Determination of non polar metabolites in human plasma using innovative sample preparation strategies coupled with gas chromatography and mass spectrometry

Maria Margarida Gonçalves1, Sandra Sagredo1, Maria Serrano Bernardo1, Benilde Mendes1, Maria Cristina Marques2, Daniel Etlin3, Ana Bragança4 and Valentina Vassilenko4

1Environmental Biotechnology Unit, Faculty of Sciences and Technology, New University of Lisbon, Portugal; 2Faculty of Pharmacy, University of Lisbon, Portugal; 3Unicam Sistemas Analíticos, SA, Portugal; 4Centre of Physics and Technological Research, New University of Lisbon, Portugal

Overview:

The main objective of this work is to develop simple,fast and selective extraction methods to isolate nonpolar metabolites or contaminants from humanplasma.

Solid phase extraction using a novel adsorbant(MonoTrapTM) was tested for direct extration fromdiluted plasma samples or from the correspondingheadspace.

Both direct extraction and headspace extraction withMonoTrapTM revealed the presence of terphenyls andpartially hydrogenated terphenyls in 39 human plasmasamples.

Dispersive liquid-liquid microextraction (DLLME) ofplasma samples was tested using CCl4 or usingheptane. Conventional DLLME procedure was adaptedto allow the use of non halogenated solvents.

A new method of simultaneous derivatisation,extraction and concentration of plasma fatty acids isproposed. This method couples directtransesterification of plasma samples with methanolicKOH and homogeneous liquid-liquid microextraction(HLLME) with halogenated or non halogenatedsolvents. All steps take place in a single vial with anextraction time of 15 min or less.

Introduction:

Metabolomics addresses the screening of compoundswith medium to low molecular weight, with biologicalrelevance as disease or exposure biomarkers [1].

Sample preparation is a key step of the process,because biological fluids are complex samples oftenavailable at limited amounts (1-2 ml in the case ofblood plasma).

The metabolic profile of plasma samples includes freeaminoacids, organic acids, amines, sugar, steroids,nucleic acid bases and other substances [2].

The analysis of these metabolites by GC-MS requiresthe use of appropriated derivatization reactions.

Non polar metabolites like aliphatic hydrocarbons havebeen determined in biological samples (breathsamples) and related to lipid peroxydation processesoccurring in cells of cancer patients or of smokers [3,4].

References

[1] Harin Kanani, Panagiotis K. Chrysanthopoulos, Maria I. Klapa “Standardizing GC–MSmetabolomics” Journal of Chromatography B, 871 (2008) 191–201

[2] Kishore K. Pasikanti, P.C. Ho, E.C.Y. Chan “Gas chromatography/mass spectrometry inmetabolic profiling of biological fluids” Journal of Chromatography B, 871 (2008) 202–211

[3] Leiliane Coelho A. Amorim, Zenilda de L. Cardeal “Breath air analysis and its use as abiomarker in biological monitoring of occupational and environmental exposure to chemicalagents” Journal of Chromatography B, 853 (2007) 1–9

[4] Michael Phillips, Renee N. Cataneo, Andrew R.C. Cummin, Anthony J., Gagliardi, KevinGleeson, Joel Greenberg, Roger A. Maxfield, William N. Rom “Detection of Lung Cancer WithVolatile Markers in the Breath” Chest 123 (2003) 2115-2123

Methods :

Blood samples from 40 volunteers (20 smokers and 20 non smokers) were collected and the corresponding plasmawas isolated and kept at -5ºC until analysis.

Plasma deproteinization was performed adding methanol and separating the precipitates by centrifugation.

Solid phase extraction with MonoTrapTM DSC18 disks (GL Sciences, Japan)

Plasma samples (500L) were diluted with distilled water and extracted with MonoTrapTM DSC18 disks using:

The SPE disks were extracted with dichloromethane, in an ultrasonic bath.

The extract was dried, concentrated to 200 µL and analysed by GC-MS.

.

a) Direct extraction (DE)

MonoTrapTM disks are placed in direct contact with the plasma sample, at 60 ºC, for 1h, under stirring.

b) Headspace extraction (HE)

MonoTrapTM disks suspended in theheadspace over the plasma sample, at60 ºC, for 1h, under stirring.

Homogeneous liquid-liquid microextraction coupled with direct transesterification(HLLME-DT): Plasma samples (500L) were diluted with methanolic KOH; a solutionof organic solvent CCl4 or heptane (50 - 150 µL) in acetone was added to thereaction mixture and an homogeneous solution was obtained. Phase separationwas induced by adding 2 ml of water. After centrifugation the organic phase wasrecovered and injected in a GC-MS-MS (Focus GC, Polaris Q, Thermounicam).

