Analyst, September, 1974, Vol. 99, pp. 565-569 565 Determination of the Tyramine Content of South African Cheeses by Gas - Liquid Chromatography BY 1:. R. KAPLAN, N. SAPEIKA (Department of Pharmacology, Medical School, Observatory, Cape Province, South Africa) (Fishing Industry Research Institute, University of Cape Town. Rondebosch, Cape Province, South Africa) AND I. M. MOODIE The tyramine content of foodstuffs and beverages is of pharmacological and therapeutic importance. A simplified method for its extraction from various cheeses and the application of gas-chromatographic analysis are presented. TYRAMINE is an indirectly acting sympathomimetic amine that releases noradrenaline from adrenergic neurones. Foods such as cheese, red wine, yeast extracts and pickled herrings may contain large amounts of tyramine.l Dietary tyramine does not normally produce any hypertensive effects because both the intestinal wall and liver contain the enzyme monoamine oxidase, which oxidises the tyramine before it can pass into the general circulation.However, if the activity of the enzyme at these sites is inhibited by a monoamine oxidase inhibitor drug, any tyramine absorbed from the gut passes freely into the circulation, thus causing the exaggerated hypertensive and other effects, which are produced by adrenergic substances and which may be mild, severe or even fatal. There is also evidence that this recognised “cheese reaction” may be partly caused by the presence of other substances such as histamine and 5-hydroxytryptamine which, like tyramine, would remain protected from destruction by inhibited monoamine oxidase.The development of the hypertensive reaction depends on the concentration of tyramine in the food. With cheese, tyramine is formed during the cheese-making process by the bacterial degradation of milk proteins to tyrosine (and other amino-acids) and subsequent decarboxylation of tyrosine to tyramine. The amount of tyramine in different cheeses varies greatly and even within a single cheese the tyramine content can vary considerably between a sample taken from the centre of the cheese and a sample taken near the rind.2 This variation may explain why the ingestion of a portion of certain cheeses may not necessarily cause a hypertensive c r i s i ~ . ~ , ~ Previous w ~ r k e r s ~ ~ ~ ~ ~ have used extraction and fluorimetric assay techniques in the determination of tyramine in cheeses.However, the methods of extraction are laborious and involve steps in which there is a real risk of introducing errors. In particular, special care must be exercised in maintaining the correct pH in order to ensure optimum extraction of tyramine as the free base. At the high pH used it is not possible to extract all the tyramine and frequent checks have to be made on the efficiency of extraction. In the present investigation, representative samples from each cheese were selected and the tyramine, as its hydrochloride salt, was extracted in acidic medium, the amount of tyramine obtained being determined by gas chromatography. As amines cannot be satisfactorily determined by direct gas-chromatographic analysis, it becomes necessary to adopt a procedure involving the forination of derivatives.Several methods for the separation and determination of biological amines by gas chromatography of their derivatives have been de~cribed.~-~ The procedure used in the proposed method is based on that of Irvine and Saxbylo for the separation of primary and secondary amines. The tyramine hydrochloride (I) is converted at room temperature into the ON-bistrifluoroacetyl (TFA) derivative (11) with a large excess of trifluoroacetic anhydride and determined by gas chromatography, using ephedrine hydrochloride as an internal standard. 