Determination of the fatty acyl profiles of phosphatidylethanolamines by tandem mass spectrometry of...

7
Determination of the fatty acyl profiles of phosphatidyl- ethanolamines by tandem mass spectrometry of sodium adducts Cla ´ udia Simo ˜ es, Vanda Simo ˜ es, Ana Reis, Pedro Domingues and M. Rosa ´rio M. Domingues * Mass Spectrometry Centre, Department of Chemistry, University of Aveiro, 3810-193 Aveiro, Portugal Received 30 June 2008; Revised 20 August 2008; Accepted 21 August 2008 Phosphatidylethanolamines (PEs) are one of the major constituents of cellular membranes, and, along with other phospholipid classes, have an essential role in the physiology of cells. Profiling of phospholipids in biological samples is currently done using mass spectrometry (MS). In this work we describe the MS fragmentation of sodium adducts of 2-oleoyl-1-palmitoyl-sn-glycero-3-phos- phatidylethanolamine (POPE) and 2-linoleoyl-1-palmitoyl-sn-glycero-3-phosphatidylethanolamine (PLPE). This study was performed by electrospray ionization tandem mass spectrometry (ESI-MS/ MS) using three different instruments and also by matrix-assisted laser desorption/ionization tandem mass spectrometry (MALDI-MS/MS). All MS/MS spectra show product ions related to the polar head fragmentation and product ions related to the loss of acyl chains. In ESI-MS/MS spectra, the product ions [MRNa-R 1 COOH-43] R and [MRNa-R 2 COOH-43] R show different relative abundance, as well as [MRNa-R 1 COOH] R and [MRNa-R 2 COOH] R product ions, allowing identi- fication of both fatty acyl residues of PEs, and their specific location. MALDI-MS/MS shows the same product ions reported before and other ions generated by charge-remote fragmentation of the C3–C4 bond (g -cleavage) of fatty acyl residues combined with loss of 163 Da. These fragment ions, [MRNa- (R 2 -C 2 H 3 )-163] R and [MRNa-(R 1 -C 2 H 3 )-163] R , show different relative abundances, and the product ion formed by the g -cleavage of sn-2 is the most abundant. Overall, differences noted that are important for identification and location of fatty acyl residues in the glycerol backbone are: relative abundance between the product ions [MRNa-R 1 COOH-43] R > [MRNa-R 2 COOH-43] R in ESI-MS/MS spectra; and relative abundance between the product ions [MRNa-(R 2 -C 2 H 3 )-163] R > [MRNa-(R 1 - C 2 H 3 )-163] R in MALDI-MS/MS spectra. Copyright # 2008 John Wiley & Sons, Ltd. Phosphatidylethanolamines (PEs) are major constituents of cell membranes and are also components of lipoproteins. 1,2 They take part in cell signaling and play different roles in biological systems and in physiopathologic processes, such as apoptosis, inflammation, aging, and diseases associated with oxidative stress. 1 The distribution of phospholipids is not homogeneous, and tissues and organelles have different amounts of phospholipids from different classes. These also differ in the composition of acyl chains, at the sn-1 and sn-2 positions in the glycerol backbone. 3 The profile of the fatty acyl residue composition of phospholipids may change substantially with external factors such as dietary fatty acids, 4 exercise training, 5 body size, 6 and cold, 7 among others. The fatty acyl residue profile influences membrane properties and functions, conditioning its susceptibility to oxidative modifications, and may have an impact on phospholipid functions and cell signaling. 3,8–11 Lipidomics is the large-scale study of lipids and phos- pholipids in biological samples. 12,13 Mass spectrometry (MS) is a key technology used in lipidomics research in identifying the phospholipid class and the fatty acyl substitutents. 12 Most of the published works in the structural characteriz- ation of PEs by MS use electrospray ionization tandem mass spectrometry (ESI-MS/MS) to study the protonated mol- ecules: 12,14,15 The product ion formed by the loss of polar head, [MþH–141Da] þ , is diagnostic for this class of phospholipids. 12,15 Studies involving PE lithium adducts under ESI-MS/MS conditions 16 show more product ions, allowing detailed structural information, namely identifi- cation of fatty acyl residues and their location in the glycerol backbone. However, there is only one study (published in 1995) reporting the fragmentation of PE sodiated adducts 17 in ESI-MS/MS, acquired using a triple quadrupole mass spectrometer. Fragmentation of [M–H] was described using ESI-MS/MS 18 and the authors focused on studying product ions generated from fragmentation involving the polar head. To our knowledge, there are no studies focusing on the RAPID COMMUNICATIONS IN MASS SPECTROMETRY Rapid Commun. Mass Spectrom. 2008; 22: 3238–3244 Published online in Wiley InterScience (www.interscience.wiley.com) DOI: 10.1002/rcm.3727 *Correspondence to: M. R. M. Domingues, Department of Chem- istry, University of Aveiro, 3810-193 Aveiro, Portugal. E-mail: [email protected] Contract/grant sponsor: FEDER, FCT-Portugal, Research Unit 62/94 ‘Quı ´mica Orga ˆnica, Produtos Naturais e Agro-Alimen- tares’, Aveiro. Copyright # 2008 John Wiley & Sons, Ltd.

Transcript of Determination of the fatty acyl profiles of phosphatidylethanolamines by tandem mass spectrometry of...

Page 1: Determination of the fatty acyl profiles of phosphatidylethanolamines by tandem mass spectrometry of sodium adducts

RAPID COMMUNICATIONS IN MASS SPECTROMETRY

Rapid Commun. Mass Spectrom. 2008; 22: 3238–3244

) DOI: 10.1002/rcm.3727

Published online in Wiley InterScience (www.interscience.wiley.com

Determination of the fatty acyl profiles of phosphatidyl-

ethanolamines by tandem mass spectrometry of

sodium adducts

Claudia Simoes, Vanda Simoes, Ana Reis, Pedro Domingues and

M. Rosario M. Domingues*Mass Spectrometry Centre, Department of Chemistry, University of Aveiro, 3810-193 Aveiro, Portugal

Received 30 June 2008; Revised 20 August 2008; Accepted 21 August 2008

*Correspoistry, UniE-mail: mContract/62/94 ‘Qtares’, Av

Phosphatidylethanolamines (PEs) are one of the major constituents of cellular membranes, and,

along with other phospholipid classes, have an essential role in the physiology of cells. Profiling of

phospholipids in biological samples is currently done using mass spectrometry (MS). In this work

we describe the MS fragmentation of sodium adducts of 2-oleoyl-1-palmitoyl-sn-glycero-3-phos-

phatidylethanolamine (POPE) and 2-linoleoyl-1-palmitoyl-sn-glycero-3-phosphatidylethanolamine

(PLPE). This study was performed by electrospray ionization tandem mass spectrometry (ESI-MS/

MS) using three different instruments and also by matrix-assisted laser desorption/ionization

tandem mass spectrometry (MALDI-MS/MS). All MS/MS spectra show product ions related to

the polar head fragmentation and product ions related to the loss of acyl chains. In ESI-MS/MS

spectra, the product ions [MRNa-R1COOH-43]R and [MRNa-R2COOH-43]R show different relative

abundance, as well as [MRNa-R1COOH]R and [MRNa-R2COOH]R product ions, allowing identi-

fication of both fatty acyl residues of PEs, and their specific location.MALDI-MS/MS shows the same

product ions reported before and other ions generated by charge-remote fragmentation of the C3–C4

bond (g-cleavage) of fatty acyl residues combined with loss of 163Da. These fragment ions, [MRNa-

(R2-C2H3)-163]R and [MRNa-(R1-C2H3)-163]

R, show different relative abundances, and the product

ion formed by the g-cleavage of sn-2 is the most abundant. Overall, differences noted that are

important for identification and location of fatty acyl residues in the glycerol backbone are: relative

abundance between the product ions [MRNa-R1COOH-43]R> [MRNa-R2COOH-43]R in ESI-MS/MS

spectra; and relative abundance between the product ions [MRNa-(R2-C2H3)-163]R> [MRNa-(R1-

C2H3)-163]R in MALDI-MS/MS spectra. Copyright # 2008 John Wiley & Sons, Ltd.

Phosphatidylethanolamines (PEs) are major constituents of

cell membranes and are also components of lipoproteins.1,2

They take part in cell signaling and play different roles in

biological systems and in physiopathologic processes, such

as apoptosis, inflammation, aging, and diseases associated

with oxidative stress.1 The distribution of phospholipids is

not homogeneous, and tissues and organelles have different

amounts of phospholipids from different classes. These also

differ in the composition of acyl chains, at the sn-1 and sn-2

positions in the glycerol backbone.3 The profile of the fatty

acyl residue composition of phospholipids may change

substantially with external factors such as dietary fatty

acids,4 exercise training,5 body size,6 and cold,7 among

others. The fatty acyl residue profile influences membrane

properties and functions, conditioning its susceptibility to

ndence to: M. R. M. Domingues, Department of Chem-versity of Aveiro, 3810-193 Aveiro, [email protected] sponsor: FEDER, FCT-Portugal, Research Unituımica Organica, Produtos Naturais e Agro-Alimen-eiro.

oxidative modifications, and may have an impact on

phospholipid functions and cell signaling.3,8–11

Lipidomics is the large-scale study of lipids and phos-

pholipids in biological samples.12,13 Mass spectrometry (MS)

is a key technology used in lipidomics research in identifying

the phospholipid class and the fatty acyl substitutents.12

Most of the published works in the structural characteriz-

ation of PEs by MS use electrospray ionization tandem mass

spectrometry (ESI-MS/MS) to study the protonated mol-

ecules:12,14,15 The product ion formed by the loss of polar

head, [MþH–141Da]þ, is diagnostic for this class of

phospholipids.12,15 Studies involving PE lithium adducts

under ESI-MS/MS conditions16 show more product ions,

allowing detailed structural information, namely identifi-

cation of fatty acyl residues and their location in the glycerol

backbone. However, there is only one study (published in

1995) reporting the fragmentation of PE sodiated adducts17

in ESI-MS/MS, acquired using a triple quadrupole mass

spectrometer. Fragmentation of [M–H]�was described using

ESI-MS/MS18 and the authors focused on studying product

ions generated from fragmentation involving the polar head.

To our knowledge, there are no studies focusing on the

Copyright # 2008 John Wiley & Sons, Ltd.

Page 2: Determination of the fatty acyl profiles of phosphatidylethanolamines by tandem mass spectrometry of sodium adducts

Determination of fatty acyl profiles of PEs 3239

fragmentation of PE sodium adducts under ESI-MS/MS not

using either a Q-TOF or ion trap instrument.

More recently, the use of matrix-assisted laser desorption/

ionization (MALDI) in the analysis of phospholipids has

become popular, showing an increased interest in this

subject,13,19–21 and, as a result of this interest, some authors

began to study the fragmentation of phospholipids, and

especially PE ions produced under MALDI conditions.22–26

Interestingly, PEs ionize preferentially as [MþNa]þ adducts

when using the most common matrix for the lipid analysis,

2,5-dihydroxybenzoic acid (DHB)13,19,20,24–26 or alpha-cyano-

4-hydroxycinnamic acid.23 Ionization as [MþH]þ and [M–

Hþ2Na]þ ions also occurs, but usually with lower relative

abundance. This is in contrast with ESI, where [MþH]þ ions

predominate, although the [MþNa]þ ions are also observed

whose relative abundance depends on the experimental

conditions used.15,17 Few studies report on the fragmentation

of ions formed under MALDI, by post-source decay (PSD)

conditions.22,24,25 These studies revealed that MALDI-PSD

spectra of PE [MþNa]þ ions show more product ions, when

compared with the described fragmentation of PE [MþH]þ

ions, thus allowing more structural information. A recent

study, using MALDI seamless PSD of both lithium and

sodium adducts of PEs, reports on fragmentation of the polar

head for the [MþNa]þ ion.22 In addition, using MALDI-MS/

MS conditions, there is only one study reporting the

fragmentation of [M–H]� ions.27 To our knowledge, there

is no MALDI-MS/MS data for the PE sodium adducts.

In this work, we report the fragmentation of the [MþNa]þ

ions of POPE (1-palmitoyl-2-oleoyl-phosphatidylethanola-

mine) and PLPE (1-palmitoyl-2-linoleoyl-phosphatidyletha-

nolamine), achieved by collision-induced dissociation (CID).

For this purposewe used different ionizationmethodologies,

ESI (ESI-MS/MS) and MALDI (MALDI-MS/MS), and used

different mass spectrometers with different analyzers,

namely Q-TOF, ion trap, triple quadrupole and MALDI-

TOF/TOF instruments.

EXPERIMENTAL

MaterialsPhosphatidylethanolamines, 2-oleoyl-1-palmitoyl-sn-glycero-

3-Phosphatidylethanolamine (POPE) and 2-linoleoyl-1-

palmitoyl-sn-glycero-3-phosphatidylethanolamine (PLPE),

were purchased from Sigma-Aldrich (Madrid, Spain). The

MALDI matrix, 2,5-dihydroxybenzoic acid (DHB), was

bought from Fluka (Seelze, Switzerland). For increasing

sensitivity for the sodium adducts in the ESI-MS exper-

iments, we added 0.5mL of sodium carbonate (1mgmL�1) to

200mL of methanolic solutions of POPE and PLPE

(50mgmL�1).

InstrumentationStudieswere performed in the positivemode using aQ-TOF2

instrument (Micromass, Manchester, UK) and a triple

quadrupole (Micromass, Manchester, UK). Samples were

introduced into the ESI source at a flow rate of 10mLmin�1.

The cone voltage was 30V and capillary voltage 3 kV. Source

temperature was 808C and desolvation temperature 1508C.

Copyright # 2008 John Wiley & Sons, Ltd.

Tandem mass spectra (MS/MS) were acquired by collision-

induced dissociation (CID), using argon as the collision gas.

The collision energies used for POPE and PLPE were,

respectively, 30 and 32 eV. Data acquisition was carried out

using a MassLynx 4 data system.

When using a linear ion trap (LXQ; ThermoFinnigan, San

Jose, CA, USA), samples were introduced into the source at

flow rate of 8mLmin�1. Typical ESI conditions were:

nitrogen sheath gas 30 psi, spray voltage 5 kV, capillary

temperature 3508C, capillary voltage 21V and tube lens

voltage 40V. CID-MS/MS experiments were performed on

mass-selected precursor ions using standard isolation and

excitation configuration. The collision energies used for

POPE and PLPE were, respectively, 23 and 26 (arbitrary

units). Data acquisition was carried out with the Xcalibur

data system.

The MALDI-MS and MALDI-MS/MS spectra were

acquired using a MALDI-TOF/TOF Applied Biosystems

4800 Proteomics Analyzer (Applied Biosystems, Framing-

ham, MA, USA) instrument equipped with a nitrogen laser

emitting at 337 nm. A volume of 4mL of matrix, 2,5-

dihydroxybenzoic acid (DHB) (10mgmL�1 in methanol

and 0.1% TFA solution), was mixed with 2mL of PE solution

(10mgmL�1), and 0.5mL of this mixture was deposited on

the MALDI plate. Spectra were subsequently acquired in the

positive ion reflector mode using delayed extraction in the

mass range between 600 and 4500Da with ca. 1500 laser

shots. For the following acquisition of MS/MS spectra, a

collision energy of 2 keV was used to induce fragmentation

and air was used as collision gas. Data acquisition was

carried out using a 4000 Series Explorer data system (Applied

Biosystems, Framingham, MA, USA).

RESULTS AND DISCUSSION

The results reveal that the ESI mass spectra of PLPE and

POPE acquired in all three instruments without addition of

sodium salt (data not shown) have both the protonated

molecule, [MþH]þ, and sodium adduct, [MþNa]þ. How-

ever, the ESI-MS spectra acquired in samples with the

addition of the sodium salt (data not shown) show only

the [MþNa]þ ion, which is an advantage, because it increases

the sensitivity in the analysis. In MALDI-MS, without salt

addition, the [MþNa]þ ion is the most abundant ion when

using DHBmatrix, which is the most commonly used matrix

for the analysis of phospholipids.13 The [MþH]þ and [M–

Hþ2Na]þ ions are also observed, but with lower relative

abundance. The yield of the different adducts ([MþH]þ,

[MþNa]þ and [M–Hþ2Na]þ) observed in the MALDI-MS

spectra depends on the cation content of the applied

solutions as well as the pH. Therefore, the intensity ratios

between the individual adducts can be changed if the salt

concentration is changed. Under the conditions we have

used, where no sodium salts were added, the ratio between

the protonated and sodiated and disodiated molecular ions

were around 20:100:45. We have observed no significant

fragmentation in both the ESI-MS and MALDI-MS spectra.

The favorable formation of the sodium adducts by PEs, and

the evidence of phospholipid cation adducts, usually giving

more structural information,16,21 corroborates the need for a

Rapid Commun. Mass Spectrom. 2008; 22: 3238–3244

DOI: 10.1002/rcm

Page 3: Determination of the fatty acyl profiles of phosphatidylethanolamines by tandem mass spectrometry of sodium adducts

Table

1.Relative

abundances�(%

)ofthePEproductionsobservedin

themassspectraacquiredusingtheQ-TOF2,lineariontrap,QqQ

andMALDImassspectrometers

[PEhead

þNa-43

]þ[PEhead

þNa-H

2O]þ

[PEhead

þNa]

þ

[MþNa-

R1COOH-

R2COOH]þ

R1CO

þR2CO

þ

[MþNa-

R1COOH-

163]

þ

[MþNa-

R2COOH-

43]þ

[MþNa-

R1COOH-

43]þ

[MþNa-

R2– –C– –O-

43]þ

[MþNa-

R1– –C– –O-

43]þ

[MþNa-

R2COOH]þ

[MþNa-

R1COOH]þ

[MþNa-

163]

þ[M

þNa-

141]

þ[M

þNa-

123]

þ[M

þNa-

43]þ

[PLPEþNa]

þFragmen

tIons

121

146

164

182

239�

263�

319�

415�

439�

433�

457�

458�

482�

575

597

615

695

Q-ToF

<5

<5

65<5

3520

1055

100

4015

1020

4010

05

40IonTrap

——

——

<5

<5

<5

32.5

100

10<5

12.5

705

205

100

QqQ

255

100

<5

1510

—60

100

3010

—15

2045

<5

10MALDI

257.5

100

<5

57.5

522

.527

.520

<5

12.5

22.5

510

<5

20

[POPEþNa]

þFragmen

tIons

121

146

164

182

239�

265�

321�

415�

441�

433�

459�

458�

484�

577

599

617

697

Q-ToF

5<5

80<5

2015

<5

4010

030

10<5

2050

100

550

IonTrap

——

——

<5

<5

<5

4010

020

515

57.5

512

.5<5

100

QqQ

25<5

100

<5

1010

<5

3010

025

<5

—<5

2030

<5

10MALDI

2510

100

<5

12.5

10<5

37.5

27.5

1517

.522

.522

.55

10<5

20

� Relativeab

undan

ce(%

)in

amassrangebetween20

0an

d50

0m/z

values.

3240 C. Simoes et al.

systematic study of fragmentation under different exper-

imental conditions.

Fragmentation of [MþNa]þ precursor ionswas induced by

low-energy CID, using ESI-MS instruments with different

analyzers, Q-TOF, triple quadrupole and linear ion trap. ESI-

CID (low-energyMS/MS) of both PLPE and POPE [MþNa]þ

ions generates product ions through the same fragmentation

pathways (summarized in Table 1), in the three instruments.

As an example, Fig. 1 shows the PLPE [MþNa]þ product ion

spectra for the different ESI instruments. Both PLPE and

POPE product ion spectra show product ions that give

information about the phospholipid class. These ions are

produced by loss of the polar head (�141Da), loss of

the polar head bearing a sodium in the phosphatidyl moiety

(�163Da), loss of part of the polar head, PO3CH2CH2NH2

(�123Da), and loss of aziridine (�43Da). It is possible to

observe in the mass spectra other ions at m/z 164.0 (sodiated

polar head), 146.0 and 121.0, in accordance with the work

performed by Han and Gross17 in sodium adducts and Hsu

and Turk16 in lithium adducts. Scheme 1 shows structures of

the product ions produced by these common pathways. In

addition, it is possible to see in Scheme 2 the structure and

fragmentation pathways of product ions generated by the

loss of sn-1 and sn-2 acyl chains, as ketene (R––C––O) or as free

fatty acid (RCOOH), combined with loss of 43Da. This

fragmentation pathway is in agreement with the fragmenta-

tion of the [MþLi]þ ions reported by Hsu and Turk.16 These

product ions and the R1COþ and R2CO

þ ions give

information about the fatty acyl substituents of the PEs.

The product ion formed by loss of R1COOH and aziridine

(m/z 439.2 for PLPE or 441.2 for POPE) shows higher relative

abundance (usually with a 2:1 ratio) when compared with

the ion formed by loss of R2COOH and aziridine (m/z 415.2),

thus [MþNa-R1COOH-43]þ> [MþNa-R2COOH-43]þ. Also,

the product ion formed by loss of R1COOH shows higher

relative abundance than the product ion formed by loss

of R2COOH ([MþNa-R1COOH]þ (m/z 482.3 for PLPE and

484.3 for POPE)> [MþNa-R2COOH]þ (m/z 458.3)). These

two pairs of product ions allow identification of both fatty

acyl residues of the PEs, as well as their specific location in

the glycerol backbone. These diagnostic product ions can be

used for differentiation of isomeric PEs in the analysis of

sodium adducts, of low-energy ESI-MS/MS spectra, inde-

pendently of the instrument used.

Due to instrumental limitations, it is not possible to

acquireMS2mass spectra in themass range belowm/z 200Da

in the linear ion trap. Thus, with the exception of the product

ions at m/z 164.0, 146.0 and 121.0, the linear ion trap MS/MS

spectra show the same product ions identified for the Q-

TOF2 instrument (Fig. 1). Is possible to observe with high

relative abundance the product ions formed by loss of the

polar head, allowing identification of the PE phospholipid

class. Besides, both spectra differ in the ion corresponding to

the base peak of the spectra.

In this work, we studied the [MþNa]þ ion fragmentation

of two PE phospholipids, formed under high-energy CID

conditions, using a MALDI-TOF/TOF. The MALDI-MS/MS

spectra of POPE and PLPE [MþNa]þ ions show extensive

fragmentation. Themass spectra of both PEs (Fig. 2) show the

fragment ion at m/z 164.0, which is the base peak of the

Copyright # 2008 John Wiley & Sons, Ltd.

Rapid Commun. Mass Spectrom. 2008; 22: 3238–3244

DOI: 10.1002/rcm

Page 4: Determination of the fatty acyl profiles of phosphatidylethanolamines by tandem mass spectrometry of sodium adducts

Figure 1. ESI-MS/MS spectra of the [MþNa]þ ion of PLPE acquired using the Q-TOF2, linear ion

trap and QqQ mass spectrometers.

Scheme 1. Common product ions generated from PLPE and POPE [MþNa]þ fragmentation

of the polar head.

Copyright # 2008 John Wiley & Sons, Ltd. Rapid Commun. Mass Spectrom. 2008; 22: 3238–3244

DOI: 10.1002/rcm

Determination of fatty acyl profiles of PEs 3241

Page 5: Determination of the fatty acyl profiles of phosphatidylethanolamines by tandem mass spectrometry of sodium adducts

Scheme 2. Common product ions generated from PLPE and POPE [MþNa]þ fragmentation of sn-1

and sn-2 acyl chains.

3242 C. Simoes et al.

spectra. We can observe other diagnostic ions of the PE polar

head at m/z 121.0, 146.0, and ions from loss of 43Da, loss of

123Da, loss of 141Da and loss 163Da from the [MþNa]þ

precursor ion, in accordance with the previous published

work performed by Fuchs et al.24 and Stubiger et al.22

Similarly to ESI-MS/MS, the MALDI-MS/MS spectra show

product ions from loss of sn-1 and sn-2 as free fatty acids or as

ketene, combined or not with the loss of 43Da. In the mass

range of m/z 200 to 500, the base peak is the fragment ion at

m/z 204.0. This ion was not observed in the MS/MS spectra

obtained under low-energy CID. It corresponds to the

sodium adduct of the ion atm/z 182.1 detected in the ESI-MS/

MS spectra and was generated by loss of R1COOH

Figure 2. MALDI-MS/MS spectrum

Copyright # 2008 John Wiley & Sons, Ltd.

and R2COOH, as represented in Scheme 2. In contrast with

the ESI-MS/MS spectra, the product ions [MþNa-R1COOH-

43]þ and [MþNa-R2COOH-43]þ have similar relative

abundance, not allowing the position of the fatty acyl

residues to be distinguished. Surprisingly, we found another

pair of product ions at m/z 367.3/391.3 (for PLPE) and at

m/z 367.3/393.3 (for POPE), that were absent in the low-

energy CID spectra. These ions result from cleavage of the

bond between C3–C4 (g-cleavage) of the fatty acyl chains,

proximal to the ester linkage, combined with the loss

of 163Da [MþNa-(R2-C2H3)-163]þ/[MþNa-(R1-C2H3)-163]

þ

(Scheme 3). The terminal double bond conjugates with the

carbonyl group, produces a stable a,b-conjugated system,

of the [MþNa]þ ion of PLPE.

Rapid Commun. Mass Spectrom. 2008; 22: 3238–3244

DOI: 10.1002/rcm

Page 6: Determination of the fatty acyl profiles of phosphatidylethanolamines by tandem mass spectrometry of sodium adducts

Scheme 3. Product ions, generated from the g-cleavage (between C3–C4) of the fatty acyl chain, by

charge-remote fragmentation, only seen in MALDI-MS/MS high-energy collision fragmentation.

Determination of fatty acyl profiles of PEs 3243

which may explain the higher abundance of the product ions

formed by the g-cleavage.28 The g-cleavage of the fatty acyl

chains is a charge-remote fragmentation, andwas reported in

high-energy CID spectra of fatty acids and triacylglycer-

ols.28–30 Gross and co-workers established that charged-

remote fragmentations are valuable for structural identifi-

cation of different molecules of lipids.30,31 Charge-remote

fragmentations occur faraway from the charge location, and

are dependent on the precursor ion energy.32 The energy of

the precursor ion depends on the energy of the compound

analyzed, of the energy transferred during the ionization

process and of the activation by the collision with a gas.32

Charge-remote fragmentation usually occurs in sector

instruments, that allow high-energy CID fragmentations,

although sometimes could result from low-energy CID

fragmentation.30 Some studies report this fragmentation in

tandem mass spectra from MALDI-TOF,33 and in a MALDI-

TOF/TOF instrument.34 The 4800 MALDI-TOF/TOF instru-

ments accelerate the ions with energy of 2 keV, allowing

high-energy CID. The three ESI instruments used in this

study accelerate the ions with energies below 100 eV,

justifying the absence of these ions in the ESI-MS/MS

spectra. Thus precursor ions activated by both high- and low-

energy CID have different energies and may yield signifi-

cantly different product ion spectra.35

The product ion [MþNa-(R2-C2H3)-163]þ, generated from

g-cleavage of the sn-2 acyl chain plus loss of the polar head

(m/z 367.3 for both PEs), is twofold more abundant than the

product ion [MþNa-(R1-C2H3)-163]þ generated from g-

cleavage of the sn-1 fatty acyl residue plus loss of 163Da

(m/z 391.3 for PLPE and m/z 393.3 for POPE). Therefore, we

propose using, when analysis is performed under MALDI

high-energy collision conditions, the pair of ions [MþNa-(R2-

C2H3)-163]þ/[MþNa-(R1-C2H3)-163]

þ as diagnostic of the

specific location of the fatty acyl residues in the glycerol

backbone of the PE.

Copyright # 2008 John Wiley & Sons, Ltd.

CONCLUSIONS

The differences found in the relative abundances of ions

resulting from losses of R1 and R2 acid chains identify

location of the fatty acyl residues in the glycerol moiety. We

propose using ESI-MS/MS product ions [MþNa-R1COOH-

43]þ> [MþNa-R2COOH-43]þ as diagnostic ions to identify

the fatty acyl residue linked to sn-1 and sn-2 positions of PEs.

When using MALDI-MS/MS, the fatty acyl residues are

identified using the product ions resulting from g-cleavage

of the sn-1 fatty acyl chain plus loss of 163 Da and from the

g-cleavage of the sn-2 fatty acyl chain plus loss of 163Da. The

pair of ions [MþNa-(R1-C2H3)-163]þ< [MþNa-(R2-C2H3)-

163]þ is diagnostic of the specific location of the fatty acid

chains in the glycerol backbone of the PE, in high-energy

MALDI-MS/MS analysis.

AcknowledgementsThe authors thank FEDER, FCT-Portugal, Research Unit 62/

94 ‘Quımica Organica, Produtos Naturais e Agro-Alimen-

tares’, Aveiro, for financial support.

REFERENCES

1. Fruhwirth GO, Loidl A, Hermetter A. Biochim. Biophys. Acta2007; 1772: 718.

2. Vance JE. J. Lipid Res. 2008; 49: 1377.3. Hulbert AJ, Rana T, Couture P. Comp. Biochem. Physiol.

B. Biochem. Mol. Biol. 2002; 132: 515.4. Guderley H, Kraffe E, Bureau W, Bureau DP. J. Comp.

Physiol. [B] 2008; 178: 385.5. Petridou A, Nikolaidis MG, Matsakas A, Schulz T, Michna

H, Mougios V. Eur. J. Appl. Physiol. 2005; 94: 84.6. Hulbert AJ, Faulks S, Buttemer WA, Else PL. J Exp Biol 2002;

205: 3561.7. Ocloo A, Shabalina IG, Nedergaard J, Brand MD. Am. J.

Physiol. Regul. Integr. Comp. Physiol. 2007; 293: R1086.8. Tsalouhidou S, Argyrou C, Theofilidis G, Karaoglanidis D,

Orfanidou E, Nikolaidis MG, Petridou A, Mougios V.J. Anim. Sci. 2006; 84: 2818.

Rapid Commun. Mass Spectrom. 2008; 22: 3238–3244

DOI: 10.1002/rcm

Page 7: Determination of the fatty acyl profiles of phosphatidylethanolamines by tandem mass spectrometry of sodium adducts

3244 C. Simoes et al.

9. Hulbert AJ. J. Theor. Biol. 2005; 234: 277.10. NikolaidisMG, PetridouA,Mougios V. Physiol. Res. 2006; 55:

259.11. Yamaji-Hasegawa A, Tsujimoto M. Biol. Pharm. Bull. 2006;

29: 1547.12. Wenk MR. Nat. Rev. Drug Discov. 2005; 4: 594.13. Schiller J, Suss R, Fuchs B, Muller M, Zschornig O, Arnold K.

Front. Biosci. 2007; 12: 2568.14. Han X, Gross RW. J. Lipid Res. 2003; 44: 1071.15. Pulfer M, Murphy RC. Mass Spectrom. Rev. 2003; 22: 332.16. Hsu FF, Turk J. J. Mass Spectrom. 2000; 35: 595.17. Han X, Gross RW. J. Am. Soc. Mass Spectrom. 1995; 6: 1202.18. Hsu FF, Turk J. J. Am. Soc. Mass Spectrom. 2000; 11: 892.19. Schiller J, Suss R, Arnhold J, Fuchs B, Lessig J, Muller M,

PetkovicM, Spalteholz H, Zschornig O, Arnold K. Prog. LipidRes. 2004; 43: 449.

20. Schiller J, Suss R, Fuchs B, Muller M, Petkovic M, ZschornigO, Waschipky H. Eur. Biophys. J. 2007; 36: 517.

21. Stubiger G, Belgacem O. Anal. Chem. 2007; 79: 3206.22. Stubiger G, Pittenauer E, Allmaier G. Anal. Chem. 2008; 80:

1664.23. Harvey DJ. J. Mass Spectrom. 1995; 30: 1333.

Copyright # 2008 John Wiley & Sons, Ltd.

24. Fuchs B, Schober C, Richter G, Suss R, Schiller J. J. Biochem.Biophys. Methods 2007; 70: 689.

25. Al-Saad KA, Zabrouskov V, Siems WF, Knowles NR, Han-nan RM, Hill HH Jr. Rapid Commun. Mass Spectrom. 2003; 17:87.

26. Estrada R, Yappert MC. J. Mass Spectrom. 2004; 39: 412.27. Jackson SN, Wang HY, Woods AS. J. Am. Soc. Mass Spectrom.

2007; 18: 17.28. Cheng C, Gross ML, Pittenauer E. Anal. Chem. 1998; 70: 4417.29. Harvey DJ. J. Am. Soc. Mass Spectrom. 2005; 16: 280.30. Cheng C, Gross ML. Mass Spectrom. Rev. 2000; 19: 398.31. Griffiths WJ. Mass Spectrom. Rev. 2003; 22: 81.32. Cheng C, Pittenauer E, Gross ML. J. Am. Soc. Mass Spectrom.

1998; 9: 840.33. Domingues MRM, Marques MGS, Vale CA, Neves MG,

Cavaleiro JA, Ferrer-Correia AJ, Nemirovskiy OV, GrossML. J. Am. Soc. Mass Spectrom. 1999; 10: 217.

34. Trimpin S, Clemmer DE, McEwen CN. J. Am. Soc. MassSpectrom. 2007; 18: 1967.

35. Domingues MRM, Nemirovskiy OV, Marques MGOS,Neves MG, Cavaleiro JAS, Ferrer-Correia AJ, Gross ML.J. Am. Soc. Mass Spectrom. 1998; 9: 767.

Rapid Commun. Mass Spectrom. 2008; 22: 3238–3244

DOI: 10.1002/rcm