CLIN. CHEM.
27/2,
328-330
(1981)
Spectrophotofluorometry
of Serotonin in Blood Platelets
Govlndankutty T. Vatassery,”2 Monica A. Sheridan,1 and Ann M. Krezowski”
A rapid, sensitive, and reproducible assay for determination
of platelet serotonin is reported. Serotonin is extracted into
an ascorbic acid solution by freezing and thawing and
sonication. Ascorbic acid stabilizes the serotonin in the
extract, and added ethanol enhances the final fluorescence
of serotonin, which is measured in a concentrated hydrochloric acid medium (activation at 295 nm and emission
at 540 nm). The method has an average coefficient of
variation of 4.1 %; 95% of added standards is accounted
for. The average serotonin content of platelets from 15
men, ages 30-65 years, was 0.67 (SE 0.03) .tg/1O9
platelets, which compares favorably with previously reported values. 5-Hydroxyindole compounds such as 5hydroxyindole-3-acetic
acid and 5-hydroxytryptophan
interfere with the assay, but there is relatively little of these
compounds in platelets.
The biogenic amine serotonin (5 HT, 5-hydroxytryptamine) occurs in man primarily in serotonergic neurons in the
brain, in the enterochromaffin cells of the intestine, and in
blood platelets. The 5HT in platelets, present in the electron-dense storage granules found in the central core, is released along with ADP, catecholamines, calcium, and platelet
factor 4 during the first phase of the release reaction (1). In
addition to its importance in the platelet-release reaction, the
amine uptake mechanisms and storage of 5HT by the platelets
resemble synaptosomes and thus serve as models for studies
of neurotransmission (2). For all of these investigations a fast,
sensitive, and reproducible assay for platelet serotonin is required.
Many methods have been reported for the chemical assay
of 5HT in blood platelets (3-10). The most commonly used
assay procedures for 5HT in biological tissues involve its extraction and spectrophotofluorometric
quantitation
(3).
Several investigators have adapted this technique to measurement of 5HT in platelets (3-7, 9, 10). The instability of
5HT under various assay conditions has resulted in problems
with reproducibility
and accuracy. It is essential that any
routine procedure for 5HT in platelets be very sensitive, so
that small samples of blood can be used.
Here we describe a method for 5HT in platelets in which
ascorbic acid is used to stabilize the 5HT and ethanol to enhance the fluorescence
of the compound
during
assay.
Serotonin was obtained from Sigma Chemical Co., St. Louis,
MO; 63178; Isoton II (azide-free balanced electrolyte solution)
from Curtin-Matheson
Scientific, Inc.; ascorbic acid from
Eastman Kodak Co., Rochester, NY 14650; and ethanol (dehydrated alcohol, reagent grade) from U.S. Industrial
Chemicals Co., New York City.
Procedure:
Draw venous blood into standard 10-mL lavender-top glass Vacutainer
Tubes (Becton-Dickinson,
Rutherford, NJ 07070). Obtain platelet-rich plasma (PRP)
by centrifugation at 250 X g for 10 mm. Collect the PRP with
a Pasteur pipet and keep it shaking gently in a plastic tube.
Mix 5 zL of the PRP with 20 mL of Isoton II and obtain the
platelet count, using the following settings of the counter:
lower threshold 5, upper threshold 50, aperture 70 sm, 1/
amplifications 1/2, 1/aperture current 1/2. Ascertain instrumental accuracy by using human platelet reference controls
(“Platelet Chex”; Streck Labs, Inc., P.O. Box 6036, Omaha,
NE 68106). For determination of platelet serotonin content,
pipet 1-mL samples of PRP into plastic culture tubes (no.
