184,86-89

ANALYTICALBIOCHEMISTRY

(1990)

High-Performance Liquid Chromatographic Determination of ,&Phenylethylamine in Human Plasma with Fluorescence Detection Junichi Ishida, Masatoshi Yamaguchi,l and Masaru Nakamura Faculty of Pharmaceutical Sciences,Fukuoka University, Nanakuma, Johnan-ku, Fukbku 814-01,Japan

Received

June

16,1989

A simple and highly sensitive method for the determination of &phenylethylamine in human plasma is investigated. The method employs high-performance liquid chromatography with fluorescence detection. & Phenylethylamine and p-methylbenzylamine (internal standard) in human plasma are isolated by cation-exchange chromatography on a Toyopak SP cartridge and then converted into the corresponding fluorescent derivatives with 3,4-dihydro-6,7-dimethoxy-4-methyl-3-oxoquinoxaline-2-carbonyl chloride, a fluorescence derivatization reagent for amines. The derivatives are separated within 30 min on a reversed-phase column, TSK gel ODS-lPOT, with isocratic elution, and detected fluorometrically. The detection limit of &phenylethylamine is 0.3 pmol/ml in plasma (S/N = 3). 0 ISSO Academic

Press,

Inc.

/3-Phenylethylamine (PEA)’ is a trace amine occurring naturally in body fluids and is now thought to be a neuromodulator in the central nervous system (1,2). The correlation between urinary PEA levels and neurological diseases such as schizophrenia (3) and depression (4,5) has been studied. It has been indicated that abnormal amounts of PEA are observed in patients with these diseases.However, the correlation concerning plasma PEA levels has remained unknown. This may be partially due to the lack of a sensitive, simple, and reproducible method. Some methods, involving gas chromatography-mass spectrometric (GC-MS) (6) and high-performance liq1 To whom correspondence should be addressed. * Abbreviations used: PEA, fi-phenylethylamine; MBA, p-methylbenzylamine; DMEQ-COCl, 3,4-dihydro-6,7-dimethoxy-4-methyl-3oxoquinoxaline-2-carbonyl chloride; GC-MS, gas chromatographymass spectrometry. 86

uid chromatographic (HPLC) (7,B) methods, have been proposed for the determination of PEA in human plasma. Although the GC-MS method (6) can offer the simultaneous determination of PEA, p-tyramine, and m-tyramine with picogram sensitivity, it requires expensive equipment and a rather tedious technique. A postcolumn HPLC method (7) with fluorescence detection using o-phthalaldehyde as a fluorogenic reagent has been applied to the analysis of PEA in biological samples. However, the method does not have a proper internal standard. Recently, a precolumn HPLC method (8) using fluorescamine has been reported for the simultaneous determination of some trace amines in plasma. The method, however, requires a complicated clean-up procedure using a two-step column separation. We previously developed 3,4-dihydro-6,7dimethoxy4-methyl-3-oxoquinoxaline-2-carbonyl chloride (DMEQCOCl) as a highly sensitive fluorescence derivatization reagent (9) for primary and secondary amines. This reagent reacts rapidly with the amines in acetonitrile in the presence of potassium carbonate to give highly fluorescent 2-substituted DMEQ carboxyamides. The purpose of the present research is to develop a simple, sensitive, and reproducible HPLC method using DMEQ-COCl for the quantification of PEA in plasma. p-Methylbenzylamine (MBA) does not occur in human plasma, and thus was used as an internal standard. REAGENTS

AND METHOD

Chemicals and solutions. All chemicals and solvents were of analytical reagent grade, unless stated otherwise. Distilled water, purified with a Milli Q II system, was used for all aqueous solutions. PEA and MBA were purchased from Tokyo Kasei (Tokyo, Japan). DMEQ-COCI was prepared as described previously (10). DMEQ-COCI solution (2 mM) was prepared in acetonitrile. The DMEQ-COCl solution was used within 3 h. MBA (IS; 0003~2697/90 $3.00 Copyright 0 1990 by Academic Press, Inc. All rights of reproduction in any form reserved.

LIQUID

CHROMATOGRAPHY

WITH

-

2

3

L

b

;(

1;

;4

87

DETECTION

particle size, 5 pm; Tosoh Corp., Tokyo, Japan). A mixture of acetonitrile and 50 mM ammonium acetate (33: 67, v/v) was used as a mobile phase. The flow rate was 1.0 ml/min. The column temperature was ambient (1825°C). Uncorrected fluorescence excitation and emission spectra of the eluates were measured with a Hitachi 65060 fluorescence spectrophotometer in 10 X lo-mm quartz cells; spectral bandwidths of 5 nm were used for both the excitation and emission monochromators. Procedure. A l.O-ml aliquot of plasma was mixed with 50 ~1of MBA and 0.5 ml of 70 mM hydrochloric acid and the mixture was poured into a Toyopak SP cartridge. The cartridge was washed successively with 5 ml of water (twice) and 1.8 ml of aqueous 40% acetonitrile (twice). The adsorbed amines were eluted with 3 ml of a mixture of acetonitrile and 1.0 M sodium chloride (2:3, v/v). To the eluate, 0.6 ml of 0.5 M sodium hydroxide and 6 ml of ethyl acetate were added. PEA and MBA were extracted by shaking for 10 min. The organic layer (ca. 6 ml) was evaporated to dryness in uucuo at 50°C and the residue was dissolved in 200 ~1 of acetonitrile containing 2.0% Triton X-405. To the solution, 100 ~1 of the DMEQ-COCI solution and ca. 3 mg of potassium car-