0 SAC and the authors.566 ---l KAPLAN et al. : DETERMINATION OF THE TYRAMINE CONTENT [Analyst, Vol. 99 OCOCF3 $ - Excess (CF3CO120 0 CH2CHzNH2.HCI CH2 CH2 NHCOCF3 I II METHOD REAGENTS- Tyramine hydrochloride (Nutritional Biochemicals Corp., m.p.271 to 273 "C) and ephe- drine hydrochloride (May and Baker, m.p. 217 to 219 "C) were checked for purity by forming their derivatives and subjecting the latter to gas chromatography (Figs. 1 and 2). L I I I I I I I I . - 18 14 10 6 2 0 Ti me/m i nu tes t E4 n 0 E k 18 14 10 i Time/minutes Fig. 1. Chromatogram of ephedrine Fig. 2. Chromatogram of tyramine derivative on 2 per cent. OV-17 - 1 per derivative on 2 per cent. OV-17 - 1 per cent. OV-210 on acid-washed, silanised cent. OV-210 on acid-washed, silanised Chromosorb W. Column programmed Chromosorb W. Column programmed from 120 to 220 "C a t 6 "C min-' from 120 to 220 "C a t 6 "C min-1 EXTRACTION- A known amount of cheese (1.0 -& 0.4 g) was homogenised with 20 ml of 0.01 N hydrochloric acid and the mixture centrifuged for 5 minutes.The supernatant fluid was removed and sampled for derivative formation. DERIVATIVE FORMATI ON- TO a 5-ml aliquot of the aqueous extract, 400 pg of ephedrine hydrochloride were added and the solution was evaporated to dryness in vacuo. The residual solid was azeotropically dried twice by repeated additions of anhydrous methylene chloride, which were removed by evaporation in vacuo. The residue was then treated with 3 ml of a 1 + 3 trifluoroacetic anhydride - methylene chloride solution followed by ultrasonic mixing, which yielded a solution of the derivatives that was immediately available for gas-chromatographic analysis.Inj ec- tions of 2 4 of this solution were made as soon as possible after derivative preparation as prolonged storage can lead to degradation of the products. The procedure differs from that used by Senll in that the excess of trifluoroacetic an- hydride is not removed from the final product, which is thereby protected from any traces of moisture with which it might come into contact. Samples were taken from different parts of each particular brand of cheese.September, 19741 GAS - LIQUID CHROMATOGRAPHIC ANALYSIS- A Hewlett-Packard 7610 gas chromatograph fitted with flame-ionisation detectors and coupled to a Hewlett-Packard 3370A electronic integrator was used. The columns used were of glass, U-shaped (1.5 m x 4 mm i.d.) and silanised, and were packed with 2 per cent.OV-17 and 1 per cent. OV-210 on acid-washed DMCS Chromosorb W (80 to 100 mesh) and conditioned for 16 hours at 235 "C with nitrogen carrier gas flowing at the rate of 20 to 30 ml min-1. Injector and detector temperatures were 205 and 260 O C , respec- tively. The column temperature for analytical runs was programmed from 115 to 220 "C at 6 "C min -l. A stock solution of tyramine hydrochloride (25.00 mg; 144 mmol) and ephedrine hydro- chloride (29-03 mg; 144 mmol) in 50 ml of 0.01 N hydrochloric acid was prepared. A typical chromatogram of cheese extract is shown in Fig. 3. OF SOUTH AFRICAN CHEESES BY GAS - LIQUID CHROMATOGRAPHY 567 t 0 Lu n LT 1 1 1 1 1 1 1 1 1 18 14 10 6 2 0 Time/minutes Fig. 3. Typical chromatogram of acid-soluble components of cheese (mature cheddar), after derivative formation, on 2 per cent.OV-17 - 1 per cent. OV-210 on acid-washed, silanised Chromosorb W. Column programmed from 115 to 220 "C at 6 "C min-1. A, ephedrine, and B, tyramine derivatives RESULTS AND DISCUSSION The amount of tyramine present in each sample was calculated after first establishing the relative molar response (RMR) of the tyramine derivative (T) with respect to that of the internal standard (IS), ephedrine hydrochloride. Thus the value of R.MRT/Is was determined by repeated analysis of derivatives prepared from aliquots of a stock standard solution con- taining equimolar amounts of tyramine hydrochloride and ephedrine hydrochloride, using the following equation- . . * * (1) RMRTp = - ... . AIBInIs where A is the peak area and n the number of moles. present in the stock solution, then As equimolar amounts of tyramine hydrochloride and ephedrine hydrochloride are Values for this relative molar response were determined from ten separate chromato- graphic analyses and are shown together with the retention and peak area values in Table I.568 KAPLAN et d.: DETERMINATION OF THE TYRAMINE CONTENT [AndySt, VOl. 99 TABLE I PEAK AREA VALUES FOR STOCK SOLUTION No. of analysis 1 2 3 4 5 6 7 8 9 10 Area of TFA- tyramine peak 6868 3850 3480 2328 6278 2435 4706 6669 2566 2162 ( A T) Area of TFA- ephedrine peak 8538 4715 4564 2739 7854 3026 6060 8085 3145 2620 ( A 1s) 0.80 0.82 0.76 0.85 0.80 0.80 0.78 0.82 0.82 0.83 After an accurately known amount of internal standard had been added to the aliquot