Spencer, K., and Price, C. P., Influence of reagent quality and re- action conditions ... Leonard M. Neckers, Lynn E. Delisi, and Richard Jed Wyatt. Phenylalanine ...
with saline or succinate buffer; he also suggested that these proteins then react immediately with BCG as if they were albumin (5). The method for albumin determination with the Vickers M300 analyzer was modified as follows. Dilute specimens fivefold with distilled water (primary dilutor) and mix a 70-jL aliquot with 1.3 mL of distilled water (secondary dilutor). Incubate the mixture at 37 #{176}C for 9 mm, add 1.3 mL of working BCG reagent and measure the absorbance at 640 nm after a further 12 s. The composition of the BCG reagent is as described above. I compared results by the modified method (12-s reaction) and unmodified method (9-mm reaction) with those by the Laurell rocket technique, analyzing 91 plasma specimens from hospital patients. Results are shown in Figures 3 and 4. In applying the “immediate” reaction to the Vickers M300 multichannel analyzer, the minimum time required for adding BCG to specimens and then measuring the absorbance is 12 s. Gustafsson (5) has reported that acceptable results for plasma albumin by BCG can be obtained if the reaction time is less than 8 to 10 s. Although shortening the reaction time from 9 mm to 12 s produces better agreement with the rocket technique, Figure 4 shows that there is still a positive intercept at 10 g/L on the y-axis. The IFCC Expert Panel on Proteins recommends that albumin should he measured by an immunospecific method, particularly at low concentrations, and that dye-binding
CLIN. CHEM. 27/1, 146-148
methods should only be used as screening procedures (7). Nevertheless, because of their ease of use and application to automated equipment, BCG dye-binding methods are still widely used. In view of this, these methods should be modified to give results that agree as closely as possible with immunospecific methods. I show here that this can be achieved by careful attention to reaction conditions.
References 1. Gustafsson, ,J. E. C., Improved specificity of serum albumin determination and estimation of ‘acute phase reactants” by use of the bromeresol green reaction. (‘un. (‘hem. 22, 616 (1977). 2. Spencer, K., and Price, C. P., Influence of reagent quality and reaction conditions on the determination of serum albumin by the hromcresol (1977).
green dye-binding
method.
Ann. (‘lin.
Riochem.
14, 105
3. lngwersen, S., and Raaho, E., Improved and more specific bromcresol green methods for the manual and automatic of serum albumin. (‘un. (‘him. ArIa. 88, 545 (1978).
determination
4. Peters, T., Serum albumin. Recent progress in the understanding of its structure and biosynthesis. (‘lin. (‘hem. 23,5(1977). Review. 5. Gustafsson, .J. E. C., Automated serum albumin determination by use of the immediate reaction with hromcresol green reagent. Clin. (‘hem. 24, 369 (1978).
6. Laurell, C. B., Quantitative estimation of proteins by electrophoresis is agarose gel containing antibodies. Anal. Biochem. 15, 45 (1966).
7. IF(’(’ Newsletter
13,4 (1976).
(1980)
Liquid-Chromatographic Quantification of Plasma Phenylalanine, Tyrosine, and Tryptophan Leonard M. Neckers, Lynn E. Delisi, and Richard Jed Wyatt Phenylalanine, tyrosine, and tryptophan are isolated and quantified by “high-pressure” liquid chromatography, with fluorescence detection. An isocratic mobile phase and reversed-phase column are used to provide rapid and reproducible measurement of these amino acids in as little as ito 2 ML of human plasma.
Additional Keyphrases: heritable disorders
.
screening
Concentrations of the aromatic amino acids tyrosine, tryptophan, and phenylalanine in plasma are of clinical interest for several reasons. Tryptophan participates in the regulation of brain serotonin metabolism (1,2), and the concentration of this amino acid in plasma has been studied in various disease states (3-6). Data on plasma phenylalanmne and tyrosine concentrations have been used in screening for heterozygotes for phenylketonuria (7, 8), and abnormal concentrations have been implicated in several diseases (6). A technique for monitoring tyrosine, tryptophan, and phenylalanine
that is both sensitive
and rapid
would
have
Adult I’syehiatrv Branch, I)ivision of Special Mental Health Research, Intramural Research Program. National Institute of Mental Health, Saint Elizabeths Hospital, Washington, IX’ 200:12. Received .Julv 10, 1980; accepted Oct. 28, 1980. 146
CLINICAL CHEMISTRY,
Vol. 27. No. 1, 1981
clear application in the clinical laboratory. In the past, plasma amino acid concentrations were best measured with an amino acid analyzer, a relatively complicated, expensive, and timeconsuming technique (9). More recently, several “highpressure” liquid-chromatographic methods for several amino acid analyses have been reported (10). We now report a method for the determination of tyrosine, tryptophan, and phenylalanine in the same plasma sample by use of an isocratic elution buffer and fluorescence detection of each amino acid, by virtue of its native fluorescence or by post-column derivatization with o- phthaldialdehyde.