3

3

FLUORESCENCE

;2

Time (min) FIG. 1. Chromatogram of the DMEQ derivatives of PEA and MBA. A portion (200 ~1) of a standard mixture of PEA and MBA (0.5 nmol/ ml, each) was treated according to the procedure. Peaks: 1, PEA; 2, MBA, 3, reagent blank.

2.0 PM) solution was prepared in acetonitrile containing 2.0% (w/v) Triton X-405. A Toyopak SP cartridge (strong cation-exchange, sulfopropyl resin; Tosoh Corp., Tokyo, Japan) was washed successively with 1.8 ml of 2.0 M sodium hydroxide (twice), 5 ml of water (twice), 1.8 ml of concentrated hydrochloric acid-acetonitrile (1:9, v/v) (twice), 5 ml of water (twice), and 1.8 ml of 10 mM sodium acetate-hydrochloric acid buffer (pH 5.0) (twice). Human blood was obtained from fasting healthy volunteers in our laboratories. The blood (10 ml) was taken into a test tube containing 38 mg of citric acid trisodium salt dihydrate and centrifuged at 1OOOgat 4°C for 15 min. The resulting plasma was stored at -40°C until just before use. Apparatus and HPLC conditions. The HPLC system consisted of a Waters 6000A high-performance liquid chromatograph equipped with a Rheodyne 7125 syringeloading sample injector valve (20 ~1 loop) and a Shimadzu RF-530 fluorescence spectrometer fitted with a 12-~1 flow cell operating at an emission wavelength of 485 nm and an excitation wavelength of 406 nm. The column was a TSK gel ODS-120T (250 X 4.6 mm i.d.;

I

I

I

0

8

16

I

24

I

32

Time (min) FIG. 2. Chromatogram of the DMEQ derivatives of PEA in human plasma. A portion (1 ml) of the plasma sample (4.1 pmol/ml) was treated according to the procedure. Peaks: see legend to Fig. 1.

88

ISHIDA, TABLE

1

Concentrations of PEA in Normal Age

Sex

41 30 24 23 22 21 25 21 21 20 Mean SD

m m m m m m f f f f

YAMAGUCHI,

Human

Plasma

PEA (pmol/ml) 3.6 3.5 4.1 4.1 9.1 7.8 5.2 5.5 10.0 5.6 5.9 2.3

bonate were added. The mixture was allowed to stand at room temperature for ca. 1 min. The supernatant (20 ~1) of the final reaction mixture was injected into the chromatograph. For the establishment of the calibration graph, 50 ~1of the MBA (IS) solution was replaced with an IS solution containing PEA (0.5-100 pmol). The net peak height ratios of PEA and MBA were plotted against the spiked PEA concentrations. RESULTS

AND

DISCUSSION

HPLC Conditions Complete baseline separation of the DMEQ derivatives of PEA and MBA was achieved using a reversedphase column, TSK gel ODS-12OT, and acetonitrile-50 mM ammonium acetate (33:67, v/v) as the eluent. Figure 1 shows a typical chromatogram obtained with a standard mixture. The retention times for PEA and MBA were 20 and 27 min, respectively. The individual amines gave single peaks in the chromatogram. The eluates of the peaks (1 and 2) showed the same fluorescence excitation (maximum, 406 nm) and emission (maximum, 485 nm) spectra. Some biogenic amines (p- and m-tyramines, octopamine, phenylethanolamine, and some polyamines) reacted with DMEQ-COCl to give fluorescent compounds (9). However, the compounds were coeluted with DMEQ-COCl under the HPLC conditions used and did not interfere with the determination of PEA. Derivatization

Conditions

DMEQ-COCl gave the most intense and constant peaks for PEA and MBA at concentrations greater than 0.5 mM; thus a concentration of 2 mM was chosen. Reproducible peak heights were not obtainable without Triton X-405. This might be due to the adsorption of the amines on the glass wall (9). Triton X-405 [l.O-3.0% (w/