2049; Falcon, Oxnard, CA 93030) and centrifuge at 2000 X g
and 4 #{176}C
for 15 mm. Remove the supernatant fluid and wash
the pellet once with Isoton H. The washed pellet can be stored
at -20 #{176}C
until analysis. Pipet 1 mL of a 5 g/L solution of
ascorbic acid into the tubes containing the platelet pellets. Use
three tubes with 1-mL solutions of 5HT (0.2 to 1 zg) in the
ascorbic acid solution as standards, and use tubes with 1 mL
of ascorbic acid solution as blanks. Freeze all tubes in a solid
COs/acetone bath. Sonicate each tube, using the freshly
cleaned probe, for lOs at a setting of 4 on the Disruptor. Chill
the sample until the mixture is just frozen, but not frozen into
a hard solid mass, and then sonicate. Repeat this operation
once more. Add 0.5 mL of the sonicated sample to 1.5 mL of
ethanol with vigorous vortex-mixing. Centrifuge at 2500 X g
for 20 mm. Carefully transfer the supernate into a clean test
tube. Pipet 1 mL of it into a glass test tube, add 0.6 mL of
concentrated HC1, and mix. Measure the fluorescence within
5 mm, with activation wavelength set at 295 nm and emission
at 540 nm. Calculate the amount of 5HT, in micrograms per
io platelets, by using the standard curve and the platelet
counts.
About 12 to 15 samples in duplicate can be processed in a
routine working day. All samples can be processed simultaneously up to the point of HC1 addition, which is done to only
two or three tubes at a time.
Materials and Methods
Results
We counted platelets in diluted aliquots of platelet-rich
plasma with an electronic counter (Model ZBI; Coulter
Electronics, Inc., Hialeah, FL 33010). Fluorometry was done
with an Aminco-Bowman
spectrophotofluorometer. Platelets
were sonicated with the Sonified Cell Disruptor, Model W185,
manufactured by Heat Systems-Ultrasonics, Inc., Plainview,
Long Island, NY.
The final fluorescence of 5HT in the concentrated HCI
medium would be expected to depend upon the other components in the medium. The effect of various aliphatic alcohols on the fluorescence was studied. We mixed 1 mL of a 1
mg/L solution of 5HT with 1 mL of alcohol. From the resulting
mixture, 1 mL was pipetted into glass tubes, mixed with 0.6
mL of concentrated HCI, and the fluorescence measured. The
results (Table 1) show that the fluorescence is most intense
with ethanol in the medium; the ethanol also serves to precipitate the proteins.
We studied the effect of ethanol concentration upon the
final fluorescence by measuring the fluorescent intensity of
the same amount of 5HT, keeping the concentration of concentrated HCI constant and varying the ethanol concentra-
1 Neurology and Geriatric Research Education and Clinical Center,
Veterans Administration Medical Center, 54th St. and 48th Ave.
South, Minneapolis, MN 55417.
2 Departments
of Neurology and Biochemistry, University of
Minnesota, Minneapolis, MN 55455.
Received Sept. 18, 1980; accepted Nov. 17, 1980.
328
CLINICAL CHEMISTRY, Vol. 27, No. 2, 1981
Table 1. Effect of Methanol, Ethanol, and
Isopropanol on Fluorescence of Serotonin
Table 4. Fluorescence of 5-Hydroxyindole
Compounds In the Final Assay Procedure
Fluor.scence
(arb. units)
Alcohol
Fluorascenca
(assuming 5HT to ho 100)
None
360
5-Hydroxytryptamine (5HT)
Methanol
650
DL-5-Hydroxytryptophan
Ethanol
820
5-Hydroxyindole-3-acetic
Isopropanol
600
Sarotonin fluorescence was measwed k the presence of the three aliphatic
alcohols. Per liter, the final solution contained 310 mL of the respective alcohol.
380 ni of concd.HCI. and 0.31 mg of 5HT.
Identical amounts of the tfree compounds wereassayedby the reccnsnended
procedue. The final assay mixtite contained 0.125 g of eachcompound.LTryptophan and lndole-3-acetlc acid gave no fluorescence under the same
conditions.
Table 2. Effect of Ethanol Concentration on
Fluorescence of Serotonin
Table 5. Serotonln Content of Platelets from 15
Healthy Men
100
70
64
acid
S.rotonln,
Ethanol concn
solution,
Ag.,
Fluonesc.nc.
