Quantitation of 13-hydroxyoctadecadienoic acid (13-HODE) by radioimmunoassay

09%3278/9O/lY339-012S/SlO.iB

ProstDglrrndins Leukotrienes and Essential Fatty Acids (199Q 39,125~129 tQ Longman Group UK Ltd 1990

Quantitation of 13,Hydroxyoctadecadienoic Radioimmunoassay

Acid (13-HODE) by

Yasuhide Tonogai and Hsin-Hsiung Tai Division of Medicinal Chemistry and Pharmacognosy, Kentucky, Lexington,

Coffege of Pharmacy,

University of

KY 40.536-0082, USA

ABSTRAC7’.

Antibodies against 13-hydroxyoctadecadienoic acid (134IODE) were produced in rabbits by immunixing the animal with 13-HODE-thyrogiobulin conjugate. The antibodies appeared to be rather specific for 13HODE since other hydroxy fatty acids showed minimal crossreaction. The radiiimmunoassay was capable of detecting 50 pg per assay tube and was applied to the study of the biosynthesis of 134IODE in platelets and leukocytes. In contrast to reported iindings from endotheliai cells, A-23187, thrombin and collagen stimulated synthesis and release of 13-HODE from platelets. However, insignificant synthesis of 13-HODE was found in leukocytes following A-23187 stimulation. Exogenous addition of iinoieic acid stimulated the synthesis of 1IHODE from both platelets and leukocytes. The majority of 1IHODE synthesized was found in the medium. These studies suggest that both types of blood ceils possess active (o-6) iipoxygenase. Platelets may use endogenously released linoleic acid to synthesize 13-HODE, whereas leukocytes may utilize linoleic acid released from other cell types for 13-HODE synthesis.

MATERIALS AND METHODS

INTRODUCTION

Materials

Buchanan et al first reported that endothelial cells produce a lipoxygenase derived chemo-repellant which influences platelet/endothelial cells’ interactions (1). This chemo-repellant was later identified as 13-hydroxy-9-cis, 11-trans-octadecadienoic acid (13-HODE) which was believed to be the lipoxygenase derived metabolite of linoleic acid (2). 13-HODE was found to produce in significantly greater quantities by resting endothelial cells than by smooth muscle cells or by fibroblasts (2). Stimulation of endothelial cells by thrombin or calcium ionophore, A-23187, resulted in a decreased production of 13-HODE. In contrast to the endothelial cells, endotoxin stimulated macrophages and increased 13-HODE content (3). In both instances 13-HODE was found to be associated with the cells after stimulation. Whether 13-HODE is synthesized and released in other cell types remains to be determined. So far, quantitation of 13-HODE has relied much on HPLC analysis which requires a great number of cells for extraction and determination. In order to facilitate the study of the biosynthesis of 13-HODE and its regulation, a more rapid and sensitive assay for 1ZHODE is needed. In this report, we describe the production of antibodies against 13-HODE, the development of radioimmunoassay (RIA) and the use of RIA in studying the biosynthesis of 13HODE in rabbit platelets and leukocytes.

Linoleic acid, A-23187, 1-ethyl-3-(dimethylaminopropyl) carbodiimide (EDC) sodium borohydride, Histopaque 1077, soybean lipoxidase (Type IV), bovine thyroglobulin, bovine serum albumin, bovine-globulin, and complete and incomplete Freund adjuvants were purchased from Sigma Chemical Co. 13-Hydroxy-9-cis, ll-trans-octadecadienoic acid (13-HODE), 12-hydroxy-5,8,10,14eicosatetraenoic acid (12-HETE) and lZhydroxy5,8,10-heptadecatrienoic acid (1ZHHT) were obtained from Cayman Chemical Co. [12,13-3H]linoleic acid (49 Ci/mmol) was supplied by DuPontNEN. Silica gel G plate was supplied by EM Science. All solvents are either reagent grade or HPLC grade. Preparation of 13-HODE Linoleic acid (20 mg) was dissolved in 0.4 ml of ethanol and then added to 50 ml of 0.1 M sodium borate buffer, pH 9.0. About 4.5 mg of soybean lipoxidase and 15 mg of sodium borohydride were added to initiate oxygenation and reduction. The mixture was stirred at room temperature for 10 min. Another 4.5 mg of soybean lipoxidase was added and the mixture was stirred for 10 min. The reaction was terminated by adding 3 ml of 1N HCI and the mixture was extracted with 75 ml of ethyl ether 125

