Document Sample
					    THE      DETERMINATION                         OF DEXTRAN    IN BLOOD                            AND
                URINE WITH                        ANTHRONE    REAGENT

                                          BY    JOSEPH         H. ROE
   (From   the Department          of Biochemistry,       School       of Medicine,    George   Washington
                                     University,    Washington,          D. C.)

                       (Received         for   publication,        January     11, 1954)

     Klevas (1) determined dextran in blood by the Hagedorn-Jensen           (2)
copper reduction method, applied to filtrate prepared by the Somogyi
technique (3), after acid hydrolysis of the filtrate for 6 hours. Hint and

                                                                                                             Downloaded from by guest, on December 27, 2011
Thorsen (4) estimated dextran in blood by alkalinizing trichloroacetic acid
filtrate, treating the filtrate with standard copper solution, and deter-
mining residual copper after 4 hours shaking. These methods are time-
consuming and cumbersome.         A considerable advance in analytical tech-
nique was made by Bloom and Willcox (5) who applied the anthrone
reaction of Dreywood (6) to dextran precipitated from solution by alcohol.
In the Bloom and Willcox method the blood proteins are removed by boil-
ing with 30 per cent KOH, a procedure that gave a blank value of 12.4 f
3.2 mg. per cent in normal subjects (5). Procedures in which the dextran
in trichloroacetic acid filtrates of blood and urine is precipitated by alcohol
and determined turbidimetrically        have been published by Metcalf and
Rousselot (7) and by Jacobsson and Hansen (8). Metcalf and Rousselot
reported their method somewhat inaccurate at concentrations of 1 mg.
per ml. but of satisfactory reliability at concentrations of 2 to 16 mg. per
     The color formed when carbohydrate is treated with anthrone reagent is
influenced by the temperature, the time of heating, and the concentration
of sulfuric acid and anthrone used. These factors, also conditions for opti-
mal precipitation of dextran by alcohol, have been studied, and a method
has been developed for the determination of clinical dextran, average mo-
lecular weight 70,000, in blood and urine.

    1. Anthrone reagent. A 0.05 per cent anthrone reagent in 72 per cent
H&S04 is prepared.       To 280 ml. of distilled water add carefully 720 ml.
of concentrated HzS04, sp. gr. 1.84, of highest purity.      Prepare 1 to 5
liters of this solution.   Estimate the amount of reagent needed for 2 days
work and dissolve 50 mg. of recrystallized anthrone in each 100 ml. of 72
per cent sulfuric acid to be used. The mixture is warmed to 70-80”, then
890                        DETERMINATION          OF   DEXTRAN

cooled to room temperature before use. Upon standing, anthrone reagent,
when boiled with carbohydrate,         yields a dark green color, less desirable,
and of lower optical density, than the bluish green color formed with fresh
reagent.      This reagent may be prepared for use one day and used during
the following day. It may be used longer, since a standard is always run
along with the unknown tubes, but it is recommended that fresh reagent
be prepared every 2 days. Anthrone recrystallized              once from alcohol is
   2. 6 per cent trichloroacetic acid.
   3. Standard glucose solutions.     (a) Stock solution.     Obtain highest purity
glucose and dry in a vacuum oven at 60-70”.           Dissolve 100 mg. of glucose