In final
mL/L
(arb. units)
63
380
94
460
156
219
313
375
438
670
750
865
840
880
890
469
900
271
Sarotonin fluorescence was measwed in the presence of various amounts
of ethenol. Per llt.r, the final solution contaIned 380 ml of concd. HCI and 0.31
mg of 5HT.
the
solution, MO
0.03 1
0.063
0.125
0.250
0.50
1.0
i09
plat.l.ts
31
32
32
35
0.72
0.66
0.62
0.83
37
0.68
38
42
42
43
0.63
0.79
0.77
0.63
44
0.44
46
0.75
51
0.64
57
0.75
0.66
0.43
Mean 0.67 (SE 0.03)
The averageof duplicate analysesis reported.
In
final
MO P
57
64
Table 3. Fluorescence of Various Final Solutions
of Serotonin
S.rotonln
yr
Fluorescence
(s*.
units)
72
140
282
571
1092
2200
Vedousamountsof serotonin weresttjected to the recomended procexe.
The fInal solution contained, per liter. 470 mL of ethanoland 380 mL of concd.
tion. The results (Table 2) show that the fluorescence is
maximal
when the ethanol concentration is 380 mL/L, beyond
which it remains constant. The final ethanol concentration
in the finally adopted procedure is 470 mL/L.
In an experiment similar to the one above, to determine the
optimal amount of ascorbic acid, we found that the fluorescence remained stable when the ascorbic acid concentration
in the final mixture exceeded 0.5 g/L. In the finally adopted
procedure, the ascorbic acid concentration in the final mixture
is 1.6 g/L.
When various amounts of standard 5 HT were analyzed by
the recommended procedure, the measured fluorescence was
linearly related to concentration
(Table 3). Essentially identical results were obtained if the standard solutions of sero-
tonin were analyzed without being subjected to freezing and
thawing
and sonication. Thus there is no loss of serotonin
during the analysis, but the standard solutions nevertheless
should be freshly prepared. Ordinarily
the platelet samples
are washed with a solution of Isoton II; results were identical
for samples washed either once or twice.
The reproducibility
of the technique was determined
by
repeated analysis of platelet samples taken from the same
individuals
at the same time. With five different
platelet
samples an average coefficient
of variation of 4.1% was obtained.
The analytical recovery was also tested. An average of 95%
of three concentrations of added standards (in the range 0.125
to 0.5 tg 5HT) could be accounted for.
Structurally
similar compounds,
such as 5-hydroxytryptophan and 5-hydroxyindole-3-acetic
acid will interfere positively (Table 4). Compounds
lacking the 5-hydroxyindole
structure will not interfere.
Blood was collected from normal men, ages 31 to 65 years,
and the platelet 5HT estimated
(Table 5). In Table 6 we
compare the mean platelet 5HT content as measured by our
technique and as reported by other investigators.
Discussion
The protective effect of ascorbic acid against oxidative loss
and Ling (14), but it has
agent in analyses for 5HT.
of 5HT was first reported
by Blum
not been widely used ass protective
Guilbault
and Froehlich (15) used
hydrochloric acids in the final assay
a mixture of ascorbic and
mixture and reported that
in the presence of ascorbic acid the fluorescence intensity
remained
constant.
The ascorbic
acid-from
the initial
step
of extraction of 5HT from platelets to the final measurement
CLINICAL CHEMISTRY, Vol. 27, No. 2, 1981
329
Table 6. Human Platelet Serotonin Content as
Reported
by Various Investigators
Platelet serotonin,
Ref. no.
Mg1109 platelets
Biological
(11)
0.57
(12)
0.76
Spectrophotofluorometric
(13)
(4)
0.87
0.65
(9)
0.94
Present report
0.67
of fluorescence-serves
to stabilize 5HT.
Most spectrophotofluorometric
methods for 5HT are based
on the technique originally proposed by Udenfriend et al. (16).
The specificity of the technique was considerably
enhanced
by measuring 5HT fluorescence in very acidic solutions (3).
Later it was reported by Das (6) that ethanol, used in the
medium, enhances the 5HT fluorescence, and this observation
has been incorporated
into our procedure.
