- Email: [email protected]
Prostaglandin endoperoxide and thromboxane generating systems and their selective inhibition
PROSTAGLANDINS
PROSTAGLANDIN ENDOPEROXIDE AND THROMBOXANE GENERATING SYSTEMS AND THEIR SELECTIVE INHIBITION Salvador Moncada, Philip Needleman*, Stuart John R. Vane
Bunting
and
Wellcome Research Laboratories Langley Court Beckenham Kent BR3 3BS England * Present address:
Department of Pharmacology, Washington University Medical School, St. Louis, Missouri, U.S.A.
ABSTRACI' Two enzyme systems and their selective inhibition are Microsomes from ram seminal vesicles (RSV) described. incubated with arachidonic acid at 22O C generated a rabbit aorta contracting substance which, after rapid ether extraction, had characteristics similar to purified Incubation of either purified standard endoperoxides. endoperoxide or the product from RSV and arachidonic acid with horse platelet microsomes (HPM) yielded a more potent rabbit aorta contracting substance characterized as thromboxane A2, with a half life of 35.9 l 2.2 s at 37O C after Two inhibitors, indomethacin and ether extracticm. benzydamine exhibited selectivity for the two enzyme The IC50 for benzydamine against thromboxane systems. synthetase was 100 ug/ml and 250 us/ml against RSV. Indomethacin shcwed a greater degree of selectivity with an IC5C of 5 us/ml for the ram seminal vesicle cyclooxygenase compared to 100 pg/ml for thromboxane synthetase. ACKNCWLEDGEMENT Part of this work was supported by grants HL-17646 and We would like to thank Sharon RCDA HL-19586 from NIH. L. Kay and Millie Parsons for excellent technical assistance.
SEPTEMBER
1976
VOL. 12 NO. 3
323
PROSTAGLANDINS
Prostaglandin endoperoxides (PGG2 and PGH2) exert They cause contraction pronounced biological effects. of strips of rabbit aorta (80 - 200 times more potent than PGE2), guinea-pig trachea (8 times more potent than PGF2,) and rat stomach (l/3 to l/2 as potent as PGE2). They also increase airway resistance in guinea-pigs and aggregate It has, therefore, been suggested human platelets (1,2). that endoperoxides are of importance in various physiopathological processes such as anaphylaxis and thrombosis. Some or all of these actions of the endoperoxides may be due to enzymic formation of thromboxane A2 (TXA2). We have recently identified an enzyme in the microsomal fraction of human and horse platelets which converts prostaglandin Sheep vesicular gland microendoperoxides into TXA2 (3). somes produce a rabbit aorta contracting substance (4,5) which is a mixture of the prostaglandin endoperoxides PGG and PGH2. We have used both enzyme systems separately ani coupled together further to differentiate the products of We have also studied the susceptibility of each reaction. thromboxane synthetase to selective inhibition. Lyophilized microsomes from horse platelets (100,000 x g pellet; 77% protein) (6) were prepared as described earlier (3) and used as the source of thromboxane synthetase. Lyophilized microsomes from ram seminal vesicles (RSV microsomes; 50% protein) were prepared according to the method of Takeguchi -et al (7) and used as the source of prostaglandin synthetase (cycle-oxygenase). Prostaglandin endoperoxides were generated by incubating RSV microsomes with arachidonic acid at room temperature They were extracted and purified (220 C) without cofactors. by column chromatography (Ubatuba, Moncada and Vane, unpublished). Biological activity was assayed on a rat stomach strip (8) and a rabbit aorta (9) superfused at 10 ml/min with Krebs' solution at 370 C containing a mixture of antagonists (10) and indomethacin (1 pg/ml) to inhibit synthesis of prostaglandins by the assay tissues (11) and to make the assay more specific for prostaglandins. Rapid ether extraction and stability of thromboxane A2 Rapid ether extraction was carried out after incubating The PGG2 with platelet microsomes for two minutes on ice. reaction was stopped by shaking vigorously for 5-10 set with an aliquot of ether (dried by addition of sodium wire) at Both phases were allowed to separate and the upper oo c. (ether phase) was then aspirated and dried under a flow of The dried extract was resuspended in 0.5 ml nitrogen.
SEPTEMBER 1976
VOL. 12 NO. 3
PROSTAGLANDINS
of Tris buffer (50 mM pH 7.8) and tested for activity on the rabbit aorta. The whole procedure took 60-75 seconds. Endoperoxide and thromboxane generating systems Incubation of arachidonate with RSVmicrosomes at 22O C produced an aorta contracting substance (Fig. 1). The reaction reached a peak at 2-5 min and declined slowly, Short term activity disappearing at about 60 min. incubation of RSV microsomes and arachidonate, in the absence of cofactors, favours endoperoxide formation over PGE2 or PGF2o (5). Incubation of PGG2 (or PGH2) with platelet microsomes at Oo C generated thromboxane A2 (Fig. l), which was characterized by the increase in contractile activity of the incubate on the rabbit aorta (3). A simple, rapid thromboxane A2 generating system was provided by coupling the two enzymic reactions together as The RSV enzyme generated endoperoxdescribed in Fig. 1. ides which then formed the substrate for the thromboxane synthetase of the platelet microsames. There is substantial evidence that the product of the RSV enzyme was a prostaglandin endoperoxide rather than TXA2. First, the RSV enzyme did not convert PGG2 into the Secondly, the product of the more active compound. incubation of RSV microsomes and arachidonate was readily converted by horse platelet microscxnesinto a substance which was much more potent in causing contractions of the Thirdly, the dose-response curve rabbit aorta (Fig. 1). for arachidonic acid plus RSV microsomes (Fig. 2) was parallel to the one observed with pure PGG2 or PGH2 in our Fourthly, horse platelet microsomes previous work (3). caused a comparable increase in potency on the rabbit aorta of either authentic PGG (or PGH2) or the product of incubation of seminal ves1 cles plus arachidonate (Fig. 1). The product of arachidonate incubated with RSV microsomes was isolated by rapid ether extraction. It decayed in aqueous solution at exactly the same rate (half life for loss of aorta contracting activity = 7 min) at room temperature as authentic PGG2 incubated without enzyme under the same conditions. This half life agrees with that previously reported (2). The product (TXA2) of the incubation of PGG2 with platelet microsomes or of the incubation of arachidonic acid with RSV microsomes followed by platelet microsomes had the same half life (2 30 set at 37o C) which was considerably shorter than that of the endoperoxides (Fig. 3). The half
SEPTEMBER 1976
VOL. 12 NO. 3
PROSTAGLANDINS
life in aqueous solution of TXA2 was previously reported to be 34 set (platelets) and 29 set (guinea-pig lung) (12). The product of the RSV cannot be transformed by the platelet microsomal fraction into a more potent aorta contracting substance after the RSV reaction is complete (>60 min incubaticm) (Fig. 4) or when it is carried out in the presence of glutathione (50 pg) which favours the breakdown of the endoperoxide to prostaglandin E (5). For all these reasons, we consider the mixture o3 R8V microsomes and arachidonate as an endoperoxide generating system (i.e. cycle-oxygenase) and the mixture of platelet microsomes and PGG as a TXA generating system (i.e. thromboxane synthez ase). T e use of these two enzyme systems either alone or in combination readily provides a source of the otherwise precious (i.e. extremely unstable and heat labile) materials.
2
Extraction of TKA, Ether extraction was carried out at neutral pH (7-7.5), for acid (pH 4-4.5) or alkali (pH 8-9) conditions, followed by extraction destroyed all rabbit aorta contracting activity; even reduction of pH to 5-5.6 reduced the recoverable activity. Resuspensicn of the ether extract in Krebs' solution at different temperatures shawed that the half life of the extracted thromboxane was 35.9 2 2.2 set at 370 C- 80 2 16 i s.e. of set at 22O C and 9.4 ,+1.2 minutes at Oo C (mean mean; three experiments) (Fig. 5). When the dried extract was dissolved in dry acetone it lost about 20% activity in 24 h at -200 C. Inhibition of thromboxane synthetase Platelet microsanes at 22O C were pre-incubated for 3 min with potential inhibitors, the tube was then placed in an ice bath, PGG2 was added and the formation of thromboxane A2 was assayed onthe rabbit aorta. When inhibition occurred the compound was also tested against the endoperoxide generating system. Benzydamine inhibited the conversion of PGG2 to TKA2 by horse platelet microsomes in a concentration dependent manner Benzydamine was somewhat less active against (Fig. 6). endoperoxide generation and thus showed some selectivity. For 50% inhibition, 100 ug/ml was needed against TKA2 synthetase and 250 pg/ml against the RSV cycle-oxygenase (Fig. 6). D2 or El (20 ug/ml) and chlorColchicine (400 pg/ml), PGD promazine (50 pg/ml) were ai'1 ineffective against thromboxane synthetase.
326
SEPTEMBER 1976
VOL. 12 NO. 3
PROSTAGLANDINS
Indomethacin showed a high degree of selectivity against the cycle-oxygenase (Fig. 6, right panel) : there was a 50% inhibition at 5 ug/ml, a concentration which did Indeed, for 50% inhibition not affect TEA*-synthetase. of the platelet microsome enzyme, indomethacin at 100 ug/ml Other compounds with differential selectivity was needed. were phenylbutazone, which produced 90% inhibition of the cycle-oxygenase at 500 ug/ml without affecting TXA2synthetase, and naproxen which produced >50% inhibition of the cycle-oxygenase at 20 ug/ml without activity on the TXA2-synthetase. Benzydamine was tested because when incubated with rabbit kidney microscnnes and arachidonate at a concentration that caused a 50% inhibition of PGF2 and PGD2 formation, In the same there was an increase in PGE2 format?on (13). system, phenylbutazone did not inhibit PGD2 formation at concentrations that inhibited PGE2 and PGF2u synthesis. Thus, these drugs appeared differentially to inhibit enzymes which transformed PGG2 to other substances such as D2 or E2. The products which arise from prostaglandin endoperoxide transformation have actions which may be beneficial or detriInhibition of cycle-oxygenase (e.g. mental to the organism. with indomethacin) blocks the formation of endoperoxides and Since some of them therefore of all these products also. may be useful to the organism, the methods we have described should allow the discovery of agents with an even greater selectivity for the inhibition of thromboxane synthetase than benzydamine.
SEPTEMBER
1976 VOL. 12 NO. 3
327
PROSTAGLANDINS
REFERENCES 1.
Hamberg, M., Hedqvist, P., Strandberg, K., Svensson, J. and Samuelsson, B. Prostaglandin endoperoxides IV. Effects on smooth muscle, Life Sci. %:451-462, 1975.
2.
Hamberg, M., Svensson, J., Wakabayashi, T. and Samuelsson, 8. Isolation and structure of two prostaglandin endoperoxides that cause platelet aggregation, Proc. Natl. Acad. Sci. USA -71: 345349, 1974.
3.
Needleman, P., Moncada. S., Bunting. S., Vane, J.R., Hamberg, M. and Samuelsson, B. Identification of an enzyme in platelet microsomes which generates thromboxane A2 from prostaglandin endoperoxides, Nature (London) 261: 558-560, 1976. Hamberg, M. and Samuelsson, B. Detection and isolation of an endoperoxide intermediate in prostaglandin biosynthesis, Proc. Natl. Acad. Sci. USA 3: 899-903, 1973.
4.
5.
Nugteren, D.H. and Hazelhof, E. Isolation and properties of intermediates in prostaglandin biosynthesis, Biochim. Biophys. Acta. 1973.
6.
Gornall, A.G., Bardawill, C.J. and David, M.M. Determination of serum proteins by means of the biuret reaction, J. Biol. Chem. 177: 751-766, 1949.
7.
Takeguchi, C., Kohno, E. and Sih, C.J. Mechanism of prostaglandin biosynthesis. I. Characterization and assay of bovine prostaglandin synthetase, Biochemistry, -10: 2372-2376, 1971.
8.
Vane, J.R. A sensitive method for the assay of 5hydroxytryptamine, Br. J. Pharmacol. 12: 344-349, 1957.
9.
Piper, P.J. and Vane, J.R. Release of additional factors in anaphylaxis and its antagonism by antiinflammatory drugs, Nature (London) 223: 29-35, 1969.
10.
Gilmore, N., Vane, J.R. and Wyllie, H.J. Prostaglandins released by the spleen, Nature (London) 218: 1135-1140, 1968.
11.
Eckenfels, A. and Vane, J.R. Prostaglandins, oxygen tension and smooth muscle tone, Br. J. Pharmacol. -45: 451-462, 1972.
328
SEPTEMBER
1976
VOL. 12 NO. 3
PROSTAGLANDINS
12.
Svensson, J., Hamberg, M. and Samuelsson, B. Prostaglandin endoperoxides IX. Characterization of rabbit aorta contracting substance (RCS) from guinea pig lung and human platelets, Acta Physiol. Stand. 2: 222-228 1975.
13.
Blackwell, G.J., Flower, R.J. and Vane, J.R. Some characteristics of the prostaglandin synthesizing system in rabbit kidney microsomes, Biophys. Acta. 398: 178-190 1975.
SEPTEMBER
1976 VOL. 12 NO. 3
329
PROSTAGLANDINS
PG.G2 2fO MIC sv
+
10 M’C
The enzymic generation of endoperoxides (SV + AA) g 1. 0":ihromboxane A2 (AA + SV MIC). The abbreviations used are: SV = ram seminal vesicle microsanes; AA = arachidonic acid; MIC = horse platelet microsomes. The RSV microsomes (300 ug protein) were incubated in 500 ~1 of Tris buffer (50 mM, pH 7.8) with various concentrations of arachidonate for 3 min at room temperature. An aliquot (100 - 200 I.rl) of the reaction was tested on rabbit aorta. Horse platelet microsomes were then added to the reaction mixture for 30 set and a further aliquot was tested on the rabbit aorta. Higher concentrations of arachidonate (100 - 200 ng) produced greater effects and the gain of the amplifier was halEd. Numbers refer to nanograms used.
330
SEPTEMBER 1976
VOL. 12 NO. 3
PROSTAGLANDINS
1
30
00
100
ARACHIDONATE
200 ng
The generation of endoperoxides by RSV microsomes
epended on the concentration of arachidonate used. M-2 The generation of TXAZ by platelet microsomes (SWMIC) was
For each of these experiments, the similarly dependent. gain of the amplifier was adjusted to give equal responses Each point also shows the to a standard dose of PGG2. standard error of the mean (n = 3).
SEPTEMBER
1976 VOL. 12 NO. 3
331
PROSTAGLANDINS
+MlCROSOMES
2
4
6
MINUTES
Spontaneous decay of rabbit aorta contracting Fig. 3. activity following ether extraction after enzyme incubations. Four incubations (n=3 for each) were carried out: a) 500 ng of PGG2 was incubated in 500 ~1 of Tris buffer (50 mM, pH 7.8) for 3 min (left panel.), b) RSV microsomes (340 pg protein) were incubated in 650 l.11 of Tris buffer with arachidonate (5 pg) at 22O C for 3 min (left panel.); c) horse platelet microscxnes (350 ug protein) were incubated in 500 ~1 of Tris buffer with 500 ng of PGG at 00 c for 2 min (right panelo) i d) same as b) WI*&l the addition at 3 min of horse platelet microsomes (350 pg protein) for 30 set before extraction The reactions were stopped by shaking (right panel*). with 2 ml of ice cold ether; the ether phase was removed and dried in a stream of nitrogen. The resultant residue was dissolved in 0.5 ml of Tris buffer (at 22O C) and 100 ~1 aliquots were tested after various times in order to measure the half life of the activity Each point also.shows the standard error on rabbit aorta. of the mean (n = 3).
332
SEPTEMBER
1976
VOL. 12 NO. 3
PROSTAGLANDINS
Fig.
4.
generation
Decay of endoperoxide activity with of prostaqlandin E2-like activity.
concomitant
The figure shows tracings from a rabbit aorta and rat stomach strip superfused in cascade with Krebs' solution at 10 ml/min. For ccnvenience, the tracings have been arranged in three panels. Arachidonate (2 pq) was incubated with RSV micl-osomes (300 ug of protein) in 500 ~1 of Tris buffer (50 mM pH 7.8) at 22O C for 3 min (AA + SV, top panel and right-hand side of bottom panel), 25 min (middle panel) and 60 min (left-hand side of bottom panel). 100 ~1 of the mixture was tested and gave the contractions shown. The remainder of the mixture was then incubated with horse platelet microsomes (380 )1g protein) at O" C for 2 min (+ MIC) and 100 ~1 were immediately tested on the assay tissues.
SEPTEMBER 1976
VOL. 12 NO. 3
333
PROSTAGLANDINS
Fig. 4. continued The endoperoxide activity (rabbit aorta) declined with time. There was a concomitant Increase in PGEq-like activity (rat stomach strip). As the endopero;ide activity declined, the ability to form TXA2 (increased activity on rabbit aorta) by addition of horse platelet microsomes also decreased. Calibrating doses of PGG2 and PGE2 are shown in the top two panels (right hand side).
I
I-
I
I
I
I
0
1
2
3’4
I
I
I
5 6 MINUTES
I
I
I
7
8
9
I
10
“J
/r 20
Half life of extracted TXA2 at different Fig. 5. PGG2 (50 ng) was incubated with horse temperatures. platelet microsomes (380 ug protein) for 2 mln in an ice bath. The reaction was then stopped by shaking with 1 ml The ether phase was removed and dried of ice cold ether. The dried extract was resuspended in a stream of nitrogen. in 500 ul of Tris buffer at O" C, 22O C or 37O C and an aliquot (100 ~1) was tested for activity on rabbit aorta at different times. The half life of TXA2 at each temperature (see text) was calculated by comparisons with standard volume/response curves for extracted TXA2.
334
SEPTEMBER 1976
VOL. 12 NO. 3
PROSTAGLANDINS
BENZYDAMINE
us/ml
INDOMETHACIN
Comparison of inhibitian of TXA2-synthetase and Fig. 6. cycle-oxygenase by benzydamine and indomethacin. Horse platelet microsomes (MIC) (154 ug/protein) were preincubated with inhibitor in 500 ul of Tris buffer (50 mM, The reaction mixture was pH 7.8) for 3 min at 220 C. transferred to an ice bath and PGG2 (100 ng) was added. An aliquot (50 ~1) was bioassayed after 2 min. RSV microsomes (300 ug) were inubated at 220 C with inhibitor Arachidonate (2 vg) for 3 min in Tris buffer (400 ~1). was then added and after a further 3 min 100 ~1 was tested for endoperoxide formation. Values are the means 2 standard error of mean (n = 4).
SEPTEMBER
1976
VOL. 12 NO. 3
335
PROSTAGLANDIN ENDOPEROXIDE AND THROMBOXANE GENERATING SYSTEMS AND THEIR SELECTIVE INHIBITION Salvador Moncada, Philip Needleman*, Stuart John R. Vane
Bunting
and
Wellcome Research Laboratories Langley Court Beckenham Kent BR3 3BS England * Present address:
Department of Pharmacology, Washington University Medical School, St. Louis, Missouri, U.S.A.
ABSTRACI' Two enzyme systems and their selective inhibition are Microsomes from ram seminal vesicles (RSV) described. incubated with arachidonic acid at 22O C generated a rabbit aorta contracting substance which, after rapid ether extraction, had characteristics similar to purified Incubation of either purified standard endoperoxides. endoperoxide or the product from RSV and arachidonic acid with horse platelet microsomes (HPM) yielded a more potent rabbit aorta contracting substance characterized as thromboxane A2, with a half life of 35.9 l 2.2 s at 37O C after Two inhibitors, indomethacin and ether extracticm. benzydamine exhibited selectivity for the two