                                                                                                    Downloaded from by guest, on December 27, 2011
in 100 ml. of saturated benzoic acid solution.
    (b) Working standard.      Place 5 ml. of the stock solution of glucose in a
100 ml. volumetric flask and make up to volume with saturated benzoic
acid solution.      2 ml. of this solution, containing 0.1 mg. of glucose, are
used as a standard.
   Dextran may be used as a standard in place of glucose. Clinical dextran
 (Commercial Solvents Corporation)         yields a color intensity with anthrone
reagent equivalent to 111 per cent of that given by an equal quantity of
   4. 95 per cent ethyl alcohol. The alcohol must be free from anthrone-
sensitive materials.
   Blood plasma is deproteinized with 5 per cent trichloroacetic         acid. Urine
is diluted with 5 per cent trichloroacetic      acid and filtered if protein is pres-
 ent. A dilution is made that will yield filtrate containing 10 to 200 y per
ml. The trichloroacetic      acid mixture of plasma or urine is filtered through
Whatman No. 42, or other acid-washed,             filter paper. Ordinary filter pa-
pers have been found to contain some water-soluble,              alcohol-precipitable
   Pipette 1 ml. of trichloroacetic    acid filtrate of plasma or 1 ml. of diluted
urine into a 15 ml. conical Pyrex centrifuge tube. Continue until 1 ml.
of each of the unknown solutions has been placed in a similar centrifuge
tube. To obtain the most reliable results duplicate tubes for each un-
known are analyzed.
   To the 1 ml. of filtrate in each centrifuge tube add 5 ml. of 95 per cent
ethyl alcohol. The alcohol is blown forcefully from a pipette into the solu-
tion in a manner that will cause thorough mixing.               Cork the tubes with
clean rubber stoppers and set aside at room temperature overnight, or, if
the results are desired upon the same day, cap the tubes loosely with clean
   1 I am indebted   to Dr. H. E. Stavely, Commercial            Solvents Corporation,   for this
                                    J.   H.    ROE                                891

 glass marbles or bulbs and place them in a 3740” water bath for 3 hours.
  Centrifuge for 15 minutes at 3000 r.p.m., decant the alcoholic mixture, and
 place the tubes in an inverted position upon a gauze mat in a beaker. Let
 the tubes drain until dry (about 15 minutes) ; then place them in a rack.
 With a pipette add 2 ml. of distilled water to each centrifuge tube, letting
 the water run down the sides of the tube in a manner that will wash down
 the inner surfaces of the tube. Tap the bottom of each tube upon the
 palm of the hand in a way that will dislodge the mat of dextran and mix
 the contents.
    For a standard, pipette 2 ml. of the working standard solution, contain-
 ing 0.1 mg. of glucose, into a 15 ml. conical Pyrex centrifuge tube. It is
 desirable to prepare two or three standard tubes for each series of unknowns

                                                                                        Downloaded from by guest, on December 27, 2011
    Into the standard and unknown tubes pipette 10 ml. of anthrone reagent.
 This reagent is blown forcefully from the pipette into the solution in the
 tube in a way that will produce a mixing of the contents.        Be careful to
 obtain a thorough mixing and make sure that this step is carried out uni-
 formly in all of the tubes. Hold the tube in a vertical position and agitate
 it in a way that will mix more thoroughly the contents in the top level and
 will incorporate the small portions of the reagent that may have splashed
 on the sides of the tube near the top. Stopper each tube by inserting
jirmly a rubber stopper through which has been passed a glass tube about
4 inches in length.    To prepare this stopper, cut off about one-third of the
 smaller end of a size 0 rubber stopper, bore a hole in the larger end, and
insert a glass tube about 3 mm. in diameter; this device is used to keep
water from getting into the tube from the water bath.
    Set the stoppered tubes in a rack suitable for immersing in a water bath
 (No. 26231, Will Corporation, or a wire test-tube basket wired with splices
crossed at right angles to each other).    Set the rack of tubes in a bath of
tap water to bring each tube to the same temperature; then place it in a
boiling water bath.      Be sure the bath is uniformly heated and each tube
is maintained at approximately the same temperature.         After 13 minutes
in the boiling water bath remove the rack of tubes to a bath of tap water
and let stand until the tubes are at approximately         room temperature.
Wipe the tubes and stoppers dry with a towel. Remove the stoppers and
decant each centrifuge tube into a matched photoelectric calorimeter tube.
Keep the tubes away from direct sunlight.         Read with a 620 rnp wave-
length filter.   To prepare a blank for setting the calorimeter add 10 ml. of
anthrone reagent to 2 ml. of water and mix

   For PIasma-DU/DS   X 0.1 X dilution        of plasma X 100 X 0.9 = mg. of dextran
per 100 ml.
892                            DETERMINATI0.N            OF   DEXTRAN

   For Urine-DU/DS          X 0.1 X dilution     of urine    X ml. in sample      X 0.9 = mg. of
dextran  per sample.
     DU = the optical  density    of the unknown;        DS = the optical density    of the stand-
ard; 0.1 = mg. of glucose      in 2 ml. of standard        solution; 0.9 = the factor      for con-
verting  the glucose  value    to the dextran     value.


