ing tryptophan are much less reactive than free tryptophan; hence proteins must ... absorbance of the resulting pinkish purple color is measured after 40 min.
ANALYTIC4L
RIOC HEMISTRY
77, 37x-386 (1977)
A Simple Calorimetric Method for the Determination of Tryptophanl S. M. MAHABOOB BASHA AND R. MICHAEL ROBERTS~ Deportmerzt
of Biochrmistry,
University
of Florida.
Gainesville.
Florida
32610
Received March 29, 1976: accepted September 14, 1976 A simple. sensitive, and reproducible calorimetric method for the determination of tryptophan in amounts as low as 2 fog is described. it is based on the oxidation of tryptophan by sodium nitrite and the coupling of the oxidized product to the leucodye N-l-(naphthyljethylenediamine dihydrochloride. The purple-pink product has an absorption maximum at 550 nm. There is no interference by carbohydrates. other amino acids, neutral salts, or a number of other compounds likely to be found in tissue hydrolysates. A number of indole derivatives including indole-3-acetic acid also react to give a colored product. Dipeptides containing tryptophan are much less reactive than free tryptophan; hence proteins must be hydrolyzed completely for the method to be useful. The assay is carried out at room temperature and can be modified easily to increase or decrease its sensitivity. It has been employed to determine the tryptophan content of a number of proteins following alkaline hydrolysis. Generally, values obtained were in close agreement with values reported in the literature.
Although there are several methods for the determination of tryptophan, its analysis remains tedious, and results are often uncertain. These methods, which include calorimetry, fluorometry, spectrophotometry, and microbiological assay, have been reviewed comprehensively (1). Most of the calorimetric methods involve oxidation of the condensation products of tryptophan with various aldehydes (2-9). Although several refinements have been reported (lo- 14), such methods are prone to various forms of interference or give unstable, colored products and inconsistent values for tryptophan (12,15- 17). Although tryptophan content is normally determined after alkaline hydrolysis of proteins, improvements to prevent oxidative destruction of tryptophan during acid hydrolysis have also been made (l&19). Nevertheless. there still exists a need for a rapid, accurate, and flexible calorimetric method for determining tryptophan in a variety of soluble and insoluble proteins of tissues, foodstuffs. and animal feeds. ’ Research supported by Grants HD-08560 from the National Institutes of Health and DMF 741A016 from the National Science Foundation. * Recipient of a Career Development Award (K4AM70389) from the National Institutes of Health. 37x Copyright 0 1977 hy Academic Preys. Inc. All ughis d rcproduct~~o in any form reserved.
ISSN 0003.2697
TRYPTOPHAN
379
DETERMINATION
.6
600
360
FIG. 1. (a) Visible absorption spectra of the products produced in the standard assay system using 0.25 pmol of: (1) tryptophan, (2) indole, (3) indole-3-acetic acid, (4) tryptophanamide, (5) L-tryptophyl-L-leucine, (6) Shydroxyindole-3-acetic acid, and (7) Shydroxytryptamine. (b) Ultraviolet absorption spectra of: (1) tryptophan, (2) tryptophan + HCI, and (3) the product produced following a 30-min incubation with HCI-sodium nitrite as described in the standard assay system.
METHODS Materials. Bovine hemoglobin and serum albumin, cr-casein and plactoglobulin (cow’s milk), ribonuclease and trypsin (bovine pancreas), myoglobin (bovine skeletal muscle), histone (calf thymus), fetuin (fetal calf serum), human transferrin, phosvitin, lysozyme and ovalbumin (hen’s eggs), cytochrome c (horse heart), ammonium sulfamate, N-l-(naphthyl)ethylenediamine dihydrochloride, the various tryptophan derivatives, and all other biochemicals were purchased from the Sigma Chemical Co., St. Louis, MO. Coforimetric procedure. The standard procedure we have used is described below. Note that the final volume of reagents is 4.5 ml, but the method could easily be miniaturized, thus increasing its sensitivity. A sample containing 1 to 100 pg of tryptophan (in a volume of 0.5 ml) is thoroughly mixed with 0.5 ml of 4 N HCl. Then 0.5 ml of sodium nitrite (0.4%, w/v) is added to the acidified sample and mixed. The solution is allowed to stand at room temperature (30 to 60 min) before addition of 1 ml of ammonium sulfamate (OS%, w/v) to destroy excess nitrite. This solution is thoroughly mixed. After 3 to 15 min, 2 ml of N-l-(naphthyl)ethylenediamine dihydrochloride (0.05% w/v, in ethanol) is added and mixed. The absorbance of the resulting pinkish purple color is measured after 40 min at 550 nm on a Beckman Model 25 spectrophotometer.
