- Email: [email protected]
Enzyme immunoassay of thromboxane B2
322
Bwchimtca
et Biophysics
Acta, 750 (19X3) 322-329 Elsevier
Biomedical
Press
BBA 51315
ENZYME IMMUNOASSAY
OF THROMBO~NE
B2
YOKO HAYASHI ‘, NATSUO UEDA ‘, KAZUSHIGE YOKOTA YOSHIKO YAMAMOTO ‘, SHOZO YAMAMOTO ‘,*, KANZEN MIYAZAKI ‘, KANEYOSHI KATO ’ and SHINJI TERAO ’ ” Deparlment
of Biochemistry
and ’ Clinical
Tokashi~~a, ’ Research Laboratories, Chemical
(Received
Kq
words
Industries,
September
Laboratory
Technicians’
P~ur~?~aceatical Diuwon,
‘, SUM10 KAWAMURA ‘, FUMITAKA NAKAMURA ‘. KOUWA YAMASHITA
School,
Nippon
Tokushima
Kqaku
Uniuersity,
OGUSHI “, “, HIROSHl
School OJ Medicme,
Co., Tokvo and ‘t Chemical
Kurumoto-cho.
Research Laborutories,
Takeda
Osaka ~~apan~
7th, 1982)
fmmunoassq:
Thromboxane
Bz: Pros~a~la~din
~le~abolite~ Platelet
qgregation
An immunoassay for thromboxane Bt was developed in which the hapten molecule was labeled with P-galactosidase. The immunoprecipitate formed after competition between enzyme-labeled and unlabeled thromboxane B, was subjected to a fluorometric assay of j3-galactosidase. Thromboxane B, was detectable in the range of 0.1-30 pmol. Both enzyme immunoassay and radioimmunoassay showed essentially the same cross-reactivities with other prostagl~dins and their meta~lites when the same antibody was used. Known amounts of thromboxane Br were added to human plasma, and the sample was applied to an octadecyl silica column. The extract was analyzed by enzyme immunoassay to examine the correlation between the added (x) and measured ( y) thromboxane B, ( y = 1.09.x + 11.07 pmol/ml, r = 0.99). A satisfactory correlation was observed between radioimmunoassay (x) and enzyme immunoassay ( y) ( y = 0.92x + 4.64 pmol/ml, r = O.%). The validity of enzyme immunoassay was also confirmed by gas chromatography-mass spectrometry of a dimethylisopropy~silyl ether derivative of throm~xane B methyl ester. The method was applicable to the assay of thromboxane BZ produced from endogenous precursor during thrombin-induced aggregation of human platelets.
quire an authentic radioactive compound of high specific activity, which is essential for radioimmunoassay and is not readily available in many cases. The new immunoassay has been utilized in the assays of hormones and drugs [2-41. Previously we reported an application of this technique to the measurement of prostag~andin FZa [5,6]. This paper describes an enzyme immunoassay for thromboxane B,, which is a stable degradation product of unstable thromboxane A,.
Investigations on the physiological and pathological role of pro-aggregatory and vasoconstrictive thromboxane A, require a sensitive, specific and handy assay method. Radioimmunoassay has been widely employed for the determination of arachidonate matebolites [l]. In view of the environmental pollution posed by the use of radioacenzyme immunoassay, which tive compound, utilizes an enzyme in place of radioisotope as a label, is a suitable non-isotopic assay method. Furthermore, enzyme immunoassay does not re* To whom correspondence OOOS-2740/83/0~0-0000/~03.00
Materials and Methods c?lt?micuis Thromboxane B,, prostaglandins F,,, FZnr E,, Ez and D,, 6-ketoprostaglandin F,,, 15-ketopros-
should be addressed. G 19X3 Elsevier Biomedical
Press
323
l-ketotetrataglandin Fzoi, 5a,7a-dihydroxy-1 norprosta-1,16-dioic acid, 13,14-dihydro-15ketoprostaglandin F,, and 13, lCdihydro- 15-ketoprostaglandin E, were kindly provided by Dr. M. Hayashi of Ono Pharmaceutical Company, Research Institute. [5,6,8,9,1 1,12,14,15-3H]Thromboxane B, (150 Ci/mmol) was purchased from Amersham International (Amersham). Synthesis of [3,3,4,4-2H,]thromboxane B, (m.p., 87-88°C) was started by catalytic deuteration of l-acetoxy5-methoxy-3-pentyne *. iso-Butyl chloroformate was obtained from Sigma (St. Louis), tri-nbutylamine from Wako Pure Chemical Industries (Osaka), 4-methylumbelliferyl-P_D-galactoside and 4-methylumbelliferone from Nakarai Chemicals (Kyoto), thrombin from Midorijuji (Osaka), and indomethacin from Sigma (St. Louis). Dimethylisopropylsilyl imidazole was purchased from Tokyo Kasei (Tokyo). /?-Galactosidase from Escherichia coli, with a specific activity of 30 pmol/min per mg protein at 25°C with lactose as substrate, was supplied from Boehringer (Mannheim), rabbit serum from Nakarai Chemicals (Kyoto), goat anti-rabbit IgG antiserum from Miles Laboratories (Elkhart), and bovine albumin (fatty acid-free) from Sigma (St. Louis). Octadecyl silica columns (Sep Pak, 10 X 10 mm) were purchased from Waters Associates (Milford), Sephadex LH-20 from Pharmacia (Uppsala), and silica gel for column chromatography from Merck (Darmstadt). All other chemicals were of reagent grade. Preparation of antibody to thromboxane B2 Thromboxane B, was conjugated with bovine serum albumin by the mixed anhydride reaction using iso-butyl chloroformate [9]. A male New Zealand white rabbit weighing about 2 kg received 1 ml of the conjugate (1 mg protein) emulsified with an equal volume of Freund’s complete ad-
* I-Acetoxy-S-methoxy-3-pentyne, synthetically readily available, was converted to l-[3,3,4,4-‘H,]acetoxy-5-methoxypentane by catalytic deuteration in the presence of tris(triphenylphosphine)chlororhodium in benzene. Hydrolysis and Jones’ oxidation gave 5-(3,3,4,4-*H,]methoxypentanoic acid which was converted to S-bromopentanoic acid by hydrogen bromide. Its conversion to a phosphonium salt (m.p. 207-208’C) followed by Wittig reaction with the lactol [7] was carried out according to the standard method for thromboxane B, synthesis [8].