126 Prostaglandins Leukotrienes and Essential Fatty Acids

twice. The ethyl ether extract after evaporation was spotted on a Silica gel G plate (20 x 20 cm) which was developed in a solvent system of ethyl ether/petroleum ether/acetic acid (50:50: 1). The 13-HODE band (Rf = 0.66) was scraped off and the gel was extracted with 2 ml of methanol twice. The methanol extract was further purified by HPLC usingvarian Micropak MCH-10 column as described in the section of HPLC analysis of 13-HODE. The product showed identical retention time with the authentic 13-HODE. About 8 mg of chromatographically pure 1bHODE can be obtained by this procedure. Preparation of [12,13-3H]-13-HODE Preparation of ]12,13-3H]-13-HODE was carried out in a similar manner as described above except on a much smaller scale. Briefly, 33 &i of [12,13-3H]linoleic acid in 10 ~1 of ethanol was incubated with 50 pg of soybean lipoxidase and 200 pg of sodium borohydride in 1 ml of 0.1 M sodium borate buffer, pH 9.0 at room temperature for 25 min. An additional 50 pg of soybean lipoxidase was then added and the incubation continued for another 10 min before addition of 0.1 ml of 1N HCl. The reaction mixture was extracted with 2 ml of ethyl ether three times. The ethyl ether extract after evaporation was purified by a silica gel G plate (2 x 20 cm) using ethyl ether/petroleum ether/acetic acid (50 : 50 :_I) .as the developing solvent. The [12,-13-3H]-13HODE band was scraped off and extracted with 1 ml of acetone three times. After evaporation of acetone [12,13-3H]-13-HODE was dissolved in 0.5 ml of ethanol and stored at -20°C. About 20 $Zi of [12,13-3H]-13-HODE was obtained. Preparation of 13-HODE-thyroglobulin conjugate 13-HODE was conjugated to thyroglobulin according to the procedure previously described for thromboxane B2 (4). Briefly, 13-HODE (2 mg) was dissolved in 0.2 ml of ethanol and diluted to 2 ml with 0.02% Na&03. Bovine thyroglobulin (6 mg) was added followed by 8 mg of EDC. The reaction mixture was adjusted to pH 5.5 and stirred for 16 hrs at 4°C. The product was extensively dialyzed against water, lyophilized and stored at -z20°c-. Production of 13-HODE antiplasma The 13-HODE-thyroglobulin conjugate (1 mg) was dissolved in 1 ml of saline and emulsified in an equal volume of complete Freund’s adjuvant. The emulsion was injected into rabbit intradermally at the back at multiple sites. Booster injection at a half dose was given at monthly intervals. One to two weeks after each injection, the animal was bled by

venous puncture collected into a x g for 10 min. further checking

of the ears. Heparinized blood was tube and then centrifuged at 2000 The plasma was kept at -20°C for of the antibody titer.

Radioimmunoassay of 1IHODE The assay was run in 10 x 75 mm glass test tube in duplicates. The standard assay buffer, 0.05 M TrisHCl buffer, pH 7.5 containing 0.1% gelatin, was used for diluting antiplasma, 13-HODE standards and samples, and labeled 13-HODE. The assay mixture (0.4 ml) contained: 0.2 ml of 13-HODE standards or samples, 0.1 ml of appropriately diluted anti-plasma, and 0.1 ml of tritiated 13HODE (ca. 7 000 cpm). The mixture was incubated at room temperature for 1 h and the separation of the bound and free 13-HODE was achieved by globulin coated charcoal as described previously (4). The radioactivity in the supernatant (bound form) was determined by liquid scintillation counting. The concentration of 13-HODE was estimated from the standard curve. HPLC analysis of 1IHODE and other hydroxy fatty acids Reversed phase HPLC was performed on a Shimadzu LCdA using Varian MicroPak MCH-10 column (4.6 x 300 mm). Elution was carried out with the solvent of eacetonitrile/water/85% H3P04 (55:45:0.1) at a flow rate of 1 ml/min as described previously (5). 13-HODE, 1ZHETE and HHT were detected by monitoring the UV absorbance at 232 nm with Shimadzu SPD-6A detector. The retention times in HPLC for 13-HODE, 12HETE and HHT were 14.4 min, 16.3 min and 9.1 min respectively. Preparation and incubation of rabbit platelets Rabbit platelets were prepared as described previously (6). Briefly, rabbit venous blood anticoagulated with one tenth volume of 3.8% sodium citrate was centrifuged at 200 X g for 15 min at room temperature. The platelet rich plasma after addition of PGEi at 0.5 fig/ml to prevent aggregation was centrifuged at 850 X g for 15 min. The platelet pellet was washed with, and finally suspended in Ca2+, Mg+ free Tyrode buffer. An aliquot of the platelet suspension (7 x 10’ 1.5 x 10’ platelets/ml) was used in various incubations. Incubation of platelets with various stimulants was carried out at 37°C for the desired length of time at the indicated concentration in a final volume of 1 ml of Ca2’, Mg2+ free Tyrode buffer and was terminated by placing the sample in ice followed by centrifugation at 1 000 x g for 10 min at 4°C. The

Quantitation

of 13-Hydroxyoctadecadienoic

supematant was removed for radioimmunoassay and HPLC analysis. For HPLC analysis, an aliquot of the supernatant was acidified by adding 0.01 ml of 1N HCl and the mixture was extracted with 1 ml of ethyl ether twice. The ether extract after evaporation under a stream of N2 was dissolved in 0.1 ml of ethanol and subjected to analysis. The cell pellet was hydrolyzed by adding 0.5 ml of 1N NaOH in CHJOH. The mixture was incubated at 37°C for 1 hr and then diluted with 0.5 ml of HZ0 and acidified with concentrated HCl before extraction with 1 ml of ethyl ether twice. The ether extract was evaporated and dissolved in ethanol for analysis as described above. Preparation

127

Acid (13-HODE) by Radioimmunoassay

% K 8

‘* \

x 0 :

‘1. \ 0

04

1

0.1

100

10

1000

NANOGRAMADDED

Fig. 1 Specificity of the 13-HODE antibodies. The compounds included in this figure are: 13-HODE, -*-a-; 15-HETE, -A-A-; lZHHT, -W-m-: lZHETJZ, -o-o-.

and incubation of rabbit neutrophils

Preparation of human neutrophils was carried out as described previously (7). Briefly, a one-half volume of 3% dextran-saline solution was added to rabbit venous blood (30 ml) anticoagulated with one-tenth volume of 3.8% sodium citrate and stood for 1 h at room temperature. The leukocyte rich tipper layer was removed and centrifuged at 400 x g for 10 min. The pellet was resuspended in Gey’s media containing 2% bovine serum albumin and treated with hypotonic solution for 20 set at 4°C followed by cetirifugation at 400 x g for 10 min. The cell pellet resuspended in 12 ml of Ca*+, Mg*+ free Hanks’ buffer was overlayered onto 3 ml of Histopaque 1077 solution and centrifuged at 550 x g for 30 min. The neutrophil pellet was washed with Ca*+, Mg*+ free Hanks’ buffer once and finally resuspended in Hanks’ buffer. An aliquot (1 ml) of the neutrophii suspension (2.2 x lo6 cells/ml) was used in various incubations. The incubation was carried out as described for platelets. The supernatant and the cell pellet were processed and assayed as described above.

Table 1 Crossreaction of 13-HODE antibodies with various fatty acids Nanograms required for 50% inhibition of maximal binding

13-HODE 1.5-HETE 1ZHHT 1ZHETJZ Linoleic Acid Arachidonic Acid

0.1 22 34 280

70 Crossreaction

100% 3.18% 2.05% 0.25% <0.2% <0.2%

assay was capable of detecting 50 pg of 13-HODE per assay tube. Using this radioimmunoassay, we examined the synthesis and release of 13-HODE into extracellular fluid from rabbit platelets and leukocytes following stimulation with calcium ionophore, A-23187. Figure 2 shows that A-23187 stimulated the synthesis and release of 13-HODE from endogenous linoleic acid in platelets but only

RESULTS AND DISCUSSION Rabbits immunized with 13-HODE-thyroglobulin conjugate produced detectable antibodies after two months of injection, and the titers increased with each booster injection. The standard displacement curve for 13-HODE and the specificity of the antibodies is shown in Figure 1. The crossreaction of the antibodies with hydroxy fatty acids and unsaturated fatty acids is summarized in Table 1. Both HHT and 15-HETE showed higher crossreactivity than 12-HETE indicating the 13-HODE antibodies recognized better the o-6 hydroxyl component of the hydroxy fatty acids since both hydroxy fatty acids but not 1ZHETE possess a was (w-6) hydroxyl group. Little crossreaction observed with tinoleic acid or arachidonic acid. The

PROST.--

B

A-23187

(Y)

Fig. 2 Stimulation of 13-HODE synthesis in rabbit platelets and leukocytes by A-23187. Rabbit platelets and leukocytes were stimulated by A-23187 at the indicated concentrations. Medium was assayed for 13-HODE by,radioimmunoassay after removing blood cells by centrifugation as described in Methods.

128 Prostanlandins Leukotrienes and Essential Fattv Acids

i

0

10

20

TIME (YIN) Fig. 3 Kinetics of the synthesis of 13-HODE, 12-HETE and 1ZHHT in rabbit platelets following stimulation with A-23187. Rabbit platelets were stimulated by 1 I.IM of A-23187 and the incubation was terminated at the indicated time by centrifugation. 13-HODE in the medium was assayed by radioimmunoassay (-o-) and HPLC (-a-), whereas 12-HETE (-A-) and 12-HHT (-a-) were assayed by HPLC as described in the Methods.

slightly in leukocytes. A-23187 at lo- %I began to stimulate the synthesis of 13-HODE. Near maximal stimulation was observed at lo- ‘MM. The time course of A-23187 induced 13-HODE synthesis and release into extracellular fluid from rabbit platelets as determined by radioimmunoassay and further confirmed by HPLC is shown in Figure 3. Simultaneous quantitation of the 1Zlipoxygenase product, lZHETE, and cyclooxygenase product, IZHHT, by HPLC is also shown in the same Figure. Levels of 13-HODE at different time points as determined by radioimmunoassay agreed reasonably well with those determined by HPLC, indicating the validity of the. method of quantitation by radioimmunoassay. Furthermore, the sensitivity of the radioimmunoassay is at least 100 times higher than that of HPLC making the assay more valuable in studies where levels of 13-HODE are found to be low. This is illustrated in Table 2 which shows that levels of 13-HODE synthesis induced by thrombin (1 U/ml) or collagen (100 pg/ml) in rabbit

Table 2 Stimulation of 1fHODE by various agonists

platelets were found to be much less than that by A-23187 (1 PM) and could be detected by radioimmunoassay but not by HPLC analysis. The formation of 13-HODE follows similar kinetics to those of 1ZHETE and HHT. The latter two products are known to derive from two separate pathways, lipoxygenase and cyclooxygenase. Although it is not clear if linoleic acid is transformed to 13-HODE by arachidonate 12- or 15-lipoxygenase reported to be present in platelets (8, 9), it appears that 13-HODE is more likely to be the product of 15-lipoxygenase since both 13-HODE and 15HETE are (o-6) hydroxy fatty acids. Platelets are known to contain linoleic acid esterified in phospholipids (10). Upon stimulation by thrombin, collagen or A-23187, linoleic acid was released and oxygenated to form 13-HODE, as shown in Table 2. However, both thrombin and collagen appeared to release much less 13-HODE than did A-23187. This is in contrast to the situation where the release of 1ZHETE and HHT or TXB2 induced by thrombin or collagen was more comparable to that by A-23187 (11). This suggests that receptor activation may release more specifically acid than linoleic acid. Rabbit arachidonic leukocytes did not appear to synthesize and release a significant amount of 13-HODE upon stimulation with A-23187 as described above. We suspect that this may be due to low levels of endogenous linoleic acid content either in free acid or in ester form but not due to the lack of lipoxygenase activity in these blood cells. To examine this point, rabbit leukocytes were incubated with exogenous linoleic acid in the absence and presence of A-23187. Table 3 indicates that A-23187 did not stimulate the production of 13” HODE either releasing into the extracellular space or existing with the cell. Addition of linoleic acid to leukocyte suspensions stimulated the production of 13-HODE which was released into the medium as well as retained in the cell. The ratio was approximately 4 to 1 in favor of release into the extracellular space. Similar experiments were car-

Table 3 Quantitation of 13-HODE in medium and in cells following addition of exogenous linoleic acid to rabbit leukocytes and platelets.

synthesis in rabbit platelets

Platelets

(ng/loR/lO min) I3-HODE RIA

A-23187 Thrombin Collagen

4.80 0.17 0.12

(ng/lO”/5min) HPLC

4.28 ND ND

ND - Not detectable Rabbit platelets were stimulated with A-23187 (1 FM), or thrombin (1 U/ml) or collagen (100 &ml) for 5 min. Medium was assayed for 13-HODE by radioimmunoassay and HPLC methods after removing platelets by centrifugation as described in Methods.

A-23187 Linoleic Acid A-23187 plus Linoleic Acid

Medium 0 260

Cell 0 65

Medium 7 88

285

75

90

Cell 2 19 25

Rabbit leukocytes or platelets were respectively stimulated with A-23187 (1 PM) or linoleic acid (200 I.LM)or A-23187 plus linoleic acid for 10 min. Following stimulation cells were removed by centrifugation and treated with alkaline CHPH as described in Methods. Both medium and cellular 13-HODE were quantitated by HPLC Method.