                  Heating Time and Concentration of Sulfuric Acid
   In some of the methods in which the anthrone reagent is employed for
the determination of carbohydrate, the heat resulting from mixing sulfuric
acid with n-at,er causes the reaction to take place (5, 9-12). This is an
inconstant source of energy for this reaction. The depth of color produced

                                                                                                          Downloaded from by guest, on December 27, 2011
    FIG. 1. Relation   of heating   time   to color production.          2 ml. of solution   containing
0.1 mg. of dextran   were mixed with       10 ml. of 0.05 per     cent    anthrone   in H&304 solution
as indicated    and heated    at 100”.

by this technique is influenced by the size and shape of the vessel used,
the volume of the solution, and the speed with which the sulfuric acid is
added and mixed.       Greater precision is obtained by those methods in
which the mixture of anthrone, sulfuric acid, and carbohydrate is heated
for a definite time in a constant temperature bath (13-17).        Scott and
Melvin (17) observed excellent precision with their procedure in which 1
volume of dextran solution is mixed with 2 volumes of 0.2 per cent an-
throne in 95 per cent sulfuric acid and heated in an ethylene glycol bath
at 90” for 16 minutes.     The standard deviation for 345 sets of duplicate
determinations, in absorbance units, was f0.0029,     corresponding to 0.48
per cent at an absorbance level of 0.6.
   In nearly all of the anthrone methods for carbohydrate determination a
66 per cent concentration of HzS04 by volume (2 volumes of 95 per cent
H&O4 to 1 volume of carbohydrate solution) is used. As shown in Fig. 1,
Curve A, 65 per cent concentration of HzS04 by volume gives a rounded
                                                                  J.     H.    ROE                                                          893

curve for the optical density when the mixture is heated in a boiling water
bath.    With H&S04 solution that is 60 per cent by volume, the optical
density is constant at the heating time of 11 to 15 minutes (Curve B).
We, therefore, adopted this concentration of HzS04 for our work, which
is the same as that used by Black (15) for the determination of methyl-
cellulose. Curve C, Fig. 1, showing results with 57 per cent H&04 by
volume, also has a flat portion that would make a good working range, but
since the solubility of anthrone, and the optical density of the color pro-
duced, decrease with a decreasing concentration of HzS04, 60 per cent by
volume of HzS04 is the concentration of choice.

                                                                    TABLE           I

                                                                                                                                                  Downloaded from by guest, on December 27, 2011
                       Factors        Affecting           Precipitation                 of Dextran    with    Alcohol
    Time      of standing,         3 hours           at room           temperature.
                                                                                              -                 -
                                                                                                                        Per cent recovery

                                 De&-an        solvent                                               Sample          Parts alcohol (95 per
                                                                                                                    cent) to parts solution



Distilled       water..                                                         .                     50                  0            67
1% sodium          chloride.                                                                          50                 97           100
Urine     diluted       with     water,        1:50.                                                  40                 38            91
    L‘          ‘(         “        “          1:25..............                                     40                              103
    <‘          it         6‘       “          1:lO.                                                  40                109           119
    “          “
                           “     5%       CCl,COOH,               1:50.                               40                              104
    “          ‘<          Cc    5%               ‘(              1:25.....                           40                              101
    L‘         “           “     5%               ‘I              l:lO.....                           40                              105

