Mass-Spectrometric Analysis of Lipids (Lipidomics) 1. Identification 2. Quantification 3. Metabolism.

Mass-Spectrometric Analysis of Lipids (Lipidomics)

1. Identification

2. Quantification

3. Metabolism

Why to do lipidomics?

Biology: Functions of different lipids?

Medicine: Diagnostics and Therapy

Industry: Healthier food, Quality control

GlyceroPhospholipids

>10 classes (PC, PE, PS, PI, PA etc)

- Each class consists of numerous species

with different fatty acid combinations

(>20 different fatty acids)

=> Thousands of different molecular

species possible

O

O

O

O

HCH2C CH2

O

PO

O

ON

H

H

H

H

HH

H

H

H

HH

H

H

Phosphatidylcholine (PC)

Neutral Glycerolipids

- Triacylglycerols (TG)- Diacylglycerols (DG)- Monoacylglycerols (MG)

-Each class consists of numerous species due to different fatty acid combinations

=> Hundreads of different molecular species

TG DG

O

O

O

O

HCH2C CH2

O

O

O

O

O

O

HCH2C CH2

OH

Lactosylceramide Ganglioside Sulfatide

O

O

HNHC

CH2

O

C O

CH

O

H

H

OH

H

O

H

H

OH

O

OH

H

H

OH

H

O

HO HH

OH

O

O

HNHC

CH2

O

C OH

CH

O

H

H

OH

H

O

H

H

OH

O

O

H

H

OH

H

O

HO HH

HO

O

COOH

O

HO

OH

AcHN

OH

HO

O

O

HNHC

CH2

O

C OH

CH

O

H

H

O

H

O

H

H

OH

H3SO3-

Sphingolipids

- Ceramides- Neutral Glycosphingolipids- Acidic Glycosphingolipids

-Each class consists of numerous species due to different fatty acid

=> Hundreads of different molecular species

The complete lipidome of no cell or tissue has ever been determined

...because of technical limitations

Advantages of MS analysis

Sensitivity >1000-fold higher than with conventional methods

Resolution - Allows quantification of hundreds of lipid species

Speed -100 times faster

Can be automated - High troughput possible

Ionization methods used in lipid MS

Electrospray (ESI)– Does not cause fragmentation – Can be easily automated – Compatible with on-line LC

Matrix-assisted laser desorption (MALDI)– Less used thus far – Suppression by PC/SM > All lipids not detected– On-line LC separation not feasible

Electrospray ionization

Competition for charge => Suppression effects!

MS spectrum of cellular lipid extract = a Mess!

700 725 750 775 800 825 850 875 m/z0

2000

4000

6000

Intens.PE 18:0/18:2

PE 16:0/18:1

PA 18:0/18:2

PE 18:0/20:4

LBPA 18:1/18:1

PE 18:0/22:5

PI !8:0/18:2

PI 18:0/20:3

PS 18:0/18:1PE 18:0/20:4

Scanning

MS1 MS2 Collision cell Detector

How to improve selectivity?

A. Lipid class -specific scanning (MS/MS)

B. On-line chromatographic separation (LC-MS)

Lipid class -specific scanning

Phospholipid class consist of species with the same polar head-group but different fatty acids

Phospholipid class Specific scan

Phosphatidylcholines Precursors of +184

Phosphatidylinositols Precursors of -241

Phosphatidylethanolamines Neutral-loss of 141

Phosphatidylserines Neutral-loss of 87

O

O O

O

CH

CH2

H2C O P

O

O

O X

Precursor ion scanning

Requires a characteristic, charged product ion

PC => Diglyceride + phosphocholine (+184)

FragmentationScanning Static (+184)

MS1 MS2 Collision cell (Helium or Argon)

Precursors of +184 => PC + SM

-Alkaline hydrolysis can be used to remove PCs

SM-16:0

Neutral-loss scanning

Scanning Fragmentation Scanning

MS1 MS2 Collision cell(Helium or Argon)

Mass interval = 141

..when the characteristic fragment is uncharged

PE => Diglyceride (+) + phosphoethanolamine (141)

Neutral-loss of 141 (= PE)

700 725 750 775 800 825 850 875 m/z0

2000

4000

6000

Intens.PE 18:0/18:2

PE 16:0/18:1

PA 18:0/18:2

PE 18:0/20:4

LBPA 18:1/18:1

PE 18:0/22:5

PI !8:0/18:2

PI 18:0/20:3

PS 18:0/18:1PE 18:0/20:4

MS-scan

700 750 800

MS analysis of Sphingolipids

Sphingosine Ceramide Lactosylceramide Ganglioside Sulfatide

O

O

HNHC

CH2

O

C O

CH

O

H

H

OH

H

O

H

H

OH

O

OH

H

H

OH

H

O

HO HH

OH

O

O

HNHC

CH2

O

C OH

CH

O

H

H

OH

H

O

H

H

OH

O

O

H

H

OH

H

O

HO HH

HO

O

COOH

O

HO

OH

AcHN

OH

HO

O

O

HNHC

CH2

O

C OH

CH

O

H

H

O

H

O

H

H

OH

H3SO3-

OOH

O

COOH

O

HO

OH

AcHN

OH

HO

H3SO3-

OO

HNHC

CH2

C OH

CH

OH

HHNHC

CH2

C OH

CH

OH

Ceramide and Neutral Glycosphingolipids

- Precursors of sphingosine (m/z +264)

Ceramides

Glucosylceramides

24:1

Sulfatides

- Precursors of Sulfate (m/z -97)

Liquid chromatography-MS (LC-MS)

Advantages - Increased sensitivity due to diminished suppression of minor species by

- Major species - Impurities

Disadvantages– Takes more time (not UPLC)– Data analysis more complex (?)

LC-MS analysis of mouse brain lipids

TimeHermansson et al. (2005) Anal Chem.77:2166-75

Data analysis => software

A. Processing of the data => Identification

=> Concentrations

B. Bioinformatics => Biomarkers?

=>Biological significance?

Quantification not simple

Signal intensity depends on:

• Lipid head-group• Acyl chain length• Acyl chain unsaturation• Ions present (adduct formation)• Detergent and other impurities (suppression)• Solvent composition and instrument settings

=> Internal standards necessary!

LIMSA

Excel add-on for Quantitative Analysis of MS data (Haimi et al. .2006. Anal Chem. 78:8324-31)

LIMSA does:

Peak picking and fitting

Peak overlap correction

Peak assignment (database of >3000 lipids)

Quantification with internal standards

Batch analysis

MS-imaging of Lipids by MALDI

PI 38:4 Sulfatide 24:1 Hydroxy-Sulfatide 24:1

Analysis of Lipid Metabolism by MS

Adds another, dynamic dimension to lipidomics

Labeled lipids can be selectively detected!

D9-PC => +193 (Unlabeled PC => +184)

D4-PE => 145 (Unlabeled PE => 141)

D4-PS => 90 (Unlabeled PS => 87)

D6-PI => -247 (Unlabeled PI => -241)

Precursors

Water soluble precursors (D9-choline etc) Exogenous lipids

Phospholipid Remodeling: Exchange of acyl chains

Glycerol

Fa

tty a

cid

Fa

tty a

cid

Alc

oh

olP

O4

Fa

tty a

cid

Glycerol

Fa

tty a

cid

Fa

tty a

cid

Alc

oh

olP

O4

Fa

tty a

cid

PLA2

GlycerolF

atty

aci

d

Alc

oh

olP

O4

OH

Acyl transferase

Analysis of phospholipid remodeling using soluble precursor is problematic

100

100

100

100

680 700 720 740 760 780 800

100

16:0/17:1

24h

unlabeled

0h

1h

5h

16:0/18:1

18:1/18:1

18:0/18:1

16:0/16:1

Rel

ativ

e in

tens

ity (

%)

m/z

0 5 10 15 20 25

0

25

50

Rel

ativ

e ab

unda

nce

(%)

Chase time (h)

16:0/16:1 16:0/17:1 16:0/18:1 16:1/18:1 18:0/18:1 18:1/18:1 18:1/18:2 18:0/20:4 18:1/20:4 18:0/22:6

D4-ethanolamine => cells => D4-PE species

Kinetics

Our approach: Use intact exogenous phospholipids with a deuterium-labeled head-group

PROTOCOL

• Synthesize a phospholipid with a deuterium-labeled head group

• Make vesicles containing the labeled phospholipid • Incubate cells with these vesicles and β-cyclodextrin

(carrier)• Extract and analyze lipids using MS/MS scans showing

the labeled (or unlabeled) lipid only• Determine the pathways and kinetics of remodeling

100%

100%

100%

100

700 750 800 850

100%

14:0/14:0

0h

18:0/20:4

18:0/18:114:0/20:4

14:0/18:1

16:0/14:0

3h

18:1/20:4

18:1/18:1

16:0/18:1

7h

Rel

ati

ve

Inte

ns

ity

24h

Endogenous

m/z

Unnatural 14:0/14:0-PE is remodeled very rapidly

14:0/14:0-D4-PE

0 5 10 15 20 25

0

25

50

14:1/14:1 16:0/14:1 18:0/14:1 18:1/14:1 16:0/18:1 14:1/20:4 18:0/18:1 18:1/18:1 18:0/20:4 18:1/20:4

KINETICS

”Natural” 18:1/18:1-PE is hardly remodeled

100%

100%

100%

100%

100%

700 750 800

100%

0 h

18:1/18:1

1 h

3 h

5 h

Re

lativ

e a

bu

nd

an

ce

24 h

endogenous

Positional isomers are remodeled with very different kinetics

14:0/18:1 18:1/14:0

700 750 800

100%

700 750 800

100%

700 750 800

100%

700 750 800

100%

700 750 800

100%

700 750 800

100%

700 750 800

100%

700 750 800

100%

0h18:1/18:1

14:0/18:1

18:1/18:1 3h18:1/18:2

16:0/18:116:1/18:17h

endogenous

Rel

ativ

e ab

unda

nce

m/z

16:1/16:10h

18:1/16:1

18:1/18:218:1/22:6

18:1/20:4

18:1/18:1

18:1/14:0

3h

7h

endogenous

m/z

Pathways of 14:0/14:0-PE remodeling

14:0/14:0

14:0/ 18:2 14:0/ 16:1 14:0/ 18:1 14:0/ 20:4 14:0/ 22:6

16:1 /18:2

16:0 /18:2 18:1 /18:2 18:1 /16:1 16:0 /18:1 16:1 /18:1 18:1 /20:4 18:1 /22:6

18:1 /18:1 18:0 /20:4

Kainu et al. (2008) J Biol Chem. 283:3676-87

Studies with >50 phospholipid species (and PLA inhibitors) indicate that

=> Multiple acyl chain specific PLAs are involved

in remodeling of phospholipids in mammalian cells

BUT which PLAs? ..and what determines their specificity?

..and which acyltransferases are involved?

Conclusions

MS-based lipidomics is highly usefull in – Biology– Medicine– Food industry

....but needs to be integrated with other “omics” and functional assays

Heavy isotope –labeling adds an important extra dimension to lipidomics

Contributors

• Martin Hermansson• Ville Kainu• Perttu Haimi