‘Reverse Analysis’. First,

the observed m/z value of the precursor ion is replaced with the

m/z value calculated from the assigned MF. The MFs of Ni (i ¼1–9), in which each element is restricted not to exceed that of A,

are analyzed with a cut-off error of 0.002–0.003 Da, but without

limitation of the molecular size. The MF of A1 is determined as

the difference of those of A and N1. The observed m/z value of

A1 is replaced with that of the calculated value. The molecular

masses (MM) of N1i (i ¼ 1–9) are also replaced accordingly.

Similarly, the MF of the fragment ion A2 is assigned from (A �N2), and (A1 � N12). The process continues until the MF of A9

is assigned from (A � N9), (A1 � N19), (A2 � N29), (A3 �N39), (A4 � A49), (A5 � A59), (A6 � N69), (A7 � N79), and

(A8 � N89). If they are not consistent, the MF of Ai is assigned

by taking the one most frequently assigned and using it in the

following steps. Table 2 shows the outcome of the reverse anal-

ysis of brefeldin A.

(3) The third step is designated as ‘Least-squares Index’,

where statistical verification is introduced on the outcome of

‘Reverse Analysis’. This step is designed to select a correct MF

out of multiple MFs that are occasionally found in ‘Forward

Analysis’. It should be recalled that ‘Reverse Analysis’ is per-

formed for all of the MFs derived from ‘Forward Analysis’. The

MFA of brefeldin A is an example that resulted in two MFs of

C16H24O4 and C17H20N4 in ‘Forward Analysis’. Their ‘Reverse

Analysis’ gives quite different MFs of the fragment ions and

neutral losses. Tables 2 and 3 show both correct and incorrect

MFA in the reverse analysis (the formats of precursor ion,

fragment ions and neutral losses are the same as those in Table

1). The two tables have quite different characteristics. The one

with the correct MF of the precursor ion produces the MFs of all

fragment ions and of most of the potential neutral losses. On the

other hand, the one with incorrect MF of the precursor ion fails

to assign the MF of four fragment ions and of several neutral

losses. In order to quantitate the different characteristics of the

tables, a least-squares index was introduced in the automated

evaluation software to evaluate the difference numerically. The

index is based on the observation (illustrated by Table 2) that the

number of possible neutral losses, NLcalc, associated with the

fragment ions linearly increases as the size of the fragment ions

decreases. This gives a range of values for NLcalc. Not all of these

neutral losses are observed in the table, i.e. NLobs is not always

equal to NLcalc. Empirically, we observe that the correlation

A5 A6 A7 A8 A9 H+

N5 N6 N7 N8 N9 NN15 N16 N17 N18 N19 N1HN25 N26 N27 N28 N29 N2HN35 N36 N37 N38 N39 N3HN45 N46 N47 N48 N49 N4H

N56 N57 N58 N59 N5HN67 N68 N69 N6H

N78 N79 N7HN89 N8H

N9H


Table 2 Reverse MFA of brefeldin A with correct formula of precursor ion

MMobs. 280.1673 262.1569 244.1458 226.1363 216.1521 198.1413 184.1261 162.1419 158.1113 130.0791

C16H24O4 C16H22O3 C16H20O2 C16H18O C15H20O C15H18 C14H16 C12H18 C12H14 C10H10

C16H24O4 — H2O H4O2 H6O3 CH4O3 CH6O4 C2H8O4 C4H6O4 C4H10O4 C6H14O4

C16H22O3 — H2O H4O2 CH2O2 CH4O3 C2H6O3 C4H4O3 C4H8O3 C6H12O3

C16H20O2 — H2O CO CH2O2 C2H4O2 C4H2O2 C4H6O2 C6H10O2

C16H18O — CO C2H2O C4O C4H4O C6H8OC15H20O — H2O CH4O C3H2O C3H6O C5H10OC15H18 — C3 C3H4 C5H8

C14H16 — C2H2 C4H6

C12H18 — H4 C2H8

C12H14 — C2H4

C10H10 —

NLcalca 1 2 3 4 5 6 7 8 9

NLobsb 1 2 3 3 5 5 6 8 9

a The number of possible neutral losses associated with each fragment ion. b The number of neutral losses, of which MFs are assigned, associated witheach fragment ion.

Table 3 Reverse MFA of brefeldin A with incorrect formula of precursor ion

MMobs 280.1673 262.1569 244.1458 226.1363 216.1521 198.1413 184.1261 162.1419 158.1113 130.0791

C17H20N4 C15H18 C14H16 C12H18 C12H14 C10H10

C17H20N4 — C2H2N4 C3H4N4 C5H2N4 C5H6N4 C7H10N4

——

——

C15H18 — C3 C3H4 C5H8

C14H16 — C2H2 C4H6

C12H18 — H4 C2H8C12H14 — C2H4

C10H10 —

NLcalca 1 2 3 4 5

NLobsb 1 1 2 4 5

a The number of possible neutral losses associated with each fragment ion (fragment ions, of which MFs are not assigned, are not counted). b Thenumber of neutral losses, of which MFs are assigned, associated with each fragment ion (fragment ions, of which MFs are not assigned, are not counted).

between NLobs and NLcalc is better when the correct precursor

ion is used. For example, in Tables 2 and 3 the Pearson corre-

lation coefficients R2 are 0.9749 and 0.9292, respectively. Hence,

the MFA giving higher R2 value is taken as the correct MF.

Nitrogen-enriched or oxygen-enriched compounds

Bases such as adenine and cytosine are typical nitrogen-enriched

groups, and saccharides are typical oxygen-enriched groups. As

the molecular masses of CN4 (68.0122 Da) and H4O4 (68.0109

Da) are very close, the MFA of nitrogen-enriched and oxygen-

enriched fragments tends to end up with two MFs. For example,

the analysis of glucose neutral loss (C6H12O6, 180.0633 Da) also

assigns a nitrogen-enriched neutral loss (C7H8N4O2, 180.0647

Da) such as p-xanthine, theophylline, and NSC265259 with

a molecular mass difference of only 0.0013 Da. In this case, they

may be distinguished by another dehydrated neutral loss

(C6H10O5, 162.0528 Da), i.e. dehydration should be observed for

glucose, but not for the base. In order to minimize the failure of

our MFA, a few commonly observed saccharides and bases are

pre-assigned in the software and the presence/absence of the


dehydrated fragment is manually confirmed later. However,

exceptional cases can still occur. For example, the analysis of

streptomycin failed as it contains both nitrogen-enriched

(C8H18N6O4) and oxygen-enriched (C7H13NO4 and C6H8O4)

sub-structures. Since none of them are commonly present in

small molecules, they are not pre-assigned, resulting in no unique

MF of the precursor molecule. Nucleoside analogs containing

bases and ribose analogs may generally have the same problem.

Results and discussion

Risk of assigning incorrect MF

The first ‘Forward Analysis’ step is typically performed with

a molecular mass cut-off of 200 Da and a mass accuracy of 0.002

Da. We have applied ‘Forward Analysis’ to 96 small molecules

(138–1569 Da) and observed the following:

(1) ‘Forward Analysis’ of 86 out of 96 compounds (90%)

resulted in a unique MF for each precursor molecule. Among

them, 83 were assigned the correct MF and 3 were assigned the

incorrect MF (97% success rate).

Analyst, 2009, 134, 690–700 | 693

Fig. 2 Proposed fragmentation pathways of brefeldin A.

(2) No MF was assigned for two of the compounds (tro-

leandomycin, 813.4511 Da; streptomycin, 581.2657 Da). In the

case of troleandomycin, only a few fragment ions were detected.

The MS/MS analysis does not work when there are not enough

detected peaks of fragment ions. However, for streptomycin,

sufficient numbers of fragment peaks were observed and yet the

analysis failed. This suggests that our analysis seems to be rather

weak in analyzing sugar-containing compounds, requiring

further improvement of the method.

(3) Two MFs were assigned for each of the seven compounds.

The ‘Least-squares Index’ on the outcome of ‘Reverse Analysis’

assigned the correct MF in all seven cases.

(4) Three MFs were assigned for protoveratrine A (793.4249

Da). The ‘Least-squares Index’ on the outcome of ‘Reverse

Analysis’ selected the correct MF.

Thus, the integrated approach reduced the risk of assigning an

incorrect MF to 3% (3 out of 96 compounds).

Mass accuracy

The accuracy of the data is crucial for the success of the MFA.

The mass spectrometer is calibrated and tuned for peak shape

(symmetry and tailing), resolving power (9000–10 000), and ion

abundance in the mass ranges of interest. Peaks for which ion

abundance was insufficient (<20 counts/scan) or exceeding

saturation (400 counts/scan) tended to have large errors, even

after accumulation of several scans, and were excluded from the

analysis. Alternatively, an isotope peak could be used instead of

the over-saturated monoisotope peak. The MFA used a total

of 850 peaks of the 96 compounds. After assigning the MF to all

of the 850 peaks, the mass accuracy of those peaks was 0.0012 Da

with 0.0016 Da precision. Out of them, the mass accuracy of the

734 peaks (86% of 850 peaks) was #0.002 Da, while 55, 36, 18

and 7 peaks had less accuracy of 0.002–0.003 Da, 0.003–0.005

Da, 0.005–0.010 Da, and 0.010–0.014 Da, respectively. The

peaks at high m/z tend to be less accurate. Only 40% and 70% of

the peaks above 500 Da achieved the accuracy of #0.001 Da and

#0.002 Da. In contrast, 60% and 90% of the peaks below 500 Da

achieved the accuracy of #0.001 Da and #0.002 Da, respec-

tively. There is no need to use many peaks in the analysis.

Instead, it is important that the neighboring fragment ions in the

MFA are within 200 Da.