                                      Precipitation                 of Dextran by Alcohol
   Effect of Ionic Concentration-The     effect of ionic concentration on the
precipitation of dextran is shown by the data of Table I. With urine di-
luted 1: 50, 1: 25, and 1: 10 volumes with water recoveries of added dextran,
when 5 parts of alcohol were added to 1 part of solution and the mixtures
were allowed to stand for 3 hours, were 38, 102, and 109 per cent, respec-
tively.    These data suggested that the greater concentration of urinary
substances in the mixture favored the precipitation of dextran.       This as-
sumption was confirmed by experiments with water and sodium chloride
solutions. In a 3 hour period no dextran was precipitated from a pure
water solution by adding 5 parts of alcohol to 1 part of dextran solution,
and only 67 per cent was precipitated by a 10: 1 ratio of alcohol to water.
With dextran in a 1 per cent sodium chloride solution, the recoveries were
97 and 100 per cent with 5: 1 and 10: 1 alcohol to water concentrations,
894                    DETERMINATION      OF   DEXTRAN

   With another sample of urine diluted 1: 50, 1: 25, and 1: 10 volumes with
water, recoveries of added dextran were 91, 103, and 119 per cent, respec-
tively, with an alcohol to water concentration of 10: 1. In the analysis
with the 1: 10 dilution, showing a 119 per cent recovery, a considerably in-
creased amount of precipitate was observed. It thus appeared that this
falsely high value was due to the precipitation by alcohol of phosphates,
or other substances, which adsorb or contain anthrone-sensitive materials.
This explanation is in agreement with further studies of Table I in which
the same urine was diluted with 5 per cent trichloroacetic acid, which pre-
vented the precipitation of phosphates by alcohol; the recoveries in these
analyses were satisfactory, regardless of the urinary dilution made.

                                                                                 Downloaded from by guest, on December 27, 2011
   The data of Table I show that filtrates of biological materials must have
a certain ionic concentration and an acid pH to permit a satisfactory pre-
cipitation of small amounts of dextran by alcohol. They reveal the un-
suitability of water as a diluent for urine when determining dextran by al-
cohol precipitation:     at high dilutions recoveries are too low and at low
dilutions recoveries are falsely high.
   Ethyl alcohol is an efficient reagent for precipitating dextran from solu-
tion.     10 y of dextran in 1 ml. of 5 per cent trichloroacetic acid are re-
covered quantitatively by the technique outlined above. If it is desired to
check the alcohol precipitation step, a freshly prepared 5 per cent trichloro-
acetic acid standard solution of dextran is used instead of a benzoic acid
standard solution, which has too low an ionic concentration to produce
quantitative    precipitation of dextran by alcohol.
   E$ect of Temperature-It       was found that the rate of precipitation of
dextran from 5 per cent trichloroacetic acid solution by alcohol is more
rapid at 40” than at 18-22” or at 3”. For this reason the tubes containing
the alcohol-unknown mixture are placed in a’water bath at 3740” for 3
hours when results are desired on the same day. If results are not needed
at once, it is simpler and preferable to allow the alcohol precipitation step
to take place at room temperatures overnight.

                              Protein Precipitant
   5 per cent trichloroacetic acid was found satisfactory for the determina-
tion of dextran in blood plasma because this acid is an efficient protein pre-
cipitant and provides an acid solution of suitable ionic concentration for
alcohol precipitation of dextran.   Deproteinization  with barium hydroxide
and zinc sulfate gave low recoveries of added dextran.     The Folin-Wu re-
agents gave slightly high recoveries, owing apparently to the precipitation
of some tungstic acid by the alcohol which probably adsorbed traces of
   ,4n important feature of this method is that trichloroacetic acid filtrates
                                                J.   H.   ROE                                             895

of blood and urine diluted with trichloroacetic acid yield negative blanks.
The method is, therefore, specific for dextran in blood and urine.