380
BASHA AND ROBERTS
FIG. 2. The linear relationship between tryptophan concentration and extinction in the standard assay system for 3- and 30-min incubation periods with NaNO,. Other conditions were as described in the Methods section.
Alkaline hydrolysis ofprotein samples. Weighed samples of protein (2 to 4 mg) were hydrolyzed with 5 M KOH in sealed ampuls in an atmosphere of argon for 10 hr in an autoclave at 120°C. This gave the maximal release of tryptophan from a-casein, but the hydrolysis time may have to be extended for highly insoluble samples of protein. The hydrolysates were cooled and acidified with a slight molar excess of HCl. The colloidal silicate was removed by centrifugation (20), and the pellet was washed at least three times with water. The clear supernatant fractions were combined and made up to a known volume (usually 5 ml), and duplicate aliquots were taken for analysis. RESULTS
Absorption of colored product. The pinkish purple product formed in the assay absorbed light strongly between 500 and 580 nm with a maximum at 550 nm (Fig. la). For comparison, the absorption spectra of the product obtained with a number of tryptophan derivatives such as tryptamine, 5-hydroxytryptamine, indole-3-acetic acid, and L-tryptophyl-L-leucine, are also included. Figure lb shows the ultraviolet absorption spectra of tryptophan and
TRYPTOPHAN
20
40
381
DETERMINATION
MINUTES
60
100
200
FIG. 3. Relationship between extinction at 550 nm and time of incubation with NaNO, using three different incubation temperatures. Procedures were otherwise carried out under standard conditions at room temperature.
the product formed after treatment with nitrous acid. Note the loss of the typical tryptophan chromophore, with its characteristic maximum at 280 nm, and the appearance of an absorption peak at 260 to 265 nm, which is probably the result of oxidation of the indole moiety to oxindole (21). Relationship of absorbance to tryptophan concentration. Under the standard conditions of assay there is a linear relationship between the absorbance at 550 nm and the amount of tryptophan added over the concentration range of 2 to 100 pg (Fig. 2). If the time of incubation with sodium nitrite is shortened to 3 min, a linear relationship is also obtained, but the color is considerably less intense. Therefore, the method has builtin flexibility in the sense that the sensitivity of the assay can be controlled by choosing appropriate incubation periods with the oxidizing agent. Effect of time and temperature of incubation with NaN02. The development of final color can be accelerated by conducting the oxidation with NaNO, at higher temperatures than used in the standard assay (Fig. 3). For example, maximum color was obtained using a IO-min incubation at 37°C compared to 23°C where the maximum was reached between 30 and 60 min. As expected, the concentration of NaNO, is also important (Table 1). Provided that the other reagents are kept constant, the optimal amount of NaNO, required for maximum sensitivity is between 2 and 3 mg. Concentrations above 4 mg gave a brownish yellow product, which is probably a result of adding insufficient ammonium sulfamate to destroy nitrous acid completely. Using 100 pg of tryptophan and the standard conditions of assay, the relative molar ratio of tryptophan, sodium nitrite, ammonium sulfamate, and dye is 1:60:90:8, respectively.
382
BASHA AND ROBERTS TABLE ASSAY
NaNO, concentration (mg)
1
FOR TRYPTOPHAN~
A 550 nm
Test color
0.00
0.000
0.25 0.50
0.212 0.279 0.342 0.408 0.510 0.550 0.542 0.536 0.520 0.401 0.361 0.244 0.218
Colorless Normal Normal Normal Normal Normal Normal Normal Normal Normal Brownish purple Light brown Golden yellow Golden yellow Golden yellow Golden yellow
1.00 1.50 2.00 2.50 3.00 3.50 4.00 4.50 5.00 7.00
10.00 15.00 20.00
0.182 0.163
a Forty micrograms of tryptophan were assayed as described in Methods, except for NaNO, concentration which was altered appropriately.