juvant. Immunization was performed every second week for about 3 months, in total seven times, and antibody was titrated by radioimmunoassay. Conjugation of thromboxane B, to /3-galactosidase The carboxyl group of thromboxane B, was linked covalently to an amino group of /3-galactosidase by the mixed anhydride reaction as described previously for prostaglandin F,, [5,6]. Thromboxane B, (3.2 mg) was mixed with [3H]thromboxane B, (25 ng, 10 PCi), and the mixture was conjugated to /%galactosidase (1 mg of protein). Based on the molecular weight of enzyme (5 18 000 [lo]) and the specific radioactivity of [‘Hlthromboxane B, (694000 cpm/pmol), the amount of thromboxane B, bound per mol of enzyme was estimated. The enzyme-labeled thromboxane B, (160 fmol) included in the standard immunoassay mixture contained a negligible amount of radioactivity. The conjugate was dissolved in buffer A (10 mM sodium phosphate, pH 7.0, containing 0.1 M NaCl, 1 mM MgCl,, 0.1% NaN, and 0.1% ovalbumin), and stored at 4°C at an enzyme concentration of 10 pgg/ml. Enzyme immunoassay Enzyme-labeled thromboxane B,, rabbit antithromboxane B, antiserum (first antibody) and non-immunized rabbit serum (carrier of first antibody) were diluted with buffer A, respectively. Thromboxane B, and goat anti-rabbit immunoglobulin G antiserum (second antibody) were diluted with buffer B (ovalbumin omitted from buffer A). Reaction of the first antibody and the antigen was performed by the sequential saturation technique [ll]. Namely, thromboxane B, antiserum diluted 106-fold (50 ~1) was mixed with a solution (50 ~1) of authentic thromboxane B, or sample to be tested. After incubation for 5 min at 4°C enzyme-labeled thromboxane B, (50 ~1, equivalent to 160 fmol of thromboxane B, and 1 ng of P-galactosidase) was added, and further incubation was carried out for 2 h at 26°C. Then, nonimmunized rabbit serum diluted 200-fold (50 ~1) and the second antiserum diluted 20-fold (50 ~1) were added. The incubation mixture was kept at 4°C for more than 12 h and centrifuged at 1200 x g for 15 min. The precipitate was washed with 1 ml of buffer B by centrifugation and suspended in
324
0.3 ml of the 0.1 mM substrate solution using a vortex mixer. A stock solution of 28 mM 4-methylumbelliferyl-P-D-galactoside in N, N’-dimethyl formamide was diluted with buffer A to a concentration of 0.1 mM. After incubation at 30°C for 1 h, the enzyme reaction was stopped by the addition of 2.5 ml of 0.1 M glycine/NaOH buffer, pH 10.3. Fluorescence intensity of 4-methylumbelliferone released was measured by a Hitachi spectrofluorometer model MPF 2A. A 1 PM solution of quinine sulfate in 0.05 M H,SO, was employed as a standard which was more stable than 4-methylumbelliferone. In fluorescence intensity 1 PM quinine sulfate was equivalent to 1.54 PM 4-methylumbelliferone. The wavelength for excitation was 360 nm and that for emission was 450 nm. 1 unit of the /3-galactosidase activity was defined as that amount of enzyme which released 1 pmol of 4methylumbelliferone per min under the conditions described above. Radioimmunoassay
[ 3H]Thromboxane B, (15 000 cpm, 0.17 pmol) was mixed with unlabeled authentic thromboxane B, or the extract prepared from biological sample as described below, and incubated with rabbit anti-thromboxane B, antiserum (5 OOO-fold final dilution) in a total volume of 0.3 ml. After l-h incubation at 37°C free ligands were removed by adsorption to dextran-coated charcoal, and thromboxane B, bound to the antibody was counted for radioactivity [ 121. 1 ml of human plasma was subjected to the extraction procedure described below. Thromboxane B, (13.5 pmol) was added to the extract, which was divided into ten portions each for radioimmunoassay. An intraassay coefficient of variation was estimated to be 10.9%. Extraction
of thromboxane
B, from blood plasma
Known amounts of thromboxane B, (O-540 pmol) were added to 1 ml of human plasma obtained by centrifugation of whole blood at 1200 X g for 20 min. Extraction of thromboxane B, was performed by the method of Powell [ 131 modified as below. Plasma was acidified to pH 3.2 with 0.2 N HCl and applied to an octadecyl silica column. After washing the column with 20 ml of water, polar lipids and fatty acids were eluted with 15%
ethanol (20 ml) and petroleum ether (20 ml), respectively. Then, thromboxane B, together with various prostaglandins was eluted with ethylacetate (10 ml) instead of methyl formate as originally described by Powell. The solvent was evaporated, and the dried material was dissolved in buffer B. Gus chromatography-mass
spectrometry
Thromboxane B, was analyzed as a methyl ester-dimethylisopropylsilyl ether derivative. The method of derivatization and gas chromatography-mass spectrometry previously described [ 14, 151 was modified as follows. Human plasma was treated with an octadecyl silica column as described above. To the extract obtained from 1 ml of human plasma were added various amounts of thromboxane B, (2-16 ng) together with 20 ng of [3,3’,4,4’-*H,]thromboxane B, as an internal standard. Ether solution of diazomethane was added. The mixture was allowed to stand at room temperature for 30 min. The resulting methyl ester of thromboxane B, was applied to a silica gel column, which was washed with 6 ml of hexane/ethyl acetate (60: 40, v/v), followed by elution with 25 ml of ethyl acetate/methanol (99 : 1, v/v). The solvent was evaporated under reduced pressure, and the dried residue was treated with 100 ~1 of a mixture of pyridine/dimethylisopropylsilyl imidazole (1 : 1, v/v) at room temperature for 1 h. The reaction mixture was applied to a Sephadex LH-20 column equilibrated with chloroform/hexane/methanol (10 : 10 : 1, v/v) to remove excess reagent. The eluate (2.5 ml) was collected, and the solvent was evaporated. The dried material was dissolved in 200 ~1 of hexane, and a 4-~1 aliquot of the solution was subjected to gas chromatography-mass spectrometry. An LKB 209 1 gas chromatograph-mass spectrometer was employed, which was equipped with a thermostable open tubular glass capillary column coated with SE-30 (25 m x 0.35 mm internal diameter). The flow rate of carrier gas (helium) was 1.4 ml/mm. The temperature of injection port was 320°C and that of the column was 285°C. Separator was kept at 300°C. Ionization energy and trap current were 70 eV and 80 PA, respectively. Accelerating voltage was 2.7 kV. Ions of m/z 641 (M - 43, -iC3H7)+ from thromboxane B, and 645 from
325
deuterated variant were monitored for quantification of thromboxane B,. The calibration curve of thromboxane B, was linear in the range of 5-43 pmol per ml of human plasma. Platelet aggregation Blood was collected from cubital vein of a healthy male donor and mixed with 0.1 vol. of 77 mM EDTA at pH 7.4. The mixture was centrifuged at 200 X g for 10 min at 20°C. The upper layer was removed and further centrifuged at 1200 x g for 30 min. After washing the resulting precipitate with 0.14 M NaCl, platelets were suspended in 50 mM Tris-HCl at pH 7.4 containing 0.90 M NaCl (5.5 3 10’ cells/ml), and were stored at room temperature during the experiment. Thrombin (4 units in 40 ~1 of 0.14 M NaCl) was added to 1 ml of the platelet suspension which had been preincubated at 37°C for 2 min, and stirred in an aggregometer cell. Aggregation was monitored by an Evans Electroselenium aggregometer model 169, and the increase in light transmission was followed. Aliquots (60 ~1) were withdrawn from the cuvette at intervals, and mixed with 2 ml of chloroform/methanol/O.02 N HCl (2/l/0.06). The mixture was evaporated to dryness. The residue was dissolved in 150 ~1 (for samples withdrawn before thrombin addition) or 2 ml (for samples withdrawn during aggregation) of buffer B, and directly subjected to enzyme immunoassay. Results and Discussion Properties of enzyme-labeled thromboxane B, After conjugation of thromboxane B, and /3galactosidase, 68% of enzyme protein and 20% of enzyme activity were recovered, resulting in a 71% decrease in specific activity. Approximately 83 mol of thromboxane B, were bound per mol of /3galactosidase. Since 95- 111 mol of lysine residue are present per mol of E. coli P-galactosidase [16], part of the 83 mol of thromboxane B, may be bound non-covalently to enzyme protein. The K, value for 4-methylumbelliferyl-P_D-galactoside was 0.11 mM with native enzyme and 0.089 mM with the conjugated enzyme. Maximum velocities were 720 and 209 units/mg protein, respectively, under the standard assay conditions. Reaction of the conjugated enzyme proceeded linearly for at least
Thromboxane
B2 ( pmol)
Fig. 1. Comparison of antigenecity between enzyme-labeled and unlabeled thromboxane B, determined by radioimmunoassay. Either enzyme-labeled thromboxane B, (0) or unlabeled thromboxane B, (0) was incubated competitively with [3H]thromboxane B, in the presence of rabbit anti-thromboxane B, antiserum. The abscissa indicates the amount of added thromboxane B, in the enzyme-bound form or in the free form. The amount of enzyme-labeled thromboxane B, was calculated on the basis of the specific radioactivity as described in Materials and Methods.