                                              Color Quality
   The agreement of the color used in this method with Beer’s law is excel-
lent, as shown in Fig. 2. Values that fall on a straight line were obtained
with quantities of 10 to 200 y of dextran, which correspond to readings of
90 to 15 with the Evelyn calorimeter.
   The color produced in this method has a high degree of stability in the
dark. A group of tubes containing the colored mixture, placed in the dark
and read at intervals, showed no change in optical density in 2 hours and

                                                                                                                 Downloaded from by guest, on December 27, 2011

                        25        so              100          IS0                             too
                                       MICROGRAIlS    OF DEXTRAN
   FIG. 2. Agreement       of color  intensity      with Beer’s        law.    2 ml. of dextran    solution
were mixed   with    10 ml. of anthrone        reagent,     consisting      of 0.05 per cent anthrone       in
72 per cent H$304 by volume,        and heated          at 100” for 12 minutes.

only a 7 per cent decrease in color intensity in 18 hours. When placed in
sunlight, the color fades: a decrease in optical density of 14 per cent in 1
hour was observed in tubes placed by a window in direct sunlight.

                                       Precision and Recovery
   A high degree of precision was observed with this method upon pure
dextran solution.    In twelve determinations upon a solution containing
0.1 mg. of dextran per ml., by the technique outlined above, there was
found, in optical density units, a standard deviation of ho.0022 at an
optical density level of 0.387, or 0.56 per cent. In twelve determinations
in which the alcohol precipitation step was omitted, the standard deviation
was f0.0014 at an optical density level of 0.387, or 0.36 per cent.
   In ten experiments in which dextran was added to blood plasma, the
following per cent recoveries were obtained: 100, 101, 102, 102, 103, 104,
104,105,105,107,   the mean recovery being 103.3 per cent. When dextran
896                                    DETERMINATION              OF   DEXTRAN

was added to ten urines, the following per cent recoveries were observed:
96, 98, 98, 101, 101, 102, 103, 103, 104, 105, the mean recovery being 101
per cent.


   A method has been developed for                           the determination of dextran in blood
and urine. Dextran is precipitated                             from trichloroacetic acid filtrate by
alcohol; it is placed in 60 per cent                        by volume HzS04 solution containing
0.042 per cent of anthrone, boiled                          for 13 minutes, and compared colori-
metrically with a standard glucose                          or dextran solution treated for color
production similarly.

                                                                                                                          Downloaded from by guest, on December 27, 2011
  Part of the expense of this work was contributed                                       by the Commercial
Solvents Corporation, Terre Haute, Indiana.

    1.   Klevas,        S., Svensk Kern.      Tidskr.,       66, 262 (1944).
   2.    Hagedorn,          H., and Jensen,        B., B&hem.           Z., 136, 46 (1923).
   3.    Somogyi,         M., J. Biol.    Chem., 86, 655 (1930).
   4.    Hint,     H. C., and Thorsen,           G., Acta them. &and.,             1, 808 (1947).
   5.    Bloom,        W. L., and Wiilcox,          M. L., Proc. Sot. Exp. Biol. and Med.,               76, 3 (1951).
   6.    Dreywood,           R., Ind. and Eng. Chem.,            Anal.     Ed., 18, 499 (1946).
   7.    Metcalf,       W., and Rousselot,          L. B., J. Lab. and Clin. Med.,            40, 901 (1952).
   8.    Jacobsson,         L., and Hansen,        H., &and.        J. Clin. and Lab. Invest.,       4, 352 (1952).
   9.    Morse,       E. E., Anal.     Chem.,      19, 1012 (1947).
10.      Morris,       D. L., Science,     107, 254 (1948).
Il.      Viles,     F. J., and Silverman,           L., Anal.     Chem., 21, 950 (1949).
12.      Samsel,       E. P., and DeLap,          R. A., AnaZ. Chem.,           23, 1795 (1951).
13.      Seifter,      S., Dayton,     S., Novic,         B., and Munt,wyler,         E., Arch.   Biochem.,     26, 191
14.      McCready,           R. M., Guggolz,       J., Silviera,     V., and Owens,       H. S., Anal.      Chem.,  22,
             1156 (1950).
15.      Black,       H. C., Jr., Anal.       Chem.,        23, 1792 (1951).
16.      Koehler,        L. H., Anal.     Chem.,       24, 1576 (1952).
17.      Scott,      T. A., and Melvin,        E. H., Anal.         Chem., 25, 1656 (1953).

Shared By: