1844
ANALYTICAL CHEMISTRY
2. The borohydride reduction of iodate appears to be an instantaneous reaction. 3. The presence of alkali in amounts greater than that necessary to neutralize all boric acid formed has no effect on the reaction between iodate and borohydride. 4. Commercial sodium borohydride deteriorates rapidly in aqueous solution, but is fairly stable in basic solutions-the higher the pH the greater the stability (Figure 2). REAGENTS
Sodium hydroxide, reagent grade, 0.5 N Sulfuric acid, reagent grade, 4 N Potassium iodate, standard, 0.25 N Sodium thiosulfate, standard, 0.10 A' Potassium iodide, reagent grade Amylose starch indicator PROCEDURE
A 20- to 25-mg. sample of sodium borohydride is dissolved i n 25 ml. of 0.5 N sodium hydroxide. Standard potassium iodatr (35.00 ml., 0.25 N )is added immediately, and the flask is swirled vigorously for 30 seconds. Two grams of potassium iodide are added, followed by 20 ml. of 4 AT sulfuric acid. The flask is allowed t o stand in darkness for 2 or 3 minutes before the liberated iodine is titrated with 0.10 A' sodium thiosulfate (starch indicator). Sodium borohydride is hygroscopic, so suitable precautions must be taken in handling it. A dry box may be used, or a larger sample may be weighed quickly and dissolved in a proportionately larger amount of 0.5 N sodium hydroxide, from which an aliquot containing approximately 20 to 25 mg. of sodium borohydride is taken. The weight of sodium borohydride can be calculated by means of the following formula: Mg. of NaBH4 = (ml. of KI03 X normality ml. of Na2S203 X normality) X 4.731 The equivalent weight of sodium borohydride is equal to one
Time in hours
Figure 2. Deterioration of Commercial Sodium Borohydride in Different Solutions
eighth of the molecular weight (BH48H+ 8e).
+
+ 3H20--+ H2B03- +
.4CKNOW LEDGM ENT
The authors are grateful to Howard S. Clark, Illinois Geological Survey, Urbana, Ill., for his interest and valuable suggestions. LITERATURE CITED
(1) Chaikin,
S.W., and Brown, W, G., J . A m .
Chem. SOC.,71, 122
(1949).
(2) Mathem, SI.B., J . Biol. Cheni., 176, 229 (194Sj. R E C E I V Efor D reyiew July 7, 1952.
;iccepted August 4, 1952.
Use of Anthrone Reaction for Determination of Carbohydrates in the Presence of Serum Protein b1. R. SHETLAR Research Laboratory, Veterans Administration Hospital, and T h e Oklahoma Medical Research Foundation, Oklahoma City, Okla. HEN compared to the tryptophan method of Shetlar, wFoster, and Everett ( 5 ) , the anthrone method for the determination of the polysaccharides in serum recently proposed by Graff, Greenspan, Lehman, and Holechek ( I ) has an advantage of greater specificity for carbohydrates. However, the absorption curves resulting from the reaction of serum proteins with anthrone show an absorption maximum at 530 mp as well as the maximum at 620 to 625 mp which is characteristic of the hexose-anthrone reaction. Morris (3) noted that a red color was obtained with certain proteins, and Seifter, Dayton, Sovic, and Muntnyler ( 4 ) reported that tryptophan reacted with anthrone to produce a solution with a maximum absorption at 515 mp. None of these investigators apparently realized that some reaction between tryptophan and carbohydrate would also undoubtedly occur as reported in the work of Shetlar, Foster, and Everett ( 5 ) . The maximum absorption for this reaction is at 500 mp. The following work was undertaken in order to clarify the question as to what the interfering reaction or reactions might be and to determine whether this reaction might interfere with the determination of polysaccharides in serum and in tissue. REAGENTS
Anthrone reagent, 0.15% anthrone in 95% sulfuric acid. As recommended by Viles and Silverman ( 6 ) , this reagent was
aged at least 4 hours and was discarded after it became more than 9 days old. The solution was kept in the refrigerator when not In use. Sulfuric acid, 95% by volume. Sine hundred and fifty milliliters of concentrated sulfuric acid (Du Pont, C.P. reagent grade, specific gravity, 1.84 at 15" C.), xere added to 50 nil. of distilled water. EQUIPMENT
All measurements of optical density were made in a >lode1 14 Coleman spectrophotometer. EXPERIMENTAL