5 h, and the reaction rate was dependent on the enzyme amount in the range of 0.1-2.5 ng of enzyme protein. The conjugate could be stored for more than a year without appreciable loss of enzyme activity and antigenicity. The antigenicity of enzyme-labeled thromboxane B, was examined by radioimmunoassay in comparison with unlabeled thromboxane B,. The amount of enzyme-labeled thromboxane B, which was required to displace 50% of the bound [ 3H]thromboxane B,, was about 10 times that of unlabeled thromboxane B, (Fig. 1). It is unknown whether the antigenicity of all the enzyme-labeled antigen molecules is uniformly reduced or only part of them is immunoreactive. Calibration curve The enzyme-labeled thromboxane B, was allowed to react with thromboxane B, antiserum. Separation of ‘bound’ and ‘free’ antigen was performed by the double antibody method. Fig. 2 shows dilution curves of the first antibody in the presence of various amounts of the second antiserum (Fig. 2A) and the carrier serum (Fig. 2B).
326
8(
(A)
!5
0
4(
0
Y
0 0
.
\\ Time(h)
I
I
I.
Al
(B)
Fig. 3. Time courses of the first and second immunoreactions. (A) Enzyme-labeled thromboxane B, (160 fmol as thromboxane B, and 1 ng as enzyme) was incubated with the first antibody (5.10”-fold final dilution) at 26’C for the time period as indicated in the abscissa. The second immunoreaction was performed at 4’C overnight. (B) After the first immunoreaction at 26°C for 2 h the second immunoreaction was carried out for the incubation time varied as indicated.
As the standard assay conditions, the first incubation was performed for 2 h at 26°C and the second incubation for more than 12 h at 4°C. The broken line in Fig. 4 shows a standard curve of thromb-
(
IO5
IO6
Anti- thromboxane
IO’
108
109
B2 dilution
Fig. 2. Dilution curves of antiserum for thromboxane B,. Enzyme-labeled thromboxane B, (160 fmol as thromboxane B, and 1 ng as enzyme) was incubated with the first antibody at various final dilutions as indicated. Enzyme immunoassay was performed under the standard conditions except that the second antibody and the carrier serum were added as indicated. (A) The second antibody was present at a final dilution of 1 : 50 (0). 1 : 100 (0) or 1 : 200 (A). Non-immunized rabbit serum as a carrier was maintained at a 1: 1000 dilution. (B) Non-immunized rabbit serum was added at a final dilution of 1 :500 (0) 1 : 1000 (0), 1 : 2500 (A) or absent (A). The second antibody was kept at a lOO-fold final dilution.
0'
0.01
I
I
0.1
1 Prostaglondin
The presence of the second antiserum at lOO-fold final dilution enhanced the immunoprecipitation of P-galactosidase. The addition of carrier serum at lOOO-fold final dilution also stimulated precipitation of the enzyme. The first antiserum at 5 . 106-fold final dilution precipitated SS-60% of the precipitable enzyme. When the time course of immunoreaction was followed, the first immunoreaction at 26°C reached a maximum after 2 h (Fig. 3A) and more than 12 h were necessary to complete the second reaction at 4°C (Fig. 3B).
.‘Y+. 10
100
(p.mol)
Fig. 4. Standard curve for thromboxane B, and cross-reactivities with other arachidonate metabolites. Closed circles and broken line show a standard curve of thromboxane B,. The cross-reactivity of anti-thromboxane B, antiserum was examined under the standard conditions by incubation of enzyme-labeled thromboxane B, with other arachidonate metabolites at concentrations as indicated in abscissa: prostaglandin D, (0, 2) prostaglandin E, (A, 4). prostaglandin E, (v, 5) prostaglandin F,, (0, 6). prostaglandin F,, (0, 3) 6-ketoprostaglandin F,, (W, 7) 1Sketoprostaglandin Faa (*, l), 13,14-dihydro-15-ketoprostaglandin E, (A, 9). 13,14-dihydro15-ketoprostaglandin F,, (v lo), 5a.7a-dihydroxy-I I-ketotetranorprosta- 1,16-dioic acid ( x . 8).
321
oxane B, determined by the standard enzyme immunoassay. Authentic thromboxane B, could be measured over the range of 0.1-30 pmol which was equivalent to the range of 90-10% binding of the labeled thromboxane B,. The sensitivity defined as the amount of thromboxane B, to reduce the enzyme precipitation from 100 to 90% was about 0.1 pmol. This value of the minimum detectable amount of thromboxane B, by enzyme immunoassay was compared with the values by radioimmunoassay which have been described previously by other investigators: 0.03 pmol [ 17-191, 0.2 pmol [20], and 0.3 pmol (defined as 15% displacement) [21]. Since the detection limit in radioimmunoassay depends on the assay conditions, such as the quality of antiserum or the specific radioactivity of labeled antigen, the minimum detectable amount of thromboxane B, was determined by radioimmunoassay using the same antiserum that was used for our enzyme immunoassay, and a value of 0.4 pmol was calculated from Fig. 1. The cross-reactivity of antiserum with other arachidonate metabolites were tested (Fig. 4). 15Ketoprostaglandin FZo showed a cross-reactivity of 2.8%. Other prostaglandins and their metabolites tested showed less than 1% cross-reaction, Inhibition of the binding of enzyme-labeled thromboxane B, (160 fmol) caused by 100 pmol of each compound was as follows: prostaglandin FZa, 40%; prostaglandin D,, 33%; prostaglandin E,, 32%; prostaglandin E,, 26%; 6-ketoprostaglandin F,,, 18%; prostaglandin F,,, 15%; 5a,7a-dihydroxy11 -ketotetranorprosta-1,16-dioic acid (main urinary metabolite of prostaglandin FZu), 10%; 13,14-dihydro-15-ketoprostaglandin E, (main plasma metabolite of prostaglandin E2), 7%; 13,14-dihydro15-ketoprostaglandin FZu (main plasma metabolite of prostaglandin F,,), 0%. Essentially the same cross-reactivity values were obtained by radioimmunoassay using the same antiserum. Enzyme immunoassay of thromboxane B2 in human plasma For extraction of thromboxane B, from human plasma and its separation from other plasma components, an octadecyl silica column was utilized as described in Materials and Methods. Recovery of
I
0
t-0
.
’
3
0
1
Added
L
I
I
10
100
1000
TXBz
I pmol /ml)
Fig. 5. Correlation between added and measured thromboxane (TX) B,. Known amounts of thromboxane B, were added to human plasma. After application of the plasma sample to an octadecyl silica column as described in Materials and Methods, the dried material from the ethyl acetate extract was dispersed in buffer B so that a 50-1.11aliquot could contain O-10 pmol of thromboxane B,, and then analyzed by standard enzyme immunoassay. The amounts of added (x) and measured (_v) thromboxane B, were plotted. Different symbols indicate the results obtained with different batches of human plasma. r. Correlation coefficient; n, number of samples.
thromboxane B, at this step was examined with 13 samples of plasma (1 ml) to which [3H]thromboxane B, (5900 cpm, 66 fmol) was added. The average recovery of radioactivity in the ethyl acetate extract was 74.5 k 2.0% (mean f S.D.). The recovery was not significantly affected by increasing the amount of thromboxane B, up to 2.7 nmol.
TABLE
I
CORRELATION BETWEEN SELECTED ING AND ENZYME IMMUNOASSAY
ION MONITOR-
Thromboxane B, (5.4-43.2 pmol) was added to 1 ml of human plasma. Ions of m/z 641 and 645 were monitored as described in Materials and Methods. Data on enzyme immunoassay are mean values of duplicate determinations, and are pmol/ml of plasma. Added thromboxane
5.4 10.8 21.6 43.2
Measured
thromboxane
B,
B, Selected ion monitoring
Enzyme immunoassay
19.4 25.1 35.2 55.4
28. I 35.1 41.2 62.6
y-0.92x
+ 4.64
r-=0.96
e
Time I
10 Radioimmunoassay
I
I 100 (Pmol/ml
1000
)
Fig. 6. Correlation between radioimmunoassay and enzyme immunoassay. The plasma samples containing added thromboxane B, were treated as described in Fig. 5 and subjected to both radioimmunoassay (x) and enzyme immunoassay OJ). r, Correlation coefficient; n, number of samples.
Varying amounts of thromboxane B, were added to human plasma, and the samples were subjected to octadecyl silica column chromatography and enzyme immunoassay. As shown in Fig. 5, a correlation was observed between the amounts of added and measured thromboxane B2 (correlation coefficient, 0.99). Intraassay variation was examined with 1 ml of plasma to which 13.5 pmol of thromboxane B2 were added. Nine assays each in duplicate were run at one time, and a coefficient of variation of intraassay values was 13.1%. The same plasma samples containing 13.5 pm01 of thromboxane B, per ml were assayed at 12 different times each in duplicate, giving a coefficient of variation of 15.3% for interassay values. The results obtained by enzyme immunoassay were compared with those by radioimmunoassay. There was a satisfactory correlation (coefficient, 0.96) between the values of thromboxane B, at higher concentrations determined by the two methods (Fig. 6). For further validation of enzyme immunoassay, the added thromboxane B, in human plasma was analyzed by gas chromatographymass spectrometry using a deuterated compound as an internal standard. Table I shows the results obtained by both selected ion monitoring and
(min)
Fig. 7. Release of thromboxane (TX) B, during platelet aggregation Thrombin (4 units) was added (arrow) to start aggregation of washed human platelets (10’ cells), which was monitored as described in Materials and Methods (solid line). Aliquots were removed at indicated time points and mixed with chloroform/methanol/O.02 N HCI (2 : 1 : 0.06). The solvent was evaporated, and the residue dissolved in buffer B was subjected to enzyme immunoassay (O------O).
enzyme immunoassay. The calibration curve of each assay was essentially linear in the range tested, and the added thromboxane B, was recovered in a good yield in all the samples. However, both assays gave higher values than the amount of added thromboxane B,, presumably due to the contribution of endogenous thromboxane B, present in the plasma samples. Furthermore, enzyme immunoassay showed still higher values than selected ion monitoring. The plots on the ordinate of Fig. 5 may give the level of endogenous thromboxane B, in plasma of four individuals; 13.0, 7.8, 5.4 and 5.1 pmol/ml. These values were higher than the previously reported values as determined by radioimmunoassay: 0.656 f 0.259 [17], 0.189 k 0.070 (aspirin treatment) [22], 0.409 & 0.180 (no aspirin) [22] and 2.9 pmol/ml [ 181. Our higher values may be attributed partly to the absence of any cyclooxygenase inhibitor when blood was collected from the donors. At lower thromboxane B, concentrations our enzyme immunoassay gave higher values, to more or less an extent, than radioimmunoassay (Fig. 6) or mass spectrometry (Table I). Possible involvement of unknown endogenous materials in the overestimation by enzyme immunoassay has
329
not yet been extensively investigated. Therefore, these values of endogenous thromboxane B, estimated by enzyme immunoassay await further investigations. Thromboxane
B2 formation
during platelet aggrega-
tion
Aggregation of human platelets was induced by thrombin, and thromboxane B, formation from endogenous arachidonic acid was determined by enzyme immunoassay (Fig. 7). Thromboxane B, accumulated as the platelet aggregation proceeded. The final level of thromboxane B, was about 126 pmol/lOX platelets. Pretreatment of washed platelets with 40 PM indomethacin caused neither aggregation of platelets nor a rise of thromboxane B, level. Acknowledgements
The authors wish to thank Miss Masako Yamamoto and Miss Hiromi Hada for technical assistance. S.Y. was supported by grants-in-aid for scientific research from the Ministry of Education, Science and Culture of Japan, a grant from the Intractable Diseases Division, Public Health Bureau, Ministry of Health and Welfare of Japan, a grant for cardiovascular diseases (56-C2), Ministry of Health and Welfare of Japan, and grants from the Japanese Foundation of Metabolism and Diseases, the Naito Foundation, the Asahi Scholarship Promotion Fund, the Japan Research Foundation for Clinical Pharmacology, Japan Heart Foundation (Tokio Marine Research Grant), Yamada Science Foundation, Suzuken, Kenzo Memorial Foundation, Takeda Science Foundation and Mishima Kaiun Memorial Foundation.
References E. and Kindahl, H. (1978) in Advances in I Granstrom. Prostaglandin and Thromboxane Research (Frolich. J.C., ed.). Vol. 5. pp. 119-210. Raven Press. New York 2 Wisdom, G.B. (1976) Clin. Chem. 22, 1243- 1255 and Van Weemen. B.K. (1977) Clin. 3 Schuurs, A.H.W.M. Chim. Acta 81. l-40 4 Ishikawa, E. (1979) in Techniques in Metabolic Research (Kornberg. H.L., Metcalfe, J.C.. Northcote, D.H.. Pogson, C.I. and Tipton. K.F., eds.). B217. pp. l-18. EIsevier/North-Holland Scientific Publishers Ltd., Limerick S. (1981) Biochim. 5 Hayashi. Y., Yano. T. and Yamamoto. Biophys. Acta 663, 661-668 6 Yano, T.. Hayashi, Y. and Yamamoto, S. (1981) J. Biothem. 90, 7733777 7 Hernandez, 0. (1978) Tetrahedron Lett.. 219-222 8 Nelson. N.A. and Jackson, P.W. (1976) Tetrahedron Lett. 327553278 9 Erlanger, B.F., Borek, F.. Beiser, SM. and Liebermann. S. (1959) J. Biol. Chem. 234. 1090-1094 10 Sund, H. and Weber. K. (1963) Biochem. Z. 337, 24-34 11 Zettner. A. and Duly, P.E. (1974) Clin. Chem. 20, 5- 14 12 Inagawa, T., Ohki, S.. Sawada, M. and Hirata, F. (1972) Yakugaku Zasshi 92. l187- 1194 13 Powell, W.S. (1980) Prostaglandins 20, 947-957 14 Miyazaki. H.. Ishibashi, M.. Yamashita. K., Nishikawa. Y. and Katori. M. (1981) Biomed. Mass Spectrom. 8. 521-526 15 Harada, Y.. Tanaka, K., Uchida. Y., Ueno, A., Oh-ishi. S.. Yamashita. K., Ishibashi, M., Miyazaki. H. and Katori. M. ( 1982) Prostaglandins 23. 88 l-895 16 Wallenfels. K. and Weil. R. (1972) in The Enzymes (Boyer, P.D.. ed.), Vol. 7, pp. 617-663, Academic Press, New York 17 Tada, M., Kuzuya. T.. Inoue, M., Fukushima. M. and Abe. H. (1980) in Advances in Myocardiology (Tajuddin, M.. Bhatia, B.. Siddiqui, H.H. and Rona, G.. eds.), Vol. 2. pp. 297-405, University Park Press, Baltimore 18 Granstrom. E.. Kindahl. H. and Samuelsson. B. (1976) Anal. Lett. 9, 61 l-627 19 Tai. H.H. and Yuan, B. (1978) Anal. Biochem. 87. 343-349 20 Anhut, H., Bernauer, W. and Peskar. B.A. (1977) Eur. J. Pharmacol. 44, 85-88 21 Fitzpatrick. F.A., Gorman, R.R.. McGuire, J.C., Kelly, R.C.. Wynalda. M.A. and Sun, F.F. (1977) Anal. Biochem. 82, 1-7 22 Viinikka, 759-766
L. and Ylikorkala,
0. (1980) Prostaglandins
20,
Bwchimtca
et Biophysics
Acta, 750 (19X3) 322-329 Elsevier
Biomedical
Press
BBA 51315
ENZYME IMMUNOASSAY
OF THROMBO~NE
B2
YOKO HAYASHI ‘, NATSUO UEDA ‘, KAZUSHIGE YOKOTA YOSHIKO YAMAMOTO ‘, SHOZO YAMAMOTO ‘,*, KANZEN MIYAZAKI ‘, KANEYOSHI KATO ’ and SHINJI TERAO ’ ” Deparlment
of Biochemistry
and ’ Clinical
Tokashi~~a, ’ Research Laboratories, Chemical
(Received
Kq
words
Industries,
September
Laboratory
Technicians’
P~ur~?~aceatical Diuwon,
‘, SUM10 KAWAMURA ‘, FUMITAKA NAKAMURA ‘. KOUWA YAMASHITA
School,
Nippon
Tokushima
Kqaku
Uniuersity,
OGUSHI “, “, HIROSHl
School OJ Medicme,
Co., Tokvo and ‘t Chemical
Kurumoto-cho.