Effect of duration of incubation with ammonium sulfamate. Ammonium sulfamate (NH,SO,NH,) is added in order to decompose excess nitrous acid. Incubation periods between 3 and 15 min were found to be optimal for final color development. Interfering substances. The possible interference in the tryptophan assay of amino acids (100 pg/lOO pg of tryptophan) and monosaccharides (up to 20 mg/lOO pg of tryptophan), which might also be present in tissue or protein hydrolysates, were tested. None of the common amino acids, amino sugars, hexoses, or hexuronic acids typically found in biological materials interfered with color production. Several amines and amides were also tested. These included spermidine, cadaverine, niacinamide, glycinamide, triethylamine, benzylamine, and 1,6-diaminohexane. None of these compounds interfered when they were included in the assay in amounts up to 100 pg. Higher concentrations were not tested but are unlikely to be encountered under normal conditions for tryptophan assay. In addition to the above organic compounds, there was no effect of neutral salts such as NaCl, KN03, KCl, MgCl*, MnCI,, and CaCl, (each up to 0.1 M final concentration in the complete assay mixture). Certain other salts, such as Na&O,, Na2HP04, NaH,PO,, and sodium acetate which are likely to influence the reaction pH, do cause some reduction in the final intensity of the color when used at these high concentrations.
TRYPTOPHAN
TABLE REACTION
OF VARIOUS
383
DETERMINATION 2
INDOLE DERIVATIVES FOR TRYPTOPHAN
IN THE STANDARD
ASSAY
Tryptophan derivatives
Color produced” (%)
Test color
Wavelength of maximum absorption (nm)
Tryptamine t.-Tryptophanamide 5-Hydroxytryptamine (serotonin) t-Tryptophan ethyl ester L-Tryptophyl-L-leucine L-Tryptophyl-L-glutamic acid L-Glycyl-L-tryptophan r-Leucyl-r-tryptophan Indole-3-acetic acid 5-Hydroxyindole-3-acetic acid 3-lndolepynrvic acid Indole
16 37 5 77 30 14 12 13 58 13 46 72
Normal Normal Light brown Normal Normal Bluish purple Pink Pink Red Brown Red Red
540- 560 535-555 400-550 545-555 540-555 540-560 525-550 530-555 530-540 400-560 530-545 525-540
a Equimolar amounts (0.25 pmol) of the tryptophan derivatives were assayed, and the color produced by each is expressed as a percentage of the color response to tryptophan.
However, it may be possible to overcome this difficulty by increasing the concentration of HCI appropriately. However, since only NaCl or KC1 are likely to be the major salt contaminants of hydrolyzed and subsequently neutralized protein samples, we conclude that the assay is not prone to interference by salts under normal circumstances. 2-Mercaptoethanol, a reducing agent, did not interfere with the assay at concentrations up to 10 mM. Above this concentration, the intensity of the purple-pink color was reduced. However, although this reagent is commonly included in tissue extracts and solutions of proteins, it is relatively unstable and decomposes under the conditions of protein hydrolysis. Effect of tryptophan derivatives. We have tested various tryptophan derivatives including dipeptides in the standard assay (Table 2). The most reactive were tryptophan ethylester, indole, indole-3-acetic acid, and 3indolepyruvic acid. The colored products in each case absorbed maximally at rather shorter wavelengths than the product formed with tryptophan alone (see Table 2 and Fig. la). Interestingly, peptide-bound tryptophan also reacted in the assay, and L-tryptophyl-L-leucine was more reactive than L-leucyl-L-tryptophan. All of the dipeptides gave a considerably lower color intensity than free tryptophan, and it is doubtful whether the method could be adapted for use with partially hydrolyzed proteins. The Shydroxyl derivatives of tryptamine and indole-3-acetic acid did not produce the usual purple-pink coloration (Fig. 1, Table 2) suggesting that substitutions on the indole ring interfere with normal color development.
384
BASHA AND ROBERTS TABLE
THE TRYPTOPHAN HYDROLYSIS,
3
CONTENT OF VARIOUS PROTEINS DETERMINED AFTER ALKALINE AND A COMPARISON OF THE RESULTS WITH PREVIOUSLY PUBLISHED VALUES
Tryptophan content (mol of tryptophan/mol of protein) Protein
Molecular weight
After alkaline hydrolysisa
Reported values
Bovine serum albumin a-Casein Ribonuclease A Myoglobin Histone F II p-Lactoglobulin Trypsin Hemoglobin Cytochrome c Ovalbumin Fetuin Transferrin Phosvitin Lysozyme
66,000 26,000 12,640 16,890 15,500 37,700 23,800 64,500 13,400 46,000 50,750 77,000 45,000 14,900
2.17 1.15 0.06 1.55 0.04 4.25 3.04 5.50 1.01 2.26 1.82 7.30 1.20 5.97
2.0b.C l.Id oe, O.lb 2.0” O.(Y 4.09 3.2b, 4.0’ 6.0’ 0.8d, l.Oe 2.7d 2.0h 8.0’ 1.0’ 6.0”
a Two to four milligrams of protein were hydrolyzed with 5 M KOH in an autoclave at 120°C for 10 hr, and the hydrolyzates were analyzed for tryptophan. b Sasaki et al. (22). c Hugli and Moore (23). d Spande and Witkop (21). e Dayhoff (24). f Tristram and Smith (25). g Spies (14). h Spiro and Spiro (26). i Parker and Beam (27). j Allerton and Perlmann (28).
Determination of tryptophan in proteins. The application of this method to determination of protein-bound tryptophan was tested on alkaline hydrolysates of various proteins. The conditions were optimized for cycasein, but no special precautions were taken to include tryptophanprotective agents, although oxygen was excluded. In addition, the presence of sugars, amino sugars, or uranic acids (1 mg/4 mg of protein) prior to hydrolysis did not interfere with the tryptophan yield. The method was in good agreement with published values for 14 proteins tested (Table 3), including the traditionally “difficult” protein, lysozyme. Hydrolysis of lysozyme and casein in the presence of added cystine and serine (0.6 mg/5 mg of protein), compounds which have been reported to destroy some tryptophan during alkaline treatment (29), did not affect the tryptophan recovery in our experiment.
TRYPTOPHAN
DETERMINATION
385
DISCUSSION
Many of the previous calorimetric methods for estimating free tryptophan in a mixture of other amino acids involve the formation of a usually colorless complex between tryptophan and an aldehyde such as p-dimethylaminobenzaldehyde. Such condensation products yield a chromophore upon oxidation. However, it has proven difficult to regulate the oxidation step; the colors produced in these assays are frequently unstable, and the assay results are inconsistent. In addition, most of the procedures are tedious to perform and prone to many forms of interference. Koshland and co-workers (30,3 1) have used 2-hydroxy-S-nitrobenzylbromide to react with tryptophan in intact proteins. This method, which assumes equal reactivity of all residues, requires an intermediate chromatographic step that renders it less attractive for routine analysis of large numbers of protein samples. Low and inconsistent results have also been reported (32). Fluorescence and direct ultraviolet-absorption methods take advantage of the absorption and fluorescent maxima of tryptophan (22,33), or the modification of these specific characteristics by reagents such as Nbromosuccinamide (34). However, these methods generally require the protein samples to be both soluble and denatured (35). In addition, contaminants, cofactors, and solution turbidity can all interfere with such spectral analyses. A recent refinement in which proteins are partially digested by proteases appears promising (22). Nevertheless, the applicability of such methods to crude samples remains questionable. The nature of the colored reaction products described here is not known. Our rationale in developing the method was to oxidize (21) and diazotize the tryptophan by nitrous acid and to couple the reaction product to leucodye in order to yield a colored chromophore. The dipeptides, probably due to steric effects, do not react to the extent of free tryptophan. Irrespective of the structure of the chromophore, it appears stable for a limited time and has a high extinction coefficient. The assay also seems free from interference and is specific for tryptophan or related compounds, including the plant growth hormone indole-3-acetic acid for whose determination it might be adapted. Because of the simplicity and flexibility of the procedure, it should be possible to automate the analysis. ACKNOWLEDGMENT The authors thank Dr. C. M. Allen for reviewing the manuscript.
REFERENCES 1. Friedman, M., and Finley, J. W. (1971) J. Agr. Food Chem. 19, 626-631. 2. Adamkiewicz, A. (1874) Arch. Physiol. (Pfiigers) 9, 156-162. 3. Hopkins. F. G., and Cole, S. W. (190l)Proc. Roy. Sm. London 68, 21-30.
386 4. 5. 6. 7. 8. 9. 10.
BASHAANDROBERTS
Shaw, J. L. D., and McFarlane, W. D. (1940) J. Biol. Chem. 132, 387-392. Rosenheim, 0. (1906) Biochem. J. 1, 233-240. Fearson, W. R. (1920) Biochem. J. 14, 548-564. Kraus, I. (1925) J. Biol. Chem. 63, 157-178. Rhode, E. (1905) 2. Physiol. Chem. 44, 161- 168. Herzfeld, E. (1913) Biochem. Z. 56, 258-266. Graham, C. E., Smith, E. P., Hier, S. W., and Klein, D. (1947) J. Biol. Chem. 168, 711-716. 11. Spies, J. R., and Chambers, D. C. (1948) Anal. Chem. 20, 30-39. 12. Spies, J. R. (1950)Anal. Chem. 22, 1447-1449. 13. Fischl, J. (1960) J. Biol. Chem. 235, 999-1001. 14. Spies, J. R. (1967)Anal. Chem. 39, 1412-1416. 15. Light, A., and Smith, E. L. (1963) in The Proteins (Neurath, H., ed.), Vol. 1, p. 1, Academic Press, New York. 16. Gruen, L. C., and Rivett, D. W. (1971)Anal. Biochem. 44, 519-522. 17. DasGupta, B. R. (1969) Anal. Biochem. 30, 284-287. 18. Matsubura, H., and Sasaki, R. M. (1969) Biochem. Biophys. Res. Commun. 35, 175- 181. 19. Liu, T. Y., and Chang, Y. H. (1971) J. Biol. Chem. 246, 2842-2848. 20. Duggan, E. D., and Udenfriend, S. (1956) J. Biol. Chem. 223, 313-319. 21. Spande, T. F., and Witkop, B. (1967) in Methods in Enzymology (Hirs, C. H. W., ed.), Vol. 9, pp. 498-506, Academic Press, New York. 22. Sasaki, T., Abrams, B., and Horecker, B. L. (1975) Anal. Biochem. 65, 396-404. 23. Hugh, T. E., and Moore, S. (1972) J. Biol. Chem. 247, 2828-2834. 24. Dayhoff, M. 0. (1972) Atlas of Protein Sequence and Structure, Vol. 5, p. 418, National Biomedical Research Foundation, Washington, D.C. 25. Tristram, G. R., and Smith, R. H. (1963) in The Proteins (Neurath, H., ed.) Vol. 1, pp. 46-51, Academic Press, New York. 26. Spiro, M. J., and Spiro, R. G. (1962) J. Biol. Chem. 237, 1.507-1510. 27. Parker, W. C., and Beam, A. G. (1967) J. Exp. Med. 115, 83-105. 28. Allerton, S. E., and Perlmann, G. E. (1967) J. Biol. Chem. 240, 3892-3898. 29. Spies, J. R., and Chambers, D. C. (1949) Anal. Chem. 21, 1249-1266. 30. Horton, R. H., and Koshland, D. E. (1965) J. Amer. Chem. Sot. 87, 1126- 1132. 31. Barman, T. E., and Koshland, E. E. (1967) J. Biol. Chem. 242, 5771-5776. 32. Burton, P. M., Hall, D. C., and Josse, J. (1970) J. Biol. Chem. 245, 4346-4352. 33. Bencze, W. L., and Schmid, K. (1957) Anal. Chem. 29, 1193-1196. 34. Patchornick, A., Lawson, W. B., and Witkop, B. (1958) J. Amer. Chem. Sot. 80, 4747-4748. 3.5. Edelhoch, H. (1967) Biochemistry 6, 1948-1954.