Method of Heating after Adding Anthrone Solution. Graff et al. ( I ) employed a method by which the anthrone was added to the carbohydrate solution at room temperature. The heat evolved due to hydration of the sulfuric acid is sufficient to cause the reaction. However, Seifter et al. ( 4 ) reported that considerable discrepancies occurred among identical samples of glucose by this procedure, and devised a method consisting of adding the anthrone to cooled sugar solutions, followed by a heating period in a boiling water bath. The question remained as to which of these procedures would favor the development of the reaction characteristic of carbohydrate-anthrone and retard the development of the reaction characterized by the absorption a t 530 mp. Consequently, human serum was diluted with twice the volume
V O L U M E 2 4 , N O . 11, N O V E M B E R 1 9 5 2
1845
of 0.9% saline solution. A 0.2-ml. aliquot was pipetted dropwise into a glass-stoppered 15-ml. test tube containing 10 ml. of absolute ethanol. The precipitate was centrifuged out and washed by suspension in 10 ml. of absolute ethanol and again centrifuged. The alcohol was drained o f f , and the precipitate was dissolved in 3 ml. of distilled water, Sets of duplicate tubes were then treated as follom: Treatment before Adding Anthrone 1. None. Reagent added a t room
Treatment after Optical Adding .inthrone 530 R.Ip 620 3 I p Stood 15 min. a t room temp. 0 198 0 238
-
*.-.-.-r2. I.
.5.
Ratio D620/ D53O
Galactose +mannose
1 20
containing 7% protein, were reacted with anthrone an absorption curve (Figure 1) with a maximum absorption at 515 to 520 mp was obtained. However, the optical density at 520 mp is not nearly enough to account for the absorption found in serumanthrone curves as shown by the theoretical curve in Figure 1. As at least three independent reactions may occur in the reaction of anthrone with serum pr0tein-i.e. , anthrone with carbohydrate] tryptophan with carbohydrate, and tryptophan with anthrone-the theoretical curve was obtained by adding the individual curves for each of these reactions together. This theoretical curve difiered markedly from serum curves, however, when tryptophan was added to the pure carbohydrate solu-
+ tryptophan.
- - - - - -,3. Galactose +mannose+anthrone.
,.-...’.............. .e’
4. Theoretical galactose+ mannose+ tryptophan t anthrone. 5. Actual ”
Changes with Heating Time in Boiling Water Bath. Samples of serum protein prepared as described above were subjected to heating in the boiling water bath for 5, 10, 15, and 20 minutes in order to establish the most effective time. The development of the greatest optical density a t 620 m p occurred after 10 minutes of heating while that a t 530 m r occurred after 20 minutes. Interaction of Anthrone, Hexose, and Tryptophan. As tryptophan has been implicated by other workers as the substance which reacts with anthrone to give a red color with a maximum absorption at 515 mp, the reaction of this substance was further investigated. When 80 micrograms of tryptophan, which is slightly more than the amount present in 0.0667 ml. of serum
Polysaccharides. It appears that the reaction between tryptophan, anthrone, and carbchydrate might be utilized to determine carbohydrate in the presence of protein by adding
aliquot was pipetted dropwise into a glassstoppered 15-ml. test tube containing 10 ml. of absolute ethanol. The precipitate was centrifuged out and washed by suspension in 10 ml. of absolute ethanol and again centrifuged. This procedure has previously been shown ( 4 ) to yield a precipitate which does not carry down an appreciable amount of free serum glucose and to precipitate the polysaccharides of serum quantitatively. After pouring off the washed ethanol the precipitates were allowed to drain for at least 30 minutes. One milliliter of a 0.1% solution of tryptophan was pipetted into each tube, followed by 2 ml. of water. A standard tube containing 25 micrcgrams of mannose, 25 micrograms of galactose, 1 ml. of 0.1% tryptophan, and 2 ml. of water was included in each set
ANALYTICAL CHEMISTRY
1846 of determinations. A blank tube containing 1 ml. of 0.1% tryptophan and 2 ml. of water was also prepared in each run. The tubes were cooled for 15 minutes in an ice bath. Six milliliters of the 0.15% anthrone solution were added with a buret or pipet t o each tube. The tubes were stoppered and mixed by inversion. They were then placed in a boiling water bath for 20 minutes. rit the end of this time they were cooled in a cold water bath for 10 minutes. They were then transferred t o colorimeter tubes, and the optical density was measured against the blank solution at 520 mfi in the Coleman 14 spectrophotometer.
Table I. Summary of Comparison of Tryptophan and Anthrone-Twptophan Methods for Determination of Serum Polysaccharide S O .
Method Tryptophan Anthronetryptophan Try tophan .intRronetryptophan
of Sera
Species Human
19
Human Dog
i
117-256 105-178
I59 i 129 0
7
96-173
133 9
12
Dog
MaxiMean mum Value, DifferM g , % 1Ig. c7c ence 113-24.5 138 4 13 Range,
Correlation Coeffirient 0 970
14
RESULTS
sults were obtained with the anthrone tryptophan method when compared to the tryptophan method, but the differences are too small to be of significance.
-4summary of comparative data obtained from 19 human sera and 7 dog $era is given in Table I. Slightly higher average re-
DlSCUSSlON
.-.
n-----.r
Serum 2 0 micrograms tryptophan
40
0
440
480
520
560
600
640
680
720
' 1
760
SbO
WAVE LENGTH Mu
Figure 2. Absorption Curves of Serum Reacted with Anthrone and of Mixtures [of 50 hlicrograms of Galactose, 50 Micrograms of Xlannose, and Various AmountsJ of Tryptophan Reacted with 4nthrone Mixture containing 40 micrograms of tryptophan closely resembles that of the serum protein.
After the original preparation of this manuscript, 11. 11. Graff, Clinical Research Cnit, Sational Cancer Institute, United States Public Health Service, Baltimore, 1Id., informed the author that Lehman ( 2 ) showed that acetone, pyruvic acid. pyruvic aldehyde, lactir acid, uric acid, and ascorbic acid all reacted with anthrone t o give solutions which have absorption maxima from 470 to530nip. Lehman noted that glutathione, glycylglycylglycine, and glycyltF1osine caused color enhancement of the reaction of tryptophan with anthrone. Khen these compounds were subjected to the procedure as described above, none of them reacted strongly enough with the anthrone tryptophan mixture to cause appreciable i n t e r f e r e n c e x i t h t h e method dexribed here for serum polysaccharides. However, in e x t r a c t s where these compounds may be found in higher concentration or where the amount. of protein i q large as compared to the hexose content, a significant interference may occur. ACKhOWLEDG\IENT
1
Acknodedgment is made of the aseistance of Virginia Richmond in determining the effects of the compounds listed by 11. 11, Graff on the tryptophan anthrone reaction.
...............,
Golactore t mannose curve. S e r u m curve.
LITERATURE CITED
I
4 h
440
480
Si0
5bO
600
640
680
720
760
800
WAVE LENGTH Mu
Figure 3. Comparison of a Serum Protein Absorption Curve with Absorption Curve of Mixture of Galactose and Mannose Mixture consisted of 25 micrograms of galactose and 25 micrograms of mannose. samples contained 3 mg. of added tryptophan.
Both
(1) Graff. hZ. It,,G r e e n s p a n , E. 11..L e h m a n , I. R., a n d Holechek, J. J., J . Lab. Clin., M e d . , 37, 736 (1951). ( 2 ) L e h m a n , I. R., unpublished d a t a . (3) Morris, D. J., Science, 107, 254 (1948). (4) Seifter, S a m . , D a y t o n , Seymour, Novic, B., a n d Rluntn-yler. E d w a r d , Arch. Biochem.. 25, 191 (1950). ( 5 ) Shetlar, 31. R., Foster, J. V., a n d E v e r e t t . 3T. R.. Proc. SOC.Exptl. Biol. M e d . , 67, 125 (1948). (6) Viles, F. J., a n d Silverman, Leslie, ANAL. CHEM..21, 950 (1949). RECEIVED for review March 13, 1952. Accepted 7 , 1952. Contribution 23 from the Oklahoma Medical Research Foundation. .4ugust