As expected, some metabolically
important
5-hydroxyindole compounds interfere with the proposed assay (Table 4),
but under normal conditions
the contribution
of such metabolites to a serotonin determination
would be less than 6%
(/7). Therefore the proposed method can be used for routine
assays of 5HT in platelets. If the study requires determination
of both 5HT and compounds metabolically
related to it, the
mixture will have to be separated by chromatography
and the
resulting components assayed individually.
These investigations were supported by Research Funds from the
Veterans Administration. We thank Mr. Mike Stanton and his coworkers from DeVac, Minneapolis, for providing blood samples for
the study, and Dr. G. J. Johnson for many discussions.
Barnhart, M. I., Platelet responses in health and disease. Mol. Cell.
Biochem.
330
22,113-137
taining neurons. Prog. Neurobiol.
1, 153-198 (1973).
3. Lovenberg, W., and Engelman, K., Assay of serotonin, related
metabolites, and enzymes. In Methods of Biochemical Analysis,
Supplementary
Volume on Analysis
of Biogenic Amines and Their
Related
Enzymes,
D, Glick, Ed., John Wiley and Sons, New York,
NY, 1971, pp 1-34.
(1978).
CLINICAL CHEMISTRY. Vol. 27. No. 2, 1981
P. E., An easy method to determine the
J. Clin. Pathol. 15, 191-193
4. Crosti, P. F., and Luchelli,
serotonin content of human
(1962).
5. Yuwiler, A., Plotkin,
platelets.
S., Geller, E., and Ritvo, E. R., A rapid, ac-
curate procedure for the determination
of serotonin
blood. Biochem. Med. 3, 426-43 1 (1970).
in whole human
6. Das, M. L., A rapid sensitive method for direct estimation
rotonin in whole blood. Biochem. Med. 6, 299-309 (1972).
of se-
7. Drummond, A. H., and Gordon, J. L., Rapid, sensitive microassay
for platelet 5HT. Thromb. Diath. Haemorrh.
31, 366-367 (1974).
8. O’Brien, R. A., and Spector, S., Determination of platelet serotonin
by radioimmunoassay
and comparison with the spectrophotofluorometric method. Anal. Biochem. 67, 336-338 (1975).
A. A., and Witkop, C. J.,
An improved method for the extraction of endogenous platelet serotonin. J. Lab. Clin. Med. 87, 129-137 (1976).
10. Goldman, J., and Thibert, R. J., Serotonin in blood platelets:
Studies on a simple assay. Microchem. J. 22, 85-91 (1977).
11. Hardisty, R. M., and Stacey, R. S., 5-Hydroxytryptamine
in
normal human platelets. J. Physiol.
(London)
130,711-720(1955).
12. Zucker, M. B., and Borelli, J., Quantity, assay and release of Serotonin in human platelets. J. AppI. Physiol. 7,425-431 (1955).
13. Weiner, M., and Udenfriend, S., Relationship of platelet serotonin
9. Rao, G. H. R., White, J. G., Jachimowicz,
to disturbances
(1957).
of clotting
and hemostasis.
Circulation
15, 353-357
14. Blum, J. J., and Ling,
N. S., Oxidation
of serotonin
and 5during the denaturation
of oxyhemoglobin.
Biochem.
(1959).
15. Guilbault, G., and Froehlich, P. M., New assay for tryptophan and
its 5-hydroxyindole metabolites in blood. Clin. Chem. 20, 812-815
(1974).
hydroxyindoles
J. 73, 530-535
16. Udenfriend,
S., Weissbach,
of 5-hydroxytryptamine
Chem.
References
I.
2. Sneddon, J. M., Blood platelets as a model for monoamine-con-
215, 337-344
H., and Clark, C. T., The estimation
(serotonin)
in biological tissues. J. Biol.
(1955).
17. Rao, G. H. R, Friedland, M., Gerrard, J. M., and White, J. G., The
influence of metabolites on the assay of platelet serotonin. Thromb.
Res. 10,791-802 (1977).