Fragmentation pathways of brefeldin A

The MS/MS spectrum of brefeldin A is shown in Fig. 1. Fig. 2

shows plausible fragmentation pathways analyzed based on the

fragment ions used in the MFA (Table 2). Further MSn (n $ 3)

Fig. 1 The MS/MS spectrum of brefeldin A.

694 | Analyst, 2009, 134, 690–700

studies are required to validate them. Also, the fragment ions

that are not used in the MFA would provide more detailed

fragmentation pathways. Nevertheless, these plausible frag-

mentation pathways provide useful information on the structure

and sub-structure of the target molecules.

McLafferty rearrangement is introduced to explain three

consecutive water losses from [C16H25O4]+ (m/z 281) to

[C16H23O3]+ (m/z 263), [C16H21O2]+ (m/z 245), and [C16H19O]+

(m/z 227) and the loss of CO from [C16H21O2]+ (m/z 245) to

[C15H21O]+ (m/z 217) and from [C16H19O]+ (m/z 227) to [C15H19]+

(m/z 199). Further fragmentations are described in two pathways

from [C16H19O]+ (m/z 227) and from [C15H21O]+ (m/z 217) in

Fig. 2. Fang et al.63 reported other fragmentation pathways of

brefeldin A. The MFs of fragment ions at m/z 227, 217, 199, and

185 are different from ours. If we use their MFA, the fragment

peaks at 227, 217, 199, and 185 m/z have errors of 0.0005, 0.0370,

0.0368 and 0.0373 Da, respectively, which are highly unlikely

based on our mass measurement accuracy.

Molecules with single structural domain

The majority of small molecules consist of a single core structure.

An example is prazosin (C19H21N5O4, 383.1594 Da) for which

the result of the MFA is shown in Table 4. The MF of prazosin is

correctly assigned to C19H21N5O4. Some of the neutral losses,

which are underlined, are easily assigned to the losses of methane

(CH4), water (H2O), and furan (C4H4O), suggesting sub-struc-

tures of CH3–(O or N)– and furan. Since the fragmentation

producing free radicals hardly occurs in the ESI Q-TOF instru-

ment, the software does not accept the free radical fragmenta-

tions. As a result, MF of the peak at m/z 232, which was

incorrectly assigned to C13H14NO3+, required manual correction

to a free radical of C11H11N4O2c based on the structure of pra-

zosin (bold in Table 4). Among the fragmentations of the 96

compounds, this was the only case that produced a free radical. It

should be emphasized that the software works even if the MFs of

a few fragment ions are incorrectly assigned. As the observed m/z

values of one or a few peaks may have errors larger than the cut-

off error, the software was designed to tolerate such errors.


Table 4 MFA of prazosin

MMobs 383.1581 367.1291 365.1492 349.1183 246.1096 231.0888 230.0805 163.0638 137.0474

— C19H21N5O4 C18H17N5O4 C19H19N5O3 C18H15N5O3 C12H14N4O2 C11H11N4O2ca C11H10NO2 C9H9NO2 C7H7NO2

C19H21N5O4 — CH4 H2O CH6O C7H7NO2 C8H10NO2c C8H11NO2 C10H12N4O2 C12H14N4O2

C18H17N5O4 — H2O C6H4NO2 C7H6NO2c C7H7NO2 C9H8N4O2 C11H10N4O2

C19H19N5O3 — CH4 C7H5NO C8H8NOc C8H9NO C10H10N4O C12H12N4OC18H15N5O3 — C6HNO C7H4NOc C7H5NO C9H6N4O C11H8N4OC12H14N4O2 — CH3c CH4 C3H5N3 C5H5N3

C11H11N4O2ca — C2H2Oc C4H4Oc

C11H10N4O2 — C2HN3 C4H3N3

C9H9NO2 — C2H2

C7H7NO2 —

a Initial incorrect assignment to C13H13NO3 required manual correction to C11H11N4O2c.

Those few incorrectly assigned MFs are easily identified and are

corrected manually, once the MFs of other fragment and

precursor ions are assigned.

Prazosin is predominantly split into two fragment ions with

a core of C12H15N4O2+ and a side chain of C7H8NO2

+ through

the cleavage of two C–N bonds (Fig. 3). The core C12H15N4O2+

loses a methyl free radical to form C11H11N4O2c. Another type of

two C–N bond cleavage of prazosin produces C9H10NO2+ and

C10H12N4O2. It is not clear why the charge is localized to

C9H10NO2+ rather than C10H12N4O2 when the charge is easily

localized on the similar fragment C12H15N4O2+ (m/zobs ¼

247.1214). Other plausible fragmentation pathways of the frag-

ment ions used in the MFA are shown in Fig. 3. Erve et al.64

reported the fragmentation pathway of prazosin that is in

agreement with ours. The major structural difference is the

fragment peak at 231 m/z. Their fragmentation includes

a cleavage of a relatively stable C–C bond, whereas our analysis

includes the cleavage of the less stable heteroatom C–O bond. It

should be mentioned that the fragmentation pathways are

analyzed using the fragment ions used in the MFA. Other peaks

could be incorporated to get more detailed analysis of the frag-

mentation pathways; however, they are all speculative and are

not worthwhile unless required.