Research Laborutories,
Takeda
Osaka ~~apan~
7th, 1982)
fmmunoassq:
Thromboxane
Bz: Pros~a~la~din
~le~abolite~ Platelet
qgregation
An immunoassay for thromboxane Bt was developed in which the hapten molecule was labeled with P-galactosidase. The immunoprecipitate formed after competition between enzyme-labeled and unlabeled thromboxane B, was subjected to a fluorometric assay of j3-galactosidase. Thromboxane B, was detectable in the range of 0.1-30 pmol. Both enzyme immunoassay and radioimmunoassay showed essentially the same cross-reactivities with other prostagl~dins and their meta~lites when the same antibody was used. Known amounts of thromboxane Br were added to human plasma, and the sample was applied to an octadecyl silica column. The extract was analyzed by enzyme immunoassay to examine the correlation between the added (x) and measured ( y) thromboxane B, ( y = 1.09.x + 11.07 pmol/ml, r = 0.99). A satisfactory correlation was observed between radioimmunoassay (x) and enzyme immunoassay ( y) ( y = 0.92x + 4.64 pmol/ml, r = O.%). The validity of enzyme immunoassay was also confirmed by gas chromatography-mass spectrometry of a dimethylisopropy~silyl ether derivative of throm~xane B methyl ester. The method was applicable to the assay of thromboxane BZ produced from endogenous precursor during thrombin-induced aggregation of human platelets.
quire an authentic radioactive compound of high specific activity, which is essential for radioimmunoassay and is not readily available in many cases. The new immunoassay has been utilized in the assays of hormones and drugs [2-41. Previously we reported an application of this technique to the measurement of prostag~andin FZa [5,6]. This paper describes an enzyme immunoassay for thromboxane B,, which is a stable degradation product of unstable thromboxane A,.
Investigations on the physiological and pathological role of pro-aggregatory and vasoconstrictive thromboxane A, require a sensitive, specific and handy assay method. Radioimmunoassay has been widely employed for the determination of arachidonate matebolites [l]. In view of the environmental pollution posed by the use of radioacenzyme immunoassay, which tive compound, utilizes an enzyme in place of radioisotope as a label, is a suitable non-isotopic assay method. Furthermore, enzyme immunoassay does not re* To whom correspondence OOOS-2740/83/0~0-0000/~03.00
Materials and Methods c?lt?micuis Thromboxane B,, prostaglandins F,,, FZnr E,, Ez and D,, 6-ketoprostaglandin F,,, 15-ketopros-
should be addressed. G 19X3 Elsevier Biomedical
Press
323
l-ketotetrataglandin Fzoi, 5a,7a-dihydroxy-1 norprosta-1,16-dioic acid, 13,14-dihydro-15ketoprostaglandin F,, and 13, lCdihydro- 15-ketoprostaglandin E, were kindly provided by Dr. M. Hayashi of Ono Pharmaceutical Company, Research Institute. [5,6,8,9,1 1,12,14,15-3H]Thromboxane B, (150 Ci/mmol) was purchased from Amersham International (Amersham). Synthesis of [3,3,4,4-2H,]thromboxane B, (m.p., 87-88°C) was started by catalytic deuteration of l-acetoxy5-methoxy-3-pentyne *. iso-Butyl chloroformate was obtained from Sigma (St. Louis), tri-nbutylamine from Wako Pure Chemical Industries (Osaka), 4-methylumbelliferyl-P_D-galactoside and 4-methylumbelliferone from Nakarai Chemicals (Kyoto), thrombin from Midorijuji (Osaka), and indomethacin from Sigma (St. Louis). Dimethylisopropylsilyl imidazole was purchased from Tokyo Kasei (Tokyo). /?-Galactosidase from Escherichia coli, with a specific activity of 30 pmol/min per mg protein at 25°C with lactose as substrate, was supplied from Boehringer (Mannheim), rabbit serum from Nakarai Chemicals (Kyoto), goat anti-rabbit IgG antiserum from Miles Laboratories (Elkhart), and bovine albumin (fatty acid-free) from Sigma (St. Louis). Octadecyl silica columns (Sep Pak, 10 X 10 mm) were purchased from Waters Associates (Milford), Sephadex LH-20 from Pharmacia (Uppsala), and silica gel for column chromatography from Merck (Darmstadt). All other chemicals were of reagent grade. Preparation of antibody to thromboxane B2 Thromboxane B, was conjugated with bovine serum albumin by the mixed anhydride reaction using iso-butyl chloroformate [9]. A male New Zealand white rabbit weighing about 2 kg received 1 ml of the conjugate (1 mg protein) emulsified with an equal volume of Freund’s complete ad-
* I-Acetoxy-S-methoxy-3-pentyne, synthetically readily available, was converted to l-[3,3,4,4-‘H,]acetoxy-5-methoxypentane by catalytic deuteration in the presence of tris(triphenylphosphine)chlororhodium in benzene. Hydrolysis and Jones’ oxidation gave 5-(3,3,4,4-*H,]methoxypentanoic acid which was converted to S-bromopentanoic acid by hydrogen bromide. Its conversion to a phosphonium salt (m.p. 207-208’C) followed by Wittig reaction with the lactol [7] was carried out according to the standard method for thromboxane B, synthesis [8].