AND NAKAMURA

v) in acetonitrile] gave maximum and constant peak heights for the amines; 2.0% was selected as the optimum concentration. Potassium carbonate was used to facilitate the derivatization of the amines with DMEQCOCl. The peak heights were maximal and constant when the amount of potassium carbonate was l-5 mg; ca. 3 mg was employed in the procedure. The derivatization reaction of PEA and MBA with DMEQ-COCl proceeded rapidly independent of the temperature (0-100°C); the reaction complete within 30 s even at 0°C. Thus, standing for ca. 1 min at room temperature was used in the procedure. Determination of PEA in Human Plasma Chromatography. Figure 2 shows a typical chromatogram obtained with normal human plasma. The components of peaks 1 and 2 (Fig. 2) were identified as the DMEQ derivatives of PEA and MBA, respectively, on the basis of their retention times, cochromatography with the standard compounds, and fluorescence excitation and emission spectra of the eluates of the peaks in comparison with those in Fig. 1. No peaks were observed in the chromatogram at the retention times for PEA and MBA when the derivatization reaction was not performed. Clean-up ofplasma. A cation-exchange cartridge column (Toyopak SP) was used for sample pretreatment. PEA and MBA were retained effectively in the cartridge when the pH of the plasma sample was adjusted to around 5.5; 0.5 ml of 70 mM hydrochloric acid was used to adjust the pH of the sample. The amines adsorbed on the cartridge were effectively eluted with a mixture (3 ml) of acetonitrile and 1.0 M sodium chloride (2:3, v/v). The recoveries of PEA and MBA from the cartridge were 62.3 f 1.3 and 80.5 -t 3.7% (mean +- SD, n = 5), respectively. When the pH of the eluate from the cartridge was changed to strong alkaline (over pH 13) by adding 0.6 ml of 0.5 M sodium hydroxide, PEA and MBA were effectively extracted with ethyl acetate (6 ml) from the eluate. The extraction recoveries of PEA and MBA (50 pmol of each added to plasma) were 93.8 + 2.6 and 90.8 + 3.3% (mean + SD, n = 5), respectively.

TABLE 2 Comparison of Plasma Concentration of PEA Obtained the Present HPLC Method and the Other Methods Method GC-MS HPLC (postcolumn) HPLC (precolumn) Present method

PEA concentration 5.3 35.3 0.33 5.9

* + + f

(pmol/ml)

0.4 10.1 0.23 2.3

by

Ref.

(‘3 (7)

(8) -

LIQUID

CHROMATOGRAPHY

WITH

Linearity, detection limits, andprecision. A linear relationship was observed between the ratios of the peak height of PEA to that of the internal standard and the amounts of PEA added to plasma in the range 0.01-100 nmol/ml. The correlation coefficient of the calibration curve was 0.998. The lower limit of detection for PEA was 0.3 pmol/ml plasma at a signal-to-noise ratio of 3. This sensitivity is comparable with that of the fluorometric method using fluorescamine (8) and about 10 times higher than that of the postcolumn derivatization method (7). The within-day precision of the method was established by repeated determinations (n = 5) using normal human plasma containing 4.1 pmol/ml of PEA. The relative standard deviation was 5.2%. PEA concentrations in human plasma. The concentrations of PEA in plasma from healthy volunteers were determined by this method (Table 1). PEA concentrations in plasma have been determined by several researchers, although these values differ considerably as shown in Table 2. The plasma levels measured by our method are in good agreement with those obtained by the GC-MS method, which offers absolute identification of PEA. CONCLUSION

The present HPLC method provides a sensitive and simple determination of PEA in 1 ml of plasma. It needs

FLUORESCENCE

89

DETECTION

one cation-exchange cartridge followed by extraction with ethyl acetate for sample pretreatment and should be useful for clinical and biological investigations. REFERENCES 1. Wolf, 390.

M. E., and Mosnaim,

A. D. (1985)

2. Boulton,

A. A., and Juorio, L. (1982) try, pp. 189-222, Plenum, New York.

Gen. Phurmucol. Handbook

of Neurochemis-

3. Potkin,

S. G., Karoum, F., Chuang, L., Cannon-Spoor, lips, I., and Wyatt, R. J. (1979) Science 206,470-471.

4. Sabelli,

H. C., Fawcett, and Jeffriess, H. (1983)

5. Boulton,

J., Gusovsky, F., Javaid, Science 220,1187-1188.

A. A., and Milward,

L. (1971)

14,385-

H. E., PhilJ., Edwards,

J. Chromatogr.

J.,

57,

287-

296. 6. Karoum, Schwing,

F., Nasrallah, H., Potkin, S., Chuang, L., MoyerJ., Phillips, I., and Wyatt, R. J. (1979) J. Neurochem.

33,201-212. 7. Tsuji, Anal.

M., Ohi, Biochm.

8. Yonekura, Watanabe,

K., Taga,

C., Myojin,

T., and Takahashi,

S. (1986)

153,116-120.

T., Kamata, S., Wasa, M., Okada, A., Yamatodani, T., and Wada, H. (1988) J. Chromatogr. 427,320-325.

9. Ishida, J., Yamaguchi, M., Iwata, Anal. Chim. Actu 223,319-326. 10. Iwata, T., Yamaguchi, Y. (1986) J. Chromatogr.

M.,

T., and Nakamura,

Hara, S., Nakamura, 362, 209-216.

M.

A.,

(1989)

M., and Ohkura,