Fig. 3 Proposed fragmentation pathways of prazosin.


Molecules with multiple core structures

Some molecules consist of two or more core structures. An

example is Taxol in which the molecule splits into two core

structures in the MS/MS fragmentation as described in our

previous paper.53 Another example is ergot alkaloid dihydro-

ergotamine (C33H37N5O5, 583.1795 Da) where the result of the

MFA of dihydroergotamine is shown in Table 5. The analysis

correctly assigns the MF to C33H37N5O5. However, the MF of

a peak at m/z 270 was incorrectly assigned to C11H19N5O3, as the

dehydrated dihydroergotamine split into a fragment ion

C17H17N2O3+ (m/z 297) and a neutral loss C16H19N3O (269 Da).

The peak at m/z 270 must be the protonated form of C16H19N3O.

Thus the MF of the peak at m/z 270 was corrected to

C16H20N3O+ (bold in Table 5). Some small neutral losses

underlined in Table 5 are easily assigned to the losses of water

(H2O), carbon monoxide (CO), carbon dioxide (CO2), and

ammonia (NH3).

The plausible fragmentation pathway of dihydroergotamine is

shown in Fig. 4. The precursor molecule is in ketal–keto equi-

librium. Both forms undergo a cleavage at an amide bond in the

linker, splitting the precursor into an oxonium ion C16H17N2O+

(m/z 253) and a neutral molecule C17H21N3O4. Cleavage of the

keto form at a C–N bond of the linker splits the precursor ion

into a fragment ion C16H20N3O+ (m/z 270) and a neutral mole-

cule C17H18N2O4. The ketal form loses formic acid to form

C32H36N5O3+ (m/z 538). The keto form similarly loses water to

form C33H36N5O4+ (m/z 566). The linkers of these fragment ions

C32H36N5O3+ (m/z 538) and C33H36N5O4

+ (m/z 566) are cleaved

at each of two amide bonds and a C–N bond, generating frag-

ment ions C19H20N3O2+ (m/z 322), C16H17N2O+ (m/z 253), and

C16H20N3O+ (m/z 270), respectively. The cleavage of the C–N

bond of the dehydrated fragment ion C33H36N5O4+ (m/z 566)

also generates a fragment ion C17H17N2O3+.

Analysis of structurally-related compounds

An extension of the use of MFA is in analyzing structurally-

related compounds. For example, drugs are metabolized in vivo

and some metabolites have biological activities and/or toxicity

that may cause adverse effects. Since metabolites are structurally

related with minor chemical modifications, MFA of metabolites

and their fragment ions can provide some structural information

to identify/characterize them. The method can also be used to

Analyst, 2009, 134, 690–700 | 695

Table 5 MFA of dihydroergotamine

MMobs 583.2810 565.2702 537.2753 321.1477 296.1163 269.1504 252.1260

— C33H37N5O5 C33H35N5O4 C32H35N5O3 C19H19N3O2 C17H16N2O3 C16H19N3Oa C16H16N2O

C33H37N5O5 — H2O CH2O2 C14H18N2O3 C16H21N3O2 C17H18N2O4 C17H21N3O4

C33H35N5O4 — CO C14H16N2O2 C16H19N3O C17H16N2O3 C17H19N3O3

C32H35N5O3 — C13H16N2O C15H19N3 C16H16N2O2 C16H19N3O2

C19H19N3O2 — C3O C3H3NOC17H16N2O3 — CO2

C16H19N3Oa — NH3

C16H16N2O —

a Initial incorrect assignment to C11H19N5O3 required manual correction to C16H19N3O.

Fig. 4 Proposed fragmentation pathways of dihydroergotamine. The

fragment ions that are also observed in the MS/MS spectrum of dihy-

droergocristine are shown in red. The precursor ion and the fragment

ions that are different from those in the MS/MS spectrum of dihy-

droergocristine are shown in blue. The fragment ion corresponding to the

one in black was not observed in the MS/MS spectrum of dihy-

droergocristine.

Fig. 5 Structures of dihydroergotamine and dihydroergocristine. The

structural difference is encircled in red.

follow a series of chemical modifications of natural products in

vivo such as in microbial transformation and others.

As a test of the concept, we first examined dihydroergotamine

and dihydroergocristine. These are not metabolically related but

they are structurally close homologues. Table 6 lists the

precursor and MS/MS fragment ions of dihydroergocristine. The

Table 6 MFA of dihydroergocristine

MMobs 611.3123 593.3027 349.1784

— C35H41N5O5 C35H39N5O4 C21H23NC35H41N5O5 — H2O C14H18NC35H39N5O4 — C14H16NC21H23N3O2 —C19H20N2O3

C16H19N3Oa

C16H16NO2

a Initial incorrect assignment to C11H19N5O3 required manual correction to

696 | Analyst, 2009, 134, 690–700

MF of a fragment ion at m/z 594 was not assigned in Reverse

Analysis; however, the neutral loss of 18.0096 Da was manually

assigned to H2O (MMcalc ¼ 18.0106 Da), allowing the assign-

ment of the MF of the fragment ion to C35H39N5O4 (bold in

Table 6).

With the plausible assumption of comparable fragmentation

pathways between these related compounds, we note the exis-

tence of two common fragment ions at m/z 270 and 253. We then

hypothesize that these correspond to the same molecular entities

in both spectra and conclude that these peaks in the dihy-

droergocristine spectrum correspond to the structures colored

red in Fig. 4. The precursor and fragment ions of dihy-

droergocristine at m/z 612, 594, 350, 325 have an extra C2H4 with

respect to the precursor and fragment ions of dihydroergotamine

at m/z 584, 566, 322, 297, respectively (fragment ion structures

shown in blue color in Fig. 4. The structures in Fig. 5 imply that

the extra C2H4 is attached to the methyl group of dihydroer-

gotamine highlighted in Fig. 5. This suggests that in dihy-

droergocristine, that methyl group is either a n-propyl group or

an isopropyl group. In fact, dihydroergocristine has an isopropyl

group.

324.1481 269.1506 252.1265

3O2 C19H20N2O3 C16H19N3Oa C16H16NO2

2O3 C16H21N3O2 C19H22N2O4 C19H25N3O4

2O2 C16H19N3O C19H20N2O3 C19H23N3O3

C5H7NO— C3H4O2

——

C16H19N3O.


Table 7 MFA of cyclazocine

MMobs 271.1928 229.1472 217.1466 174.1043 172.0883 158.0729

— C18H25NO C15H19NO C14H19NO C12H14O C12H12O C11H10OC18H25NO — C3H6 C4H6 C6H11N C6H13N C7H15NC15H19NO — C3H5N C3H7N C4H9NC14H19NO — C2H5N C2H7N C3H9NC12H14O — H2 CH4

C12H12O — CH2

C11H10O —

Table 8 MFA of N-allylnormetazocine

MMobs. 257.1773 174.1041 172.0876 158.0725

— C17H23NO C12H14O C12H12O C11H10OC17H23NO — C5H9N C5H11N C6H13NC12H14O — CH4

C12H12O — CH2

C11H10O —

Cyclazocine and N-allylnormetazocine

Another example of metabolites is cyclazocine and N-allylnor-

metazocine, where we take cyclazocine as a known compound

and N-allylnormetazocine as its analog. Tables 7 and 8 list the

precursor and fragment ions of benzomorphan opioids cyclaz-

ocine and N-allylnormetazocine, respectively.

The plausible fragmentation pathway of cyclazocine is shown

in Fig. 6. The fragment ions of cyclazocine at m/z 175, 173 and

159 are the same as those of N-allylnormetazocine. Another

fragment ion at m/z 216, which was not used in the MFA, is also

common and is added in the fragmentation pathway. The MF

difference between cyclazocine and N-allylnormetazocine is CH2.

Since the peak at m/z 216 is the largest fragment ion of N-

allylnormetazocine, it is very likely that the nitrogen is alkylated

in a similar way as the cyclopropylmethyl group of cyclazocine

with a cyclopropyl or allyl group. Indeed, N-allylnormetazocine

has an N-allyl group.

Fig. 6 Proposed fragmentation pathways of cyclazocine. The fragment

ions in common with N-allylnormetazocine are shown in red. The

molecular ions in blue or black are uncommon with N-allylnormetazo-

cine.


Peptides

The MFA of peptides is useful for proteomics and identification

of bioactive peptide metabolites. 5-Leucine enkephalin was

analyzed as an example (ESI,† Table S1). Fourteen fragment

ions were used for the analysis, more than used for small organic

molecules as the analysis was extended from MFA to peptide

sequencing. The MFs of the fragment ions at m/z 538 and 510

were assigned manually (shown in bold) because the observed

m/z values have errors larger than the allowed cut-off error of

0.002 Da. The MFs of the neutral losses that are assigned

manually are also shown in bold.

The use of MFA for peptides was not the final goal. Proteo-

mics requires further sequence analysis, which was carried out

with the following stepwise analysis of the MS/MS data. The first

step is the analysis of the amino acid composition of the peptide,

Fig. 7 Stepwise analysis of 5-leucine enkephalin sequences.

Analyst, 2009, 134, 690–700 | 697

resulting in Gly, Leu, Phe and Tyr. Leu could be Ile as they are

not distinguished by our MS/MS fragmentation energy. Other

amino acids are not found in the neutral losses. Adding the

molecular masses of the four amino acid residues and H2O for

the C-terminal residue resulted in the molecular mass of (2Gly +

Leu + Phe + Tyr + H2O) matching the precursor molecule. Thus,

the amino acid composition of the peptide is 2Gly, Leu (or Ile),

Phe, and Tyr with a C-terminal free carboxyl group. The MFs of

the fragment ions are assigned to the amino acid residues, H2O,

CO, NH3, and HCO2H reflecting different amide bond cleavage

sites and N- or C-terminal residues. The MF of the fragment

C11H12N2O2 (204 Da) is, for example, assigned to (Gly + Phe)

and other assignments are listed in the second row from the

bottom (see ESI,† Table S1). The bottom row of Table S1 lists

the amino acid residues, H2O, CO, NH3 and HCO2H lost from

the precursor molecule to form the corresponding fragment ions.

Alternatively, the amino acid composition can be obtained by

tracing fragmentation pathways, which release amino acid resi-

dues (or di-peptides). In the case of 5-leucine enkephalin, one

such fragmentation pathway is shown in Fig. 7A, giving

a composition of 2Gly, Leu, Phe, Tyr and H2O. The second step

is the identification of the N- or C-terminal residues (Fig. 7B).

Since Leu and Tyr are the first amino acid residues that can be

fragmented off from the precursor ion, they are the N- or

C-terminal residues. The third step is the identification of the

C-terminal residue. As peptides often have a free carboxyl group

Fig. 8 Overall detail of analysis of 5-leucine enkephalin.

Table 9 MFA of quinacrine

MMobs 399.2080 326.1167

— C23H30ClN3O C19H19ClN2OC23H30ClN3O — C4H11NC19H19ClN2O —C14H11ClN2OC14H10ClNOC9H19N

698 | Analyst, 2009, 134, 690–700

or are blocked with an amide, fragment ions that contain Leu-

OH, Leu-NH2, Tyr-OH, or Tyr-NH2 were searched, resulting in

a fragment ion of (Phe, Leu)-OH. Therefore, Leu was assigned to

the C-terminal residue, and, automatically, Tyr was assigned to

the N-terminal residue. The fourth step is the sequencing of

internal amino acid residues. Starting from the C-terminal Phe-

Leu-OH fragment ion, in the fragmentation pathway amino acid

residues are sequentially added (or di-peptides if necessary) as of

Fig. 7C. Similarly, starting from the N-terminal Tyr-Gly frag-

ment ion, amino acid residues (or di-peptides if necessary) are

sequentially added as shown in Fig. 7D.

The final step applies the sequence Tyr-Gly-Gly-Phe-Leu-OH

to all fragment ions in order to confirm the sequence. Each line

shown in Fig. 8 represents the sequence region of the fragment of

which its MF is shown in the left column. As all of the fragments

used in the MFA fit to the sequence, the Tyr-Gly-Gly-Phe-Leu-

OH sequence is confirmed.

Chloro- or bromo-containing compounds

The MFA was extended to organic compounds that contain

chlorine and/or bromine atoms in addition to C, H, N and O.

The inclusion of chlorine and bromine atoms in the automated

MFA is not possible with the precision of 0.002 Da. However, as

chlorine and bromine have characteristic isotopes of 37Cl (24.47%

natural abundance) and 81Br (49.48% natural abundance),

chlorine and/or bromine atoms in precursor and fragment ions

estimated based on their isotope peaks are replaced with

hydrogen atom(s). The molecular masses of these ions are

modified accordingly and are submitted to the MFA. Chlorine

258.0573 243.0446 141.1510

C14H11ClN2O C14H10ClNO C9H19NC9H19N C9H20N2 C14H11ClN2OC5H8 C5H9N C10ClNO—

——

Fig. 9 Plausible fragmentation pathways of quinacrine, where the

cleavages occurred at C–N bonds.


and/or bromine atoms are then restored to the MFs manually.

The method was applied to five compounds beclometasone

dipropionate (C28H37ClO7, 520.2228 Da), benzamil

(C13H14ClN7O, 319.0948 Da), brimonidine (C11H10BrN5,

291.0120 Da), phenoxybenzamine (C18H22ClNO, 303.1390 Da),

and quinacrine (C23H30ClN3O, 399.2077 Da), resulting in correct

MF assignments for all of them.

Table 9 shows the MFA after restoring a chlorine atom and

Fig. 9 shows the chemical fragmentation pathway of quinacrine

based on the analysis in Table 9. Since 5 compounds containing

Cl or Br may not be enough to generalize the analysis, these

compounds are not included in the 96 compounds analyzed in

this article.

Conclusions

The automated software developed to determine the MFs (C, H,

N, and O) of precursor molecules demonstrated a high success

rate of 95%. The software also determined the MFs of fragment

ions and neutral losses. Although a few MFs of the fragment ions

were incorrectly assigned due to the large errors of the observed

m/z values, the software managed to get the correct MF of the

precursor ion. The incorrectly assigned MFs of fragment ions

and neutral losses were then easily corrected manually. Using the

MFs of the precursor ions, fragment ions and neutral losses,

plausible fragmentation pathways were estimated. These were

used for the structural characterization of homologous

compounds with possible applications to dereplication of natural

products and identification of drug metabolites. The analysis was

further successfully extended to compounds containing Cl and

Br.

Acknowledgements

We acknowledge Beata Usakiewicz for her technical assistance,

and Misato Konishi for her useful suggestions. S. A. was sup-

ported by a McGill Chemical Biology Scholarship. NRCC

Publication 50655.

References

1 D. J. Newman, J. Med. Chem., 2008, 51, 2589–2599.2 F. E. Koehn and G. T. Carter, Nat. Rev. Drug Discovery, 2005, 4, 206–

220.3 A. L. Harvey, Drug Discovery Today, 2008, 13, 894–901.4 T. Beghyn, R. Deprez-Poulain, N. Willand, B. Folleas and B. Deprez,Chem. Biol. Drug Des., 2008, 72, 3–15.

5 G. M. Cragg and D. J. Newman, Pure Appl. Chem., 2005, 77, 7–24.6 F. E. Koehn, Prog. Drug Res., 2008, 65, 176–210.7 G. A. Cordell and Y. G. Shin, Pure Appl. Chem., 1999, 71, 1089–1094.8 K. Hostettmann, A. Marston and J.-L. Wolfender, Chimia, 2005, 59,

291–294.9 L. Dinan, in Natural Products Isolation (Methods in Biotechnology),

ed. S. D. Sarker, Z. Latif and A. I. Gray, Humana Press, Totowa,NJ, 2nd edn, 2005, vol. 20, pp. 297–321.

10 K. B€oer€oeczky, H. Laatsch, I. Wagner-D€obler, K. Stritzke andS. Schulz, Chem. Biodiversity, 2006, 3, 622–634.

11 J. Bitzer, B. K€opeke, M. Stadler, V. Hellwig, Y.-M. Ju, S. Seip andT. Henkel, Chimia, 2007, 61, 332–338.

12 K. U. Bindseil, J. Jakupovic, D. Wolf, J. Lavayre, J. Leboul andD. van der Pyl, Drug Discovery Today, 2001, 6, 840–847.

13 J. Bradshaw, D. Butina, A. J. Dunn, R. H. Green, M. Hajek,M. M. Jones, J. C. Lindon and P. J. Sidebottom, J. Nat. Prod.,2001, 64, 1541–1544.


14 J. L. L�opez-P�erez, R. Ther�on, E. del Olmo and D. Dı́az,Bioinformatics, 2007, 23, 3256–3257.

15 I. D. Wilson and U. A. Th. Brinkman, J. Chromatogr., A, 2003, 1000,325–356.

16 M. E. Hansen, J. Smedsgaard and T. O. Larsen, Anal. Chem., 2005,77, 6805–6817.

17 C. A. Nesbitt, H. Zhang and K. K.-C. Yeung, Anal. Chim. Acta, 2008,627, 3–24.

18 E. Edwards and J. Thomas-Oates, Analyst, 2005, 130, 13–17.19 X.-G. He, J. Chromatogr., A, 2000, 880, 203–232.20 T. Murakami, N. Fukutsu, J. Kondo, T. Kawasaki and F. Kusu,

J. Chromatogr., A, 2008, 1181, 67–76.21 M. Sandvoss, L. H. Pham, K. Levsen, A. Preiss, C. M€ugge and

G. W€unsch, Eur. J. Org. Chem., 2000, 7, 1253–1262.22 M. Lambert, D. Strk, S. H. Hansen, M. Sairafianpour and

J. W. Jaroszewski, J. Nat. Prod., 2005, 68, 1500–1509.23 D. Louden, A. Handley, S. Taylor, E. Lenz, S. Miller, I. D. Wilson,

A. Sage and R. Lafont, J. Chromatogr., A, 2001, 910, 237–246.24 M. Pepaj, S. R. Wilson, K. Novotna, E. Lundanes and T. Greibrokk,

J. Chromatogr., A, 2006, 1120, 132–141.25 R. L. Sleighter and P. G. Hatcher, J. Mass Spectrom., 2007, 42, 559–

574.26 G. Chen, B. N. Pramanik, Y.-H. Liu and U. A. Mirza, J. Mass

Spectrom., 2007, 42, 279–287.27 A. Nita-Lazar, H. Saito-Benz and F. M. White, Proteomics, 2008, 8,

4433–4443.28 A. Espada, M. Molina-Martin, J. Dage and M.-S. Kuo, Drug

Discovery Today, 2008, 13, 417–423.29 R. T. Kurulugama, S. J. Valentine, R. A. Sowell and D. E. Clemmer,

J. Proteomics, 2008, 71, 318–331.30 X. Han, A. Aslanian and J. R. Yate III, Curr. Opin. Chem. Biol., 2008,

12, 483–490.31 C.-H. (Winston) Chen, Anal. Chim. Acta, 2008, 624, 16–36.32 T. Sangster, H. Major, R. Plumb, A. J. Wilson and I. D. Wilson,

Analyst, 2006, 131, 1075–1078.33 X. Lu, X. Zhao, C. Bai, C. Zhao, G. Lu and G. Xu, J. Chromatogr., B,

2008, 866, 64–76.34 J. Trygg, E. Holmes and T. Lundstedt, J. Proteome Res., 2007, 6, 469–

479.35 E. Werner, J.-F. Heilier, C. Ducruix, E. Ezan, C. Junot and

J.-C. Tabet, J. Chromatogr., B, 2008, 871, 143–163.36 D. J. Crockford, A. D. Maher, K. R. Ahmadi, A. Barrett,

R. S. Plumb, I. D. Wilson and J. K. Nicholson, Anal. Chem., 2008,80, 6835–6844.

37 M. Katajamaa and M. Ore�si�c, J. Chromatogr., A, 2007, 1158, 318–328.

38 S. Mahadevan, S. L. Shah, T. J. Marrie and C. M. Slupsky, Anal.Chem., 2008, 80, 7562–7570.

39 W. B. Dunn, Phys. Biol., 2008, 5, 1–24.40 I. Spasi�c, D. Schober, S.-A. Sansone, D. Rebholz-Schuhmann,

D. B. Kell and N. W. Paton, BMC Bioinformatics, 2008, 9(Suppl.5), 1–16.

41 M. Oldiges, S. L€utz, S. Pflug, K. Schroer, N. Stein and C. Wiendahl,Appl. Microbiol. Biotechnol., 2007, 76, 495–511.

42 W. A. Korfmacher, Drug Discovery Today, 2005, 10, 1357–1367.43 C. Petucci and L. Mallis, Rapid Commun. Mass Spectrom., 2005, 19,

1492–1498.44 O. Potterat, K. Wagner and H. Haag, J. Chromatogr., A, 2000, 872,

85–90.45 R. S. Plumb, K. A. Johnson, P. Rainville, B. W. Smith, I. D. Wilson,

J. M. Castro-Perez and J. K. Nicholson, Rapid Commun. MassSpectrom., 2006, 20, 1989–1994.

46 A. Fredenhagen, C. Derrien and E. Gassmann, J. Nat. Prod., 2005,68, 385–391.

47 A. W. T. Bristow, Mass Spectrom. Rev., 2006, 25, 99–111.48 A. C. Aiken, P. F. DeCarlo and J. L. Jimenez, Anal. Chem., 2007, 79,

8350–8358.49 I. V. Chernushevich, A. V. Loboda and B. A. Thomson, J. Mass

Spectrom., 2001, 36, 849–865.50 L.-K. Zhang, D. Rempel, B. N. Pramanik and M. L. Gross, Mass

Spectrom. Rev., 2005, 24, 286–309.51 A. W. T. Bristow and K. S. Webb, J. Am. Soc. Mass Spectrom., 2003,

14, 1086–1098.52 B. P. Koch, T. Dittmar, M. Witt and G. Kattner, Anal. Chem., 2007,

79, 1758–1763.

Analyst, 2009, 134, 690–700 | 699

53 Y. Konishi, T. Kiyota, C. Draghici, J.-M. Gao, F. Yeboah, S. Acoca,S. Jarussophon and E. Purisima, Anal. Chem., 2007, 79, 1187–1197.

54 C. Eckers, J.-C. Wolff, N. J. Haskins, A. B. Sage, K. Giles andR. Bateman, Anal. Chem., 2000, 72, 3683–3688.

55 J.-C. Wolff, C. Eckers, A. B. Sage, K. Giles and R. Bateman, Anal.Chem., 2001, 73, 2605–2612.

56 J.-C. Wolff, T. R. Fuentes and J. Taylor, Rapid Commun. MassSpectrom., 2003, 17, 1216–1219.

57 K. F. Blom, Anal. Chem., 2001, 73, 715–719.58 M. Colombo, F. R. Sirtori and V. Rizzo, Rapid Commun. Mass

Spectrom., 2004, 18, 511–517.59 J. Wu and H. McAllister, J. Mass Spectrom., 2003, 38, 1043–1053.

700 | Analyst, 2009, 134, 690–700

60 G. Viv�o-Truyols and P. J. Schoenmakers, Anal. Chem., 2006, 78,4598–4608.

61 L. Sleno, D. A. Volmer and A. G. Marshall, J. Am. Soc. MassSpectrom., 2005, 16, 183–198.

62 K. Clauwaert, S. V. Casteele, B. Sinnaeve, D. Deforce, W. Lambert,C. Van Peteghem and J. Van Bocxlaer, Rapid Commun. MassSpectrom., 2003, 17, 1443–1448.

63 M. Fang, J. Wang, Y. Huang and Y. Zhao, Front. Chem. China, 2006,1, 15–19.

64 J. C. L. Erve, S. C. Vashishtha, O. Ojewoye, A. Adedoyin,R. Espina, W. DeMaio and R. E. Talaat, Xenobiotica, 2008, 38,540–558.


automated molecular formula determination by tandem mass spectrometry (ms/ms

Documents