juvant. Immunization was performed every second week for about 3 months, in total seven times, and antibody was titrated by radioimmunoassay. Conjugation of thromboxane B, to /3-galactosidase The carboxyl group of thromboxane B, was linked covalently to an amino group of /3-galactosidase by the mixed anhydride reaction as described previously for prostaglandin F,, [5,6]. Thromboxane B, (3.2 mg) was mixed with [3H]thromboxane B, (25 ng, 10 PCi), and the mixture was conjugated to /%galactosidase (1 mg of protein). Based on the molecular weight of enzyme (5 18 000 [lo]) and the specific radioactivity of [‘Hlthromboxane B, (694000 cpm/pmol), the amount of thromboxane B, bound per mol of enzyme was estimated. The enzyme-labeled thromboxane B, (160 fmol) included in the standard immunoassay mixture contained a negligible amount of radioactivity. The conjugate was dissolved in buffer A (10 mM sodium phosphate, pH 7.0, containing 0.1 M NaCl, 1 mM MgCl,, 0.1% NaN, and 0.1% ovalbumin), and stored at 4°C at an enzyme concentration of 10 pgg/ml. Enzyme immunoassay Enzyme-labeled thromboxane B,, rabbit antithromboxane B, antiserum (first antibody) and non-immunized rabbit serum (carrier of first antibody) were diluted with buffer A, respectively. Thromboxane B, and goat anti-rabbit immunoglobulin G antiserum (second antibody) were diluted with buffer B (ovalbumin omitted from buffer A). Reaction of the first antibody and the antigen was performed by the sequential saturation technique [ll]. Namely, thromboxane B, antiserum diluted 106-fold (50 ~1) was mixed with a solution (50 ~1) of authentic thromboxane B, or sample to be tested. After incubation for 5 min at 4°C enzyme-labeled thromboxane B, (50 ~1, equivalent to 160 fmol of thromboxane B, and 1 ng of P-galactosidase) was added, and further incubation was carried out for 2 h at 26°C. Then, nonimmunized rabbit serum diluted 200-fold (50 ~1) and the second antiserum diluted 20-fold (50 ~1) were added. The incubation mixture was kept at 4°C for more than 12 h and centrifuged at 1200 x g for 15 min. The precipitate was washed with 1 ml of buffer B by centrifugation and suspended in
324
0.3 ml of the 0.1 mM substrate solution using a vortex mixer. A stock solution of 28 mM 4-methylumbelliferyl-P-D-galactoside in N, N’-dimethyl formamide was diluted with buffer A to a concentration of 0.1 mM. After incubation at 30°C for 1 h, the enzyme reaction was stopped by the addition of 2.5 ml of 0.1 M glycine/NaOH buffer, pH 10.3. Fluorescence intensity of 4-methylumbelliferone released was measured by a Hitachi spectrofluorometer model MPF 2A. A 1 PM solution of quinine sulfate in 0.05 M H,SO, was employed as a standard which was more stable than 4-methylumbelliferone. In fluorescence intensity 1 PM quinine sulfate was equivalent to 1.54 PM 4-methylumbelliferone. The wavelength for excitation was 360 nm and that for emission was 450 nm. 1 unit of the /3-galactosidase activity was defined as that amount of enzyme which released 1 pmol of 4methylumbelliferone per min under the conditions described above. Radioimmunoassay
[ 3H]Thromboxane B, (15 000 cpm, 0.17 pmol) was mixed with unlabeled authentic thromboxane B, or the extract prepared from biological sample as described below, and incubated with rabbit anti-thromboxane B, antiserum (5 OOO-fold final dilution) in a total volume of 0.3 ml. After l-h incubation at 37°C free ligands were removed by adsorption to dextran-coated charcoal, and thromboxane B, bound to the antibody was counted for radioactivity [ 121. 1 ml of human plasma was subjected to the extraction procedure described below. Thromboxane B, (13.5 pmol) was added to the extract, which was divided into ten portions each for radioimmunoassay. An intraassay coefficient of variation was estimated to be 10.9%. Extraction
of thromboxane
B, from blood plasma
Known amounts of thromboxane B, (O-540 pmol) were added to 1 ml of human plasma obtained by centrifugation of whole blood at 1200 X g for 20 min. Extraction of thromboxane B, was performed by the method of Powell [ 131 modified as below. Plasma was acidified to pH 3.2 with 0.2 N HCl and applied to an octadecyl silica column. After washing the column with 20 ml of water, polar lipids and fatty acids were eluted with 15%
ethanol (20 ml) and petroleum ether (20 ml), respectively. Then, thromboxane B, together with various prostaglandins was eluted with ethylacetate (10 ml) instead of methyl formate as originally described by Powell. The solvent was evaporated, and the dried material was dissolved in buffer B. Gus chromatography-mass
spectrometry
Thromboxane B, was analyzed as a methyl ester-dimethylisopropylsilyl ether derivative. The method of derivatization and gas chromatography-mass spectrometry previously described [ 14, 151 was modified as follows. Human plasma was treated with an octadecyl silica column as described above. To the extract obtained from 1 ml of human plasma were added various amounts of thromboxane B, (2-16 ng) together with 20 ng of [3,3’,4,4’-*H,]thromboxane B, as an internal standard. Ether solution of diazomethane was added. The mixture was allowed to stand at room temperature for 30 min. The resulting methyl ester of thromboxane B, was applied to a silica gel column, which was washed with 6 ml of hexane/ethyl acetate (60: 40, v/v), followed by elution with 25 ml of ethyl acetate/methanol (99 : 1, v/v). The solvent was evaporated under reduced pressure, and the dried residue was treated with 100 ~1 of a mixture of pyridine/dimethylisopropylsilyl imidazole (1 : 1, v/v) at room temperature for 1 h. The reaction mixture was applied to a Sephadex LH-20 column equilibrated with chloroform/hexane/methanol (10 : 10 : 1, v/v) to remove excess reagent. The eluate (2.5 ml) was collected, and the solvent was evaporated. The dried material was dissolved in 200 ~1 of hexane, and a 4-~1 aliquot of the solution was subjected to gas chromatography-mass spectrometry. An LKB 209 1 gas chromatograph-mass spectrometer was employed, which was equipped with a thermostable open tubular glass capillary column coated with SE-30 (25 m x 0.35 mm internal diameter). The flow rate of carrier gas (helium) was 1.4 ml/mm. The temperature of injection port was 320°C and that of the column was 285°C. Separator was kept at 300°C. Ionization energy and trap current were 70 eV and 80 PA, respectively. Accelerating voltage was 2.7 kV. Ions of m/z 641 (M - 43, -iC3H7)+ from thromboxane B, and 645 from
325
deuterated variant were monitored for quantification of thromboxane B,. The calibration curve of thromboxane B, was linear in the range of 5-43 pmol per ml of human plasma. Platelet aggregation Blood was collected from cubital vein of a healthy male donor and mixed with 0.1 vol. of 77 mM EDTA at pH 7.4. The mixture was centrifuged at 200 X g for 10 min at 20°C. The upper layer was removed and further centrifuged at 1200 x g for 30 min. After washing the resulting precipitate with 0.14 M NaCl, platelets were suspended in 50 mM Tris-HCl at pH 7.4 containing 0.90 M NaCl (5.5 3 10’ cells/ml), and were stored at room temperature during the experiment. Thrombin (4 units in 40 ~1 of 0.14 M NaCl) was added to 1 ml of the platelet suspension which had been preincubated at 37°C for 2 min, and stirred in an aggregometer cell. Aggregation was monitored by an Evans Electroselenium aggregometer model 169, and the increase in light transmission was followed. Aliquots (60 ~1) were withdrawn from the cuvette at intervals, and mixed with 2 ml of chloroform/methanol/O.02 N HCl (2/l/0.06). The mixture was evaporated to dryness. The residue was dissolved in 150 ~1 (for samples withdrawn before thrombin addition) or 2 ml (for samples withdrawn during aggregation) of buffer B, and directly subjected to enzyme immunoassay. Results and Discussion Properties of enzyme-labeled thromboxane B, After conjugation of thromboxane B, and /3galactosidase, 68% of enzyme protein and 20% of enzyme activity were recovered, resulting in a 71% decrease in specific activity. Approximately 83 mol of thromboxane B, were bound per mol of /3galactosidase. Since 95- 111 mol of lysine residue are present per mol of E. coli P-galactosidase [16], part of the 83 mol of thromboxane B, may be bound non-covalently to enzyme protein. The K, value for 4-methylumbelliferyl-P_D-galactoside was 0.11 mM with native enzyme and 0.089 mM with the conjugated enzyme. Maximum velocities were 720 and 209 units/mg protein, respectively, under the standard assay conditions. Reaction of the conjugated enzyme proceeded linearly for at least
Thromboxane
B2 ( pmol)
Fig. 1. Comparison of antigenecity between enzyme-labeled and unlabeled thromboxane B, determined by radioimmunoassay. Either enzyme-labeled thromboxane B, (0) or unlabeled thromboxane B, (0) was incubated competitively with [3H]thromboxane B, in the presence of rabbit anti-thromboxane B, antiserum. The abscissa indicates the amount of added thromboxane B, in the enzyme-bound form or in the free form. The amount of enzyme-labeled thromboxane B, was calculated on the basis of the specific radioactivity as described in Materials and Methods.
5 h, and the reaction rate was dependent on the enzyme amount in the range of 0.1-2.5 ng of enzyme protein. The conjugate could be stored for more than a year without appreciable loss of enzyme activity and antigenicity. The antigenicity of enzyme-labeled thromboxane B, was examined by radioimmunoassay in comparison with unlabeled thromboxane B,. The amount of enzyme-labeled thromboxane B, which was required to displace 50% of the bound [ 3H]thromboxane B,, was about 10 times that of unlabeled thromboxane B, (Fig. 1). It is unknown whether the antigenicity of all the enzyme-labeled antigen molecules is uniformly reduced or only part of them is immunoreactive. Calibration curve The enzyme-labeled thromboxane B, was allowed to react with thromboxane B, antiserum. Separation of ‘bound’ and ‘free’ antigen was performed by the double antibody method. Fig. 2 shows dilution curves of the first antibody in the presence of various amounts of the second antiserum (Fig. 2A) and the carrier serum (Fig. 2B).
326
8(
(A)
!5
0
4(
0
Y
0 0
.
\\ Time(h)
I
I
I.
Al
(B)
Fig. 3. Time courses of the first and second immunoreactions. (A) Enzyme-labeled thromboxane B, (160 fmol as thromboxane B, and 1 ng as enzyme) was incubated with the first antibody (5.10”-fold final dilution) at 26’C for the time period as indicated in the abscissa. The second immunoreaction was performed at 4’C overnight. (B) After the first immunoreaction at 26°C for 2 h the second immunoreaction was carried out for the incubation time varied as indicated.
As the standard assay conditions, the first incubation was performed for 2 h at 26°C and the second incubation for more than 12 h at 4°C. The broken line in Fig. 4 shows a standard curve of thromb-
(
IO5
IO6
Anti- thromboxane
IO’
108
109
B2 dilution
Fig. 2. Dilution curves of antiserum for thromboxane B,. Enzyme-labeled thromboxane B, (160 fmol as thromboxane B, and 1 ng as enzyme) was incubated with the first antibody at various final dilutions as indicated. Enzyme immunoassay was performed under the standard conditions except that the second antibody and the carrier serum were added as indicated. (A) The second antibody was present at a final dilution of 1 : 50 (0). 1 : 100 (0) or 1 : 200 (A). Non-immunized rabbit serum as a carrier was maintained at a 1: 1000 dilution. (B) Non-immunized rabbit serum was added at a final dilution of 1 :500 (0) 1 : 1000 (0), 1 : 2500 (A) or absent (A). The second antibody was kept at a lOO-fold final dilution.
0'
0.01
I
I
0.1
1 Prostaglondin
The presence of the second antiserum at lOO-fold final dilution enhanced the immunoprecipitation of P-galactosidase. The addition of carrier serum at lOOO-fold final dilution also stimulated precipitation of the enzyme. The first antiserum at 5 . 106-fold final dilution precipitated SS-60% of the precipitable enzyme. When the time course of immunoreaction was followed, the first immunoreaction at 26°C reached a maximum after 2 h (Fig. 3A) and more than 12 h were necessary to complete the second reaction at 4°C (Fig. 3B).
.‘Y+. 10
100
(p.mol)
Fig. 4. Standard curve for thromboxane B, and cross-reactivities with other arachidonate metabolites. Closed circles and broken line show a standard curve of thromboxane B,. The cross-reactivity of anti-thromboxane B, antiserum was examined under the standard conditions by incubation of enzyme-labeled thromboxane B, with other arachidonate metabolites at concentrations as indicated in abscissa: prostaglandin D, (0, 2) prostaglandin E, (A, 4). prostaglandin E, (v, 5) prostaglandin F,, (0, 6). prostaglandin F,, (0, 3) 6-ketoprostaglandin F,, (W, 7) 1Sketoprostaglandin Faa (*, l), 13,14-dihydro-15-ketoprostaglandin E, (A, 9). 13,14-dihydro15-ketoprostaglandin F,, (v lo), 5a.7a-dihydroxy-I I-ketotetranorprosta- 1,16-dioic acid ( x . 8).
321
oxane B, determined by the standard enzyme immunoassay. Authentic thromboxane B, could be measured over the range of 0.1-30 pmol which was equivalent to the range of 90-10% binding of the labeled thromboxane B,. The sensitivity defined as the amount of thromboxane B, to reduce the enzyme precipitation from 100 to 90% was about 0.1 pmol. This value of the minimum detectable amount of thromboxane B, by enzyme immunoassay was compared with the values by radioimmunoassay which have been described previously by other investigators: 0.03 pmol [ 17-191, 0.2 pmol [20], and 0.3 pmol (defined as 15% displacement) [21]. Since the detection limit in radioimmunoassay depends on the assay conditions, such as the quality of antiserum or the specific radioactivity of labeled antigen, the minimum detectable amount of thromboxane B, was determined by radioimmunoassay using the same antiserum that was used for our enzyme immunoassay, and a value of 0.4 pmol was calculated from Fig. 1. The cross-reactivity of antiserum with other arachidonate metabolites were tested (Fig. 4). 15Ketoprostaglandin FZo showed a cross-reactivity of 2.8%. Other prostaglandins and their metabolites tested showed less than 1% cross-reaction, Inhibition of the binding of enzyme-labeled thromboxane B, (160 fmol) caused by 100 pmol of each compound was as follows: prostaglandin FZa, 40%; prostaglandin D,, 33%; prostaglandin E,, 32%; prostaglandin E,, 26%; 6-ketoprostaglandin F,,, 18%; prostaglandin F,,, 15%; 5a,7a-dihydroxy11 -ketotetranorprosta-1,16-dioic acid (main urinary metabolite of prostaglandin FZu), 10%; 13,14-dihydro-15-ketoprostaglandin E, (main plasma metabolite of prostaglandin E2), 7%; 13,14-dihydro15-ketoprostaglandin FZu (main plasma metabolite of prostaglandin F,,), 0%. Essentially the same cross-reactivity values were obtained by radioimmunoassay using the same antiserum. Enzyme immunoassay of thromboxane B2 in human plasma For extraction of thromboxane B, from human plasma and its separation from other plasma components, an octadecyl silica column was utilized as described in Materials and Methods. Recovery of
I
0
t-0
.
’
3
0
1
Added
L
I
I
10
100
1000
TXBz
I pmol /ml)
Fig. 5. Correlation between added and measured thromboxane (TX) B,. Known amounts of thromboxane B, were added to human plasma. After application of the plasma sample to an octadecyl silica column as described in Materials and Methods, the dried material from the ethyl acetate extract was dispersed in buffer B so that a 50-1.11aliquot could contain O-10 pmol of thromboxane B,, and then analyzed by standard enzyme immunoassay. The amounts of added (x) and measured (_v) thromboxane B, were plotted. Different symbols indicate the results obtained with different batches of human plasma. r. Correlation coefficient; n, number of samples.
thromboxane B, at this step was examined with 13 samples of plasma (1 ml) to which [3H]thromboxane B, (5900 cpm, 66 fmol) was added. The average recovery of radioactivity in the ethyl acetate extract was 74.5 k 2.0% (mean f S.D.). The recovery was not significantly affected by increasing the amount of thromboxane B, up to 2.7 nmol.
TABLE
I
CORRELATION BETWEEN SELECTED ING AND ENZYME IMMUNOASSAY
ION MONITOR-
Thromboxane B, (5.4-43.2 pmol) was added to 1 ml of human plasma. Ions of m/z 641 and 645 were monitored as described in Materials and Methods. Data on enzyme immunoassay are mean values of duplicate determinations, and are pmol/ml of plasma. Added thromboxane
5.4 10.8 21.6 43.2
Measured
thromboxane
B,
B, Selected ion monitoring
Enzyme immunoassay
19.4 25.1 35.2 55.4
28. I 35.1 41.2 62.6
y-0.92x
+ 4.64
r-=0.96
e
Time I
10 Radioimmunoassay
I
I 100 (Pmol/ml
1000
)
Fig. 6. Correlation between radioimmunoassay and enzyme immunoassay. The plasma samples containing added thromboxane B, were treated as described in Fig. 5 and subjected to both radioimmunoassay (x) and enzyme immunoassay OJ). r, Correlation coefficient; n, number of samples.
Varying amounts of thromboxane B, were added to human plasma, and the samples were subjected to octadecyl silica column chromatography and enzyme immunoassay. As shown in Fig. 5, a correlation was observed between the amounts of added and measured thromboxane B2 (correlation coefficient, 0.99). Intraassay variation was examined with 1 ml of plasma to which 13.5 pmol of thromboxane B2 were added. Nine assays each in duplicate were run at one time, and a coefficient of variation of intraassay values was 13.1%. The same plasma samples containing 13.5 pm01 of thromboxane B, per ml were assayed at 12 different times each in duplicate, giving a coefficient of variation of 15.3% for interassay values. The results obtained by enzyme immunoassay were compared with those by radioimmunoassay. There was a satisfactory correlation (coefficient, 0.96) between the values of thromboxane B, at higher concentrations determined by the two methods (Fig. 6). For further validation of enzyme immunoassay, the added thromboxane B, in human plasma was analyzed by gas chromatographymass spectrometry using a deuterated compound as an internal standard. Table I shows the results obtained by both selected ion monitoring and
(min)
Fig. 7. Release of thromboxane (TX) B, during platelet aggregation Thrombin (4 units) was added (arrow) to start aggregation of washed human platelets (10’ cells), which was monitored as described in Materials and Methods (solid line). Aliquots were removed at indicated time points and mixed with chloroform/methanol/O.02 N HCI (2 : 1 : 0.06). The solvent was evaporated, and the residue dissolved in buffer B was subjected to enzyme immunoassay (O------O).
enzyme immunoassay. The calibration curve of each assay was essentially linear in the range tested, and the added thromboxane B, was recovered in a good yield in all the samples. However, both assays gave higher values than the amount of added thromboxane B,, presumably due to the contribution of endogenous thromboxane B, present in the plasma samples. Furthermore, enzyme immunoassay showed still higher values than selected ion monitoring. The plots on the ordinate of Fig. 5 may give the level of endogenous thromboxane B, in plasma of four individuals; 13.0, 7.8, 5.4 and 5.1 pmol/ml. These values were higher than the previously reported values as determined by radioimmunoassay: 0.656 f 0.259 [17], 0.189 k 0.070 (aspirin treatment) [22], 0.409 & 0.180 (no aspirin) [22] and 2.9 pmol/ml [ 181. Our higher values may be attributed partly to the absence of any cyclooxygenase inhibitor when blood was collected from the donors. At lower thromboxane B, concentrations our enzyme immunoassay gave higher values, to more or less an extent, than radioimmunoassay (Fig. 6) or mass spectrometry (Table I). Possible involvement of unknown endogenous materials in the overestimation by enzyme immunoassay has
329
not yet been extensively investigated. Therefore, these values of endogenous thromboxane B, estimated by enzyme immunoassay await further investigations. Thromboxane
B2 formation
during platelet aggrega-
tion
Aggregation of human platelets was induced by thrombin, and thromboxane B, formation from endogenous arachidonic acid was determined by enzyme immunoassay (Fig. 7). Thromboxane B, accumulated as the platelet aggregation proceeded. The final level of thromboxane B, was about 126 pmol/lOX platelets. Pretreatment of washed platelets with 40 PM indomethacin caused neither aggregation of platelets nor a rise of thromboxane B, level. Acknowledgements
The authors wish to thank Miss Masako Yamamoto and Miss Hiromi Hada for technical assistance. S.Y. was supported by grants-in-aid for scientific research from the Ministry of Education, Science and Culture of Japan, a grant from the Intractable Diseases Division, Public Health Bureau, Ministry of Health and Welfare of Japan, a grant for cardiovascular diseases (56-C2), Ministry of Health and Welfare of Japan, and grants from the Japanese Foundation of Metabolism and Diseases, the Naito Foundation, the Asahi Scholarship Promotion Fund, the Japan Research Foundation for Clinical Pharmacology, Japan Heart Foundation (Tokio Marine Research Grant), Yamada Science Foundation, Suzuken, Kenzo Memorial Foundation, Takeda Science Foundation and Mishima Kaiun Memorial Foundation.
References E. and Kindahl, H. (1978) in Advances in I Granstrom. Prostaglandin and Thromboxane Research (Frolich. J.C., ed.). Vol. 5. pp. 119-210. Raven Press. New York 2 Wisdom, G.B. (1976) Clin. Chem. 22, 1243- 1255 and Van Weemen. B.K. (1977) Clin. 3 Schuurs, A.H.W.M. Chim. Acta 81. l-40 4 Ishikawa, E. (1979) in Techniques in Metabolic Research (Kornberg. H.L., Metcalfe, J.C.. Northcote, D.H.. Pogson, C.I. and Tipton. K.F., eds.). B217. pp. l-18. EIsevier/North-Holland Scientific Publishers Ltd., Limerick S. (1981) Biochim. 5 Hayashi. Y., Yano. T. and Yamamoto. Biophys. Acta 663, 661-668 6 Yano, T.. Hayashi, Y. and Yamamoto, S. (1981) J. Biothem. 90, 7733777 7 Hernandez, 0. (1978) Tetrahedron Lett.. 219-222 8 Nelson. N.A. and Jackson, P.W. (1976) Tetrahedron Lett. 327553278 9 Erlanger, B.F., Borek, F.. Beiser, SM. and Liebermann. S. (1959) J. Biol. Chem. 234. 1090-1094 10 Sund, H. and Weber. K. (1963) Biochem. Z. 337, 24-34 11 Zettner. A. and Duly, P.E. (1974) Clin. Chem. 20, 5- 14 12 Inagawa, T., Ohki, S.. Sawada, M. and Hirata, F. (1972) Yakugaku Zasshi 92. l187- 1194 13 Powell, W.S. (1980) Prostaglandins 20, 947-957 14 Miyazaki. H.. Ishibashi, M.. Yamashita. K., Nishikawa. Y. and Katori. M. (1981) Biomed. Mass Spectrom. 8. 521-526 15 Harada, Y.. Tanaka, K., Uchida. Y., Ueno, A., Oh-ishi. S.. Yamashita. K., Ishibashi, M., Miyazaki. H. and Katori. M. ( 1982) Prostaglandins 23. 88 l-895 16 Wallenfels. K. and Weil. R. (1972) in The Enzymes (Boyer, P.D.. ed.), Vol. 7, pp. 617-663, Academic Press, New York 17 Tada, M., Kuzuya. T.. Inoue, M., Fukushima. M. and Abe. H. (1980) in Advances in Myocardiology (Tajuddin, M.. Bhatia, B.. Siddiqui, H.H. and Rona, G.. eds.), Vol. 2. pp. 297-405, University Park Press, Baltimore 18 Granstrom. E.. Kindahl. H. and Samuelsson. B. (1976) Anal. Lett. 9, 61 l-627 19 Tai. H.H. and Yuan, B. (1978) Anal. Biochem. 87. 343-349 20 Anhut, H., Bernauer, W. and Peskar. B.A. (1977) Eur. J. Pharmacol. 44, 85-88 21 Fitzpatrick. F.A., Gorman, R.R.. McGuire, J.C., Kelly, R.C.. Wynalda. M.A. and Sun, F.F. (1977) Anal. Biochem. 82, 1-7 22 Viinikka, 759-766
L. and Ylikorkala,
0. (1980) Prostaglandins
20,