Results : Detection of terphenyls and hydrogenated terphenyls in human plasma SPE with MonoTrap DSC18 disks (DE and HE) + GC-MS

Partially hydrogenated terphenyls are producedin large amounts for industrial use – they areincluded in the HPV (high production volume)chemicals list by US-EPA. They are mainly used asheat transfer fluids and polymer plasticizers.

Extracted Chromatograms with base peak = 230 m/z

DE(Direct Extraction)

HE(Headspace Extraction)

Terphenyls

Diphenylcyclohexanes

Cyclohexylbiphenyls anddiphenylcyclohexanes

RT: 51.79 - 55.33

52.0 52.5 53.0 53.5 54.0 54.5 55.0

Time (min)

0

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100

Re

lative

Ab

un

da

nce

RT: 52.89

AA: 308491

RT: 53.31

AA: 95245RT: 54.28

AA: 80384

NL:

8.15E4

Base Peak

m/z=

241.50-

242.50 F:

MS ICIS

MQ8366NF1

RT: 55.17 - 58.72

55.5 56.0 56.5 57.0 57.5 58.0 58.5

Time (min)

0

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100

Re

lative

Ab

un

da

nce

56.39

56.89

57.87

57.0156.47

57.09 58.0057.7356.26 57.4455.88 58.5355.69 58.3355.29

NL:

2.84E4

Base Peak

m/z=

241.50-

242.50 F:

MS

Soro1-

MonotrapC18-

1

Extracted Chromatograms with base peak = 242 m/z

These compounds have been found in recycledpaper and cardboard, as well as in food productspacked with these recycled materials. Our resultsshow the presence of these contaminants inhuman plasma samples of 39 volunteers, bothsmokers and non smokers.

RT: 51.73 - 61.14

52 53 54 55 56 57 58 59 60 61

Time (min)

0

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100

Re

lative

Ab

un

da

nce

59.31

60.58

57.72

53.8551.95 58.3356.41 57.54 59.5553.95 58.4352.11 53.10 56.3355.11

NL:

1.06E5

Base Peak

m/z=

235.50-

236.50 F:

MS

Soro1-

MonotrapC18-

1

RT: 53.34 - 57.65

53.5 54.0 54.5 55.0 55.5 56.0 56.5 57.0 57.5

Time (min)

0

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100

Re

lative

Ab

un

da

nce

RT: 55.53

AA: 5395533

RT: 56.75

AA: 3780512

RT: 54.02

AA: 1045938

RT: 54.54

AA: 161266

RT: 54.26

AA: 72594

NL:

1.39E6

Base Peak

m/z=

235.50-

236.50 F:

MS ICIS

MQ8366NF1

Extracted Chromatograms with base peak = 230 m/z

Dicyclohexylbenzenes

RT: 52.59 - 62.52

53 54 55 56 57 58 59 60 61 62

Time (min)

0

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100

105

Re

lative

Ab

un

da

nce

59.31

60.48

56.53

60.5856.39

53.04

57.7256.89

53.84

57.88

57.00 58.3355.25 60.7460.18 62.0854.58 56.03

NL:

6.01E5

TIC F: MS

Soro1-

MonotrapC

18-1

DE(Direct Extraction)

RT: 52.59 - 61.56

53 54 55 56 57 58 59 60 61

Time (min)

0

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100

Re

lative

Ab

un

da

nce

RT: 59.31

AA: 2569240

RT: 60.48

AA: 3042690

RT: 56.53

AA: 1709616

RT: 56.39

AA: 1073826RT: 53.04

AA: 959376

RT: 57.72

AA: 438797

RT: 56.89

MA: 408434

RT: 53.84

AA: 460400 RT: 57.88

AA: 258017

RT: 58.33

AA: 41344RT: 55.25

AA: 33574

60.58

57.0060.7460.1854.5853.20 56.03

NL:

6.01E5

TIC F: MS

Soro1-

MonotrapC

18-1

HE(Headspace Extraction)

Results : Extraction of terphenyls and hydrogenated terphenyls using HLLME

Peaknº

NameBase peak area (m/z= 230)

MT-DE MT-HE HLLME

1 o-terphenyl 2429532 241665 1200775

2 m-terphenyl 333761 - 115186

Peaknº

NameBase peak area (m/z= 236)

DE HE HLLME

3 1,2-diphenylcyclohexane 164368 17267 79147

4 1,3-diphenylcyclohexane 1045938 45041 353403

5 1,4-diphenylcyclohexane 72594 5981 61418

6 3-cyclohexylbiphenyl 5395533 457725 2872927

7 4-cyclohexylbiphenyl 3780512 237143 1819588

Peaknº

Name Base peak area (m/z= 242)

DE HE HLLME

8 1,3-dicyclohexylbenzene 308491 117248 231020

9 phenyldicyclohexane 95245 35084 62574

10 1,4-dicyclohexylbenzene 80384 26736 62484

0

1000000

2000000

3000000

4000000

5000000

6000000

1 2 3 4 5 6 7 8 9 10

Pe

ak a

rea

Peak nº

Monotrap - Direct Extraction

Monotrap - Headspace

HLLME

RT: 51.01 - 62.99

52 53 54 55 56 57 58 59 60 61 62

Time (min)

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100

Re

lative

Ab

un

da

nce

59.07

53.76

59.23

60.01

60.38

52.86

57.5253.04

56.20

57.1652.01 57.69 61.5153.99 61.8955.7555.07

NL:

6.52E6

TIC F: MS

dllme_100_

150CCL4_

MQ8365NF

_18_06_02

Base peak 230

Base peak242

Base peak 236

TIC profile from HLLME of a nonsmoker plasma

Results : direct transesterification + HLLME CCl4 or heptane

0

5

10

15

20

25

30

35

40

45

C12:0 C14:0 C16:0 C16:1 C18:0 C18:1 C18:2 C18:3 C20:3 C20:4 C20:5 C22:5 C22:6

Re

lati

ve C

on

cen

trat

ion

(%

pe

ak a

rea)

Fatty Acid

Smoker, CCl4

Smoker, Heptane

Non smoker, CCl4

Non smoker, Heptane

RT: 35.08 - 90.01

40 45 50 55 60 65 70 75 80 85 90

Time (min)

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100

Re

lative

Ab

un

da

nce

59.36

54.04

59.50

64.06 85.15

60.26

64.5569.29

60.59

53.03

69.66 88.6079.5057.7247.02 71.43 79.9665.0849.66 76.3773.8638.15 41.96

NL:

5.42E7

TIC F: MS

dllme_100_

150CCL4_

MQC535F_

23_06_01

Tic profile, FFA methyl estersfrom smoker

plasma

26 Fatty Acid Methyl Esters Identified

(C12:0) Lauric acidC14:0) Myristic acid(C14:0) 12-Methyltetradecanoic acid(C16:1) Palmitoleic acid(C16:0) Palmitic acid(C16:0) 15-methylhexadecanoic acid(C16:0) 14-methylhexadecanoic acid(C17:0), Heptadecanoic acid

( (C20:2)Eicosadienoic acid(C20:1) Eicosenoic acid(C20:0) Eicosanoic acid(C22:6) Docosahexaenoic (2 isomers)(C22:5) Docosapentaenoic(C22:4) Docosatetraenoic

• Fatty acid profile can be obtained with direct transesterification and HLLMEeither with heptane or with CCl4.

• Comparison of 13 reference fatty acid showed no significant differencesbetween smokers and non smokers.

• Although heptane extracts are less concentrated than CCl4 extracts wecould identify 26 fatty acid methyl esters.

(C18:3), Linolenic acid(C18:2), Linoleic acid(C18:1), Oleic acid(C18:1) Elaidic acid(C18:0) Stearic acid(C20:5) Eicosapentaenoic acidC20:4) Eicosatetraenoic acid (3 isomers)(C20:3) Eicosatrienoic acid (2 isomers)

Conclusions :

• Terphenyls and partially hydrogenated terphenyls were extracted from human plasma.

• Extracts of the materials in contact with the plasma samples and blanks of the analytical process revealed no contamination with these compounds.

• SPE of deproteinized and diluted plasma with a new monolithic adsorbant (MonoTrapTM), both in direct contact orfrom the headspace, allows the selective extraction of these plasma contaminants without significant co-extraction ofplasma metabolites. The method with direct contact provides better extraction yield than the headspace extraction.

• To the best of our knowledge this is the first report on the use of non halogenated solvents for any kind of liquid-liquid microextraction. Although showing smaller enrichment factors then halogenated solvents, the non halogenatedsolvents are generally less toxic and therefore can be a environmentally safer alternative.

• When performing DLLME with plasma samples diluted in water the addition of the organic phase (heptane or CCl4), causes the formation of a gel like structure that prevents a good phase separation.

• A new method for the direct derivatization (transesterification) of plasma lipids and simultaneous extraction andconcentration of the corresponding methyl esters was developed. The method uses organic solvents in the microliterrange, is performed in a single vial and has a total derivatization + extraction time of 15 minutes or less.

• The homogeneous liquid-liquid microextraction coupled with direct transesterification allows the fast and sensitivedetermination of up to 40 fatty acid methyl esters derivatives of plasma lipids.

• Terphenyls and partially hydrogenated terphenyls were found, at variable concentrations in 39 out of 40 humanplasma samples analysed.

• No significant differences were found between the smokers and the non smokers group in what concerns fatty acidprofile or aromatic contaminants level.

Aknowledgements: The authors thank Dr. Etsudo Ikeda (GL Sciences, Japan) for the kind offer of MonoTrapTM samples. Thanks are also due to Dr. Carlos Marques (Nova Era Clinical Analysis Laboratory, Lisbon) for supplying the plasma samples.