of cheese extract, derivatives were formed in the mixture and chromatographed, the peak areas for tyramine and internal standard (shown in Table 11) being obtained.By taking equation (1) above in expanded form and rearranging the terms, an equation for the amount of tyramine is readily obtained: where A is the peak area, g the amount and M the relative molecular mass, then By using the results shown in Table 11, the average value for RMRTilS (0.81, derived from values in Table I) and the relative molecular masses of tyramine and ephedrine hydrochloride, the amounts of tyramine in the extracts can be calculated. The tyramine contents of the original cheese samples are then determined and are shown in Table 111. TABLE I1 PEAK AREA VALUES FOR VARIOUS CHEESES Sample of cheese Roquefort Cheshire Camembert Cottage (cream) Pont 1’Eveque Cheddar (mature) Cheddar (1st grade) Area of tyramine peak 3108 62 1 36 10 124 1818 2884 TFA- (AT) Area of TFA- ephedrine peak 1553 2762 3643 11 510 15 430 3094 6029 !A IS) Sample of cheese Brie Cottage (low fat) Cheddar (wedges) Cheddar (wedges) Cheddar (wedges) Gruy&re Sweet milk Area of tyramine peak 63 1 64 2358 424 1846 122 152 TFA- (AT) Area of ephedrine peak 5865 12 640 10 230 3183 5179 5419 9261 TFA- ( A 1s) In order to ensure that the extraction was operating satisfactorily, a sample of cheese was split into four portions.Three portions were “spiked” with different levels of tyramine and then together with the “unspiked” fourth portion were extracted and the mixtures centrifuged.Internal standard was added to 5-ml aliquots of the extract and the analysis conducted as already described. The tyramine contents (Table IV) of the four samples were calculated by using the results shown in Table IV and equation (3). A comparison of the calculated value of added tyramine with the actual value of added tyramine shows good agreement and reflects an efficient extract ion procedure.September, 19741 OF SOUTH AFRICAN CHEESES BY GAS - LIQUID CHROMATOGRAPHY TABLE I11 TYRAMINE CONTENTS OF VARIOUS CHEESES 569 Sample of cheese Roquefort Cheshire Camembert Cottage (cream) Pont 1’Eveque Cheddar (mature) Cheddar (1st grade) Brie Cottage (low fat) Cheddar (wedges) Cheddar (wedges) Cheddar (wedges) Gruybre Sweet milk Amount of cheese sample/ g 1.0226 1.0177 1.0334 1.0315 1.0015 1.0188 1.0156 1.0197 1.0300 1-0026 1.0200 1~0000 1~0000 1.0080 Amount of tyramine (gT) in 5 ml of extract/ 167.8 75.5 3.3 1.0 2.7 197.4 160.7 36.2 1.7 77.4 44.7 119.7 7.6 5.5 rLg Tyramine content of cheese/ 656.4 296.7 12.8 5.0 10.8 775.0 632.9 142.0 6.6 308.8 175.3 478.8 30-4 21.8 r g g-’ From the results it can be seen that aged (mature) cheese ( e g ., Cheddar) contains relatively large concentrations of tyramine compared with those found in cheese of the “cottage” variety. Such cheese with a high tyramine content must especially be avoided by persons who take anti-depressive drugs of the monoamine oxidase inhibitor variety; certain other foods con- taining tyramine should also be avoided. TABLE IV EFFICIENCY OF EXTRACTION OF TYRAMINE FROM CHEESE Area of Area of TFA-tyramine TFA-ephedrine Tyramine Amount of cheese Tyramine peak peak found in sample/g added/pg (AT) ( A IS) extract/ p g 0.9400 - 14 688 24 026 82 1 1.0712 1580 32 342 18 567 2340 1.0980 790 19 169 15 611 1650 1.0440 395 14 954 16 611 1209 Apparent tyramine added/pg 1519 829 388 - 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. REFERENCES Stockley, I. H., Phavm. J., 1973, 210, 590. Price, K. and Smith, E., Lancet, 1971, i, 130. Sjoqvist, F., Proc. R. Soc. Med., 1965, 58, 967. Spector, S., Melmon, K., Lovenberg, W., and Sjoerdsma, A., J . Pharmac. Exp. They., 1963, 140, 229. Horwitz, D., Lovenberg, W., Engelman, K., and Sjoerdsma, A., J . Amer. Med. Ass., 1964, 188, 1108. Sen, N. P. and McGreer, P. L., Biochem., Biophys. Res. Commun., 1963, 13, 390. Beckett, A. H. and Wilkinson, G. R., J. Pharm. Pharmac., 1965, 17, 104s. Capella, P. and Horning, E. C . , Analyt. Chem., 1966, 38, 316. Cancalon, P. and Klingman, J. D., J. Chromat. Sci., 1972, 10, 253. Irvine, W. J., and Saxby, M. J.. J. Chromat., 1969, 43, 129. Sen, N. P., J. F d Sci., 1969, 34, 22. Received September loth, 1973 Amended Mavch llth, 1974 Accepted Mawh 19th, 1974