Materials and Methods Chemicals. Unless otherwise stated, chemicals used in preparing the lithium citrate elution buffer and the borate buffer were of analytical grade (Fisher Chemical Co., Silver Spring, MD 20910). o-Phthaldialdehyde was from Sigma Chemical Co., St. Louis, MO 63178, and was used without further purification. Perchloric acid (60%, analytical grade) was obtained from Fisher Chemical Co. /3-2-Thienyl-DLalanine was obtained from Sigma Chemical Co. Apparatus. The chromatograph we used was assembled in our laboratory according to the basic details given by Meek (11). Both the elution buffer and the reagent were contained in individual stainless-steel tanks pressurized by nitrogen gas.
1.S.
Try
1 Try
‘.5.
Tyr
Ty r
1 II II
0
5$
1520
0
$
1216
Fig. 1. Chromatogram of tyrosine and tryptophan phenylalanine (right) in deproteinized plasma Tyr, tyrosine;
(left), and
IS., internal standard; Try, tryptophan; and P. phenylalanine. Time
scale is in minutes; 0’ is the injection point Fifty-microliter samples were introduced into the elution buffer stream by on-line injection, using a six-port highpressure injector valve (Valco Instruments, Houston, TX 77024). The 25 cm X 0.4 cm column we used was an RP-18 reversed-phase column (Bio-Rad Labs., Rockville Ctr., NY 11571). Fluorescence was detected with a FS 970 fluorometer equipped with a 10-ML flow cell (Schoeffel Instruments, Westwood, NJ 07675). All tubing precolumn was 316-grade stainless steel; all postcolumn (or post-reagent reservoir) tubing was of Teflon. Sample preparation. Collect venous blood in evacuated blood-collection tubes containing anticoagulant (ACD, NIH Formula “A”) and, without delay,centrifugeat 1000 X g for 15 mm. Remove the plasma layer and store it at -50 #{176}C until used. Thaw samples to be assayed and centrifuge them for 2 mm at 8000 X g. Mix 100 ML of plasma with 100 ML of 1 mol/L perchloric acid containing 10 mmol of mercaptoethanol per liter and 10 ML of /3-2-thienyl-DL-alanine (10mg in 20 mL of water) as internal standard. After centrifugation, dilute 20 ML of the supernate to 1 mL with water. Inject 50-ML aliquots of this finaldilution, equivalent to 0.5 ML of original plasma, into the chromatograph. Inject standards (50 ng per 50-ML injectu)n), prepared similarly, at the beginning and end of each run.
Operation
I
I’
0
20 0
Fig. 2. Chromatogram of tyrosine and 10 pmol, respectively)
20 and tryptophan
standards
(50
Left: all amino acids except Tyr, Try, and internal standard. Right: same sample with these added. Time in minutes
a 330-nm sharp-cutoff is used. The flow rate here is 0.5 mL/ mm, and one sample can be injected every 15 mm.
Results Figures 1,2, and 3 show typical chromatograms of phenylalanine, tyrosine, and tryptophan standards and of deproteinized plasma. 9-2-Thienyl-DI.-alanine is included as an internal standard. The standard curve for each amino acid is linear over the range 10 to 500 pmol (r = 0.97-0.98). It can be estimated from the signal-to-noise ratio of the amino acid peaks that the limit of detectability for phenylalanine, tyrosine, and tryptophan is 4, 5, and 1 pmol, respectively. This would mean that the smallest concentration (in mg/L) detectable in plasma would be 0.4 for phenylalanine, 0.5 for tyrosine, and 0.1 for tryptophan. Repeated injection of the same
LS
of the Instrument
Phenylalanine. The elutionbuffer contains,per liter, 0.6 mol of lithium hydroxide and 0.2 mol of sodium citrate and has a pH of 5.0. Pump the buffer at 0.4 mL/min through the
r
heated (55 #{176}C) RP-18 column. Pump o-phthaldialdehyde(100 mg in 50 mL of 0.5 mol/L borate buffer, pH 10.3, containing 1 mL of ethanol and 100 ML of mercapthoethanol) at 0.5 mL/min and mix with the column effluent by use of a Teflon T (Altex, Berkeley, CA 94710). Allow the column effluent to mix with o- phthaldialdehyde in a 2-m loop of 28-gauge Teflon tubing before entering the fluorometer. The wavelengths used for detection are 330 nm (excitation) and 450 nm (emission), and a 450-nm sharp-cutoff filter is used. One 50-ML sample can be injectedevery 20 mm. Tvrosine and tryptophan. Conditions identical to those for phenylalanine are used except that o-phthaldialdehyde and the mixing loop of tubing are eliminated and the detection wavelengths are 280-nm (excitation) and 330 nm (emission);
Fig. 3. Chromatogram
of phenylalanine
standard (50 pmol)
Left: amino acids except Phe and internal standard. Right: chromatogram with these added to the sample. Time in minutes.
