47

Clinica Chimica Acta, 71 (1976) 47-54 @ Elsevier Scientific Publishing Company,

Amsterdam

- Printed in The Netherlands

CCA 7908

AUTOMATIC POTENTIOMETRIC TITRATION AND GAS-LIQUID CHROMATOGRAPHY OF UNDERIVATISED FREE FATTY ACIDS

A. WIRTH *, J. ECKHARD Department (G.F.R.) (Received

and H. WEICKER

of Metabolic and Athletic Research,

University of Heidelberg,

Hospitalstmsse 3

March, 2nd. 1976)

Summary A rapid and precise method is described for the determination of total free fatty acids (FFA) in the serum followed by the separation of individual FFA by gas-liquid chromatography (GLC). After the extraction the FFA are directly injected into the gas-liquid chromatograph without esterification. DEGS-PS serves as stationary phase. The esterified fatty acids such as cholesterol esters, phospholipids or triglycerides do not interfere with the FFA determined by the GLC method. This method is compared to other commonly used titrimetric and gas-chromatographic procedures. The changes of the composition of individual FFA during and after ergometer work is reported.

Introduction Though free fatty acids (FFA) as a supply for the energy metabolism are very important, simple and exact methods for their determination are still missing. A reliable method which is often used for the extraction of FFA from the serum is the one developed by Dole [l] which was originally used for titrimetric techniques [ 1,2,3]. Since then it has also been applied in calorimetric [4], radiochemical [ 51, and gas-chromatographic analysis [ 6,7]. Several investigators found a recovery of more than 93% [ 1,5,8]. Detailed findings over interfering substances such as phospholipids, lactate, aceto-acetic and beta-hydroxybutyric acids have also been reported [ 1,9,10]. Titration in a two-phase system has been performed in previous techniques and the result was that the changing point of the indicator could not be read off exactly [1,3,9]. Some investigators dissolved FFA and the base in solvents with different dissociation [ 311,121. This method contains a systematic error which will be shown later. * TO whom

alle correspondence should be addressed.

Usually the FFA are isolated by means of column- or thin-layer chromatography and then derivatised into methyl esters [13,14] . Esterification of FFA in the presence of other lipids is possible with diazomethane [6] since no transesterification can take place, whereas transesterification may occur using carbonyldiimidazol [7]. Recently a simple and quick determination of FFA without isolation and esterification was reported [ 151. Since the recovery was between 75--80% it is recommended as a screening method. The combination of titration and gas-liquid chromatography increases the accuracy of the FFA evaluation considerably. Materials and methods Reagents Solvent system for extraction: isopropan-2-01/n-heptane/sulfuric acid (40 : 10 : 1, v/v; E. Merck, Darmstadt). Titrant: 0.002 N tetrabutylammoniumhydroxide (TBAH) (out of 0.1 N TBAH by E. Merck) in isopropan-2-al/methanol (1 : 1, v/v). Titration standards: palmitic, lauric, and linolenic acid in concentrations of 300,500,1000 and 2000 pequiv./l, dissolved in isopropan-2-ol/methan01 (1 : 1, v/v). Standards for GLC: lauric, myristic, pentadecanoic, palmitic, palmitoleic, stearic, oleic, linoleic, and linolenic acid as a mixture in a concentration of 10 mg/ml. Internal standard: pentadecanoic acid in a concentration of 6.666 g/l. Purity grade of fatty acids: at least 99% (Serva, Heidelberg). Extraction of FFA For extraction, 5 ml of the solvent system is added to 1 ml serum. The mixture is then shaken vigorously for at least 5 min. After standing for 10 min 2 ml HZ0 and 3 ml n-heptane are added and this mixture is shaken again for 3 min. After standing for 10 min the sample separates into 3 layers. The upper layer containing the FFA is transferred into a centrifuge tube. After the addition of 5 ml H,S04 (0.05%) the sample is centrifuged for 5 min at 3000 rev./min. Thereupon the upper layer is transferred into a tube fitting in to the rotation evaporator and the n-heptane is removed by vacuum evaporation. By using a distillation spider 6 aliquots can simultaneously be dried. Titration Titrations were carried out automatically with a combititrator (Metrohm AG, Switzerland) consisting of a pH-meter E 512, an impulsomate E 473, and a microdosimate E 412-1. The dry residue is dissolved in 2 ml isopropanolln-heptane (1 : 1, v/v) and then the tube is attached to the titrator. The glass electrode, on which the two glass capillaries for the titrant and nitrogen are attached, is then inserted. It is important that the distance between the tip of the glass electrode and the outlet of the titrant be 3 mm. During titration the mixture is stirred by nitrogen. Titrant is added automatically until the endpoint has been reached. The pH of the endpoint had been previously ascertained by monitoring titration curves automatically. In the described procedure the endpoint is achieved at a pH of 9.8, The titration of 15 aliquots requires about 1 h. The titrations of blanks, standards and samples were performed in the same way.