Quantitation of 13-Hydroxyoctadecadienoic Acid (IIHODE)

ried out for platelets, as shown in the same Table. A-23187 did stimulate the synthesis and release of 13-HODE in platelets. Addition of linoleic acid to platelet suspension stimulated the synthesis of 13HODE which was found to be present in both extracellular space and in cell pellet. The ratio was nearly 4 to 1 in favor of release into extracellular space. These results indicate that both leukocytes and platelets exhibit active linoleate 13-lipoxygenase which appears to be identical with arachidonate 15 lipoxygenase reported to be present in both cell types (9, 12). The significance of the synthesis of 13-HODE was first reported by Buchanan et al (1) when they described that endothelial cells produced 13-HODE which maintained the thromboresistant property of the vessel wall. The production of 13-HODE appeared to be active under basal conditions and is decreased by thrombin, calcium ionophore or trypsin stimulation (2). We have found that both platelets and leukocytes do not synthesize 13HODE under basal conditions unless they are stimulated by exogenous agonists. Furthermore, 13HODE synthesized following stimulation appeared to release into extracellular space. This is in direct contrast with endothelial cells in which 13-HODE is retained in the cells after stimulation. The significance of the production of 13-HODE particularly in platelets following thrombin or collagen stimulation is not clear. Further investigation into the role of 13-HODE in platelet function may prove to be rewarding in our understanding of the nature of the interaction between platelets and other blood cells or vascular cells.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

Acknowledgements This study was supported in part by a grant from The Kentucky Tobacco Research Board. We are indebted to Ms. Jean Cavenee for typing the manuscript.

References 1. Buchanan M R, Butt R W, Magas Z, Van Ryn J, Hirsh J, Nazir D J. Endothelial cells produce a

12.

by Radioimmunoassay

lipoxygenase derived chemo-repellent which influences platelet leudothelial cell interactions-effect of aspirin and salicylate. Thromb. Haemostas 53: 306-311, 1985. Buchanan M R, Haas T A, Lagarde M, Guichardant M. 13-Hydroxyoctadecadienoic acid is the vessel wall chemo-repellant factor, LOX. J. Biol. Chem. 260: 16056-16059, 1985. Schade U F, Burmeister I, Engel R. Increased 13-hydroxyoctadecadienoic acid content in lipopolysaccharide stimulated macrophages. Biochem. Biophy. Res. Commun. 147: 695-700, 1987. Tai H H, Yuan B. Development of radioimmunoassay for thromboxane B2. Anal. Biochem. 87: 343-349. Liu Y, Yoden K, Shen R F, Tai H H. 12-Hydroxy-5,8,10-heptadecatrienoic acid (HHT) is an excellent substrate for NAD+-dependent 15-hydroxyprostaglandin dehvdrogenase. B&hem. Biophys. Rcs. Commun. 129: 268-274. Yamazaki H. Tai H H. Phorbol ester induces inhibition of arachidonate incorporation into phospholipids in human neutrophils. Prostaglandins Leukotrienes Ess. Fatty Acids (in press), 1989. Fuse 1, Tai H H. Stimulations of arachidonate release and inositol-1,4,5_triphosphate formation are mediated by distinct G-proteins in human platelets. Biochem. Biophys Res. Commun 146: 657-665, 1987. Hamberg M, Samuelsson B. Prostaglandin endoperoxides, novel transformations of arachidonic acid in human platelets. Proc. Natl. Acad. Sci. 71: 3400-3404, 1974. Wong P Y K, Westlund P, Hamberg M, Granstrom E, Chao P H W, Samuelsson B. 15-Lipoxygenase in human platelets. J. Biol. Chem. 260: 9162-9165, 1985. Imai A, Yano K, Kameyama Y, Nozawa Y. Reversible thrombin-induced modification of positional distribution of fatty acids in platelet phospholipids: involvement of deacylation-reacylation. Biochem. Biophys. Res. Commun. 103: 1092-1099, 1981. Mobley A. Tai H H. Synergistic stimulation of thromboxane biosynthesis by calcium ionophore and phorbol ester or thrombin in human platelets. Biochem. Biophys. Res. Commun. 130: 717-723. 1985. Narumiya S, Salmon J A. Cottee F H, Weatherley B C. Flower R J. Arachidonic acid 15-lipoxygenase from rabbit peritoneal polymorphonuclear leukocytes: partial purification and properties. J. Biol. Chem. 256: 9583-9582. 1981.

129