CLINICAL CHEMISTRY. Vol.27.No. 1,1981
147
Table 1. Plasma Amino Acid Concentrations Concentration, ± SEM (and n)
Amino acid
Phenylalanine Tyrosine Tryptophan
11.2±
2.2(10)
6.3± 10.3±
1.4(10) 1.6(10)
Values in literature
108±21 7,9
2.4
9,8b (5.1-14.9)
#{149} see reference 13. See reference
14.
_____________
standard solution on the same day, 100 pmol per injection, yielded a CV of 2.1% (n = 10). Repeated assay of the same plasma specimen showed the method to be adequately reproducible (SD 3.6%, n = 8). Analytical recovery, as determined after adding known amounts of amino acid to plasma
samples, was 97-100%. The new method sufficiently resolves phenylalanmne, tyrosine, and tryptophan from the other amino acids in plasma. A synthetic standard, containing 21 amino acids (Sigma, kit LAA-21) in equimolar concentrations (100 pmol per injection) was analyzed by our technique. The results were, within experimental error, identical to the quantity of amino acid injected. Finally, the values we obtain for these three amino acids in normal individuals compare very closely to previously reported values (Table 1).
Discussion Of the variousmethods developed for assay of amino acids 13), only the amino acid analyzer allows for simultaneous determination of phenylalanine, tyrosine, and tryptophan. With our relatively easy and rapid method, one now can routinely monitor these three amino acids in as little as 1 or 2 jzL of plasma in only 40 mm. We have been using this method for 18 months to measure the plasma amino acids in schizophrenic patients; neuroleptic drugs, as well as many others, have no effect on results by the procedure, either in vivo or in vitro, nor have we encountered other problems. The column has only had to be replaced once.
in plasma (/2,
148
CLINICAL CHEMISTRY, Vol. 27, No. 1, 1981
References 1. Neckers, L. M., Biggio, G., Moja, E., and Meek,J.
L., Modulation
. of brain tryptophan hydroxylase activity by brain tryptophan J. Pharmacol. Exp. Ther. 201,110-116(1977).
content.
2. Fernstrom, J. D., and Wurtman, R. J., Brain serotonin content: Physiological dependence on plasma tryptophan levels. Science 173, 139-140
(1971).
3. Lehmann, 3., Tryptophan malabsorption in levodopa-treated parkinsonian patients. Effects of tryptophan on mental disturbances. Acta Med. Scand. 194,181-189 (1973). 4. Coppen, A., Eccleston, E. G., and Peet, M., Plasma tryptophan binding and depression. Adu. Biochem. Psvchopharmacol. 11, 325-333 (1974). 5. Domino, E. F., and Krause, R. R., Free and hound serum tryptophan in drug-free normal controls and chronic schizophrenic patients. Rio!. Psychiatry 8,265-279(1974). 6. Lowe, C. V., and Auerbach, V. H., Inborn errors of metabolism. In Text book of Pediatrics, W. E. Nelson, Ed., W. B. Saunders Co., Philadelphia, PA, 1964, pp 290-294.
7. Perry, T. L., Hansen, S., Tischler, B., and Hunting, R., Determination of heterozygosity for phenylketonuria on the amino acid analyser. Clin. (‘him. Acta 18, 51-56 (1967). 8. Jagenburg, R., Regardh, C.-G., and Rodjer, S., Detection of beterozygotes for phenylketonuria. Total body phenyalanine clearance and tyrosine in the plasma of fasting subjects compared. Clin. Chem. 23,1654-1660 (1977). 9. Benson, .J. V., Grodon, M. .J., and Patterson, J. A.. Accelerated chromatographic analysis of amino acids in physiological fluids containing glutamine and asparagine. Anal. Riochem. 18, 228-240 (1967). 10. Neckers, L. M., and Bohlen, P., HPLC: Applications to psychiatric research. In Physiochemical Methodologies in Psychiatric Research, I. Hanin and S. H. Koslow, Eds., Raven Press, New York, NY, 1980, pp 23-36. II. Meek, J. L., Application of inexpensive equipment for high pressure liquid chromatography to assays for taurine, GABA and 5-hydroxytryptophan. Anal. (‘hem. 48, 375-379 (1976).
12. Guilbault, C. C., and Froehlich, P. M., Rapid, sensitive, and sein serum. Clin. Chem. 19, 1112-1113
lective assay for tryptophan (1973).
13. Shen, R., and Abel, C. W., Phenylketonuria: A new method for the simultaneous determination of plasma phenylalanine and tyrosine. Science 197, 665-667 (1977). 14. Handbook of Biochemistry, H. A. Sober, Ed., The Chemical Rubber Co., Cleveland, OH, 1968, p B-55.