49

Gas-liquid chromatography The quantitative evaluation of FFA requires the addition of an internal standard (pentadecanoic acid dissolved in n-heptane). When gas-chromatographic determination of FFA is carried out without a preceding titration, the dry residue is transferred to a small tube by chloroform. Then chloroform is removed up to a volume of 0.1 ml and an aliquot of 1 ~1 is injected into the column of the gas chromatograph. When gas chromatography and titration are applied together, then following titration 1 ml 0.002 N HzS04 and 2 ml n-heplane are pipetted into the tube. The mixture is shaken vigorously and centrifuged at 3000 rev./min. The upper layer of the sample is transferred to a glassstoppered tube and the chloroform is removed. The residue is treated as mentioned above. The gas-chromatographic conditions are presented in the following table : Apparatus: Carlo-Erba. Fractovap Model GI Detector: flame-ionisation detector oven: 2oo”c Injector: 25ooc Detector: 25O’C Carrier-gas: nitrogen, 30 ml/min Burning-gas: hydrogen. 30 ml/min air, 300 ml/min Column: glass column, 6 ft X 3 mm I.D.: 10% DEGS-PS on 80-100

mesh Supelcopoti -

-

Results Po ten tiome tric endpoint titration A typical titration curve of FFA is shown in Fig. 1. When weak acids e.g. fatty acids are titrated with a strong base (e.g. TBAH), 2 turning points are visible. The first one corresponds to the half-neutralisation point (HNP) and the second one to the equivalent point (E,P). The EP was ascertained by drawing the middle tangent [16] . The EP of pH 9.8 is entered into the combititrator and then titrations can be carried out automatically. The amount of titrant decreases considerably, close to the EP, so that the electrode can adjust itself to the ion concentrations and the exact endpoint can be achieved. Since FFA represent a mixture of different FA with distinctive pK values causing different EPs in the same medium, some concentrations of lauric, palmitic and linolenic acid were titrated and their EPs were determined. The difference of pH 0.15 between the EPs of lauric and linolenic acid implies an error of about l%, which demonstrates that FFA are titratable (Table I). Palmitic acid was repeatedly titrated in various concentrations (300, 500, 1000 and 2000 (pequiv./l) and the results can be seen in Fig. 2. The differences between the theoretical and obtained values were less than 2%. The reproducibility of the method was checked by repeated endpoint titrations. The variation coefficient amount to 0.7% which is mostly due to the error in pipetting. Gas-chromatographic determination A gas chromatogram obtained by the‘method described can be seen in Fig. 3. The individual FFA are completely separated. The separation factor for

\J2

I

200 v VfVdumo)

,

I

I

,

600

400

800

tow

Tllranl

/Ji

Fig. 1. Titrigram of FFA with equivalent point (EP). half-neutralisation point (HNP). PK value. Titrant: 0.002 N tetrabutylammoniumhydroxide in isopropanol/methanol.

and approximative

JJMOI/I

I

I

I

I

200

I

400

I

I

600

I

I

’

800

loab

1’

TITRANT Fig. 2. Titration (n = 5).

TABLE

of palmitic

acid of various concentrations

with 0.002 N tetrabutylammoniumhydroxide

I

Equivalent points (EPs) of different fatty acids (1 mequiv./l). dissolved in isopropanolln-heptane trated with 0.002 N tetrabutylammoniumhydroxide (TBAH) in isopropanol/n-heptane.

PH of equivalent point (EP)

Laurie acid

Palmitic acid

Linolenic acid

FFA

9.7

9.8

9.85

9.8

and ti-

51

n 'l&l

I

1

I

I

0

5

10

I5

1

I

25

20

MINUTES Fig. 3. Gas-liquid cbromatogram of underivatised FFA. Operating conditions: glass column 6 it X 3 mm I.D., 10% DEGS-PS on SO-100 mesh Supelcoport, column temperature 2OO’C.

1coo 1400-

1200< 5 1000g

800-

E 2 GOOc" 400-

0

200

I

400

GAS-LIQUID

6th

I

800

I

I

I

I

loo0 1200 1400 icm

CHROMATOGRAPHY

@q/l

Fig. 4. Comparison of total FFA determined by titration (endpoint titration with tetrabutylammoniumhydroxide) and gas chromatography (underivatised FFA by use of 10% DEGS-PSI (n = 30).

52

TABLE

II

Serum

individual

work

(periods

Fatty

free

for

fatty

12 min).

acid

acids

(FFA)

n = 8; n.s..

of healthy not

subjects

at rest and

Rest

End

(%)

(%)

of

exercise

C(16 C(18

: 0)

10.8

+ 0.9

Cc18

: 1) : 2)

33.8

f 0.8

33.3

-r 0.7

n.s.

16.9

f 0.9

16.2

I? 0.5

n.s.

C(16

C(18

of exhaustive

ergometer

Significance

: 0) : 0) : 0) : 1)

C(12 C(14

at the end

significant

1.2

f 0.2

1.2

+ 0.2

ms.

3.2

+ 0.2

3.1

t 0.2

n.s.

28.5

? 0.7

31.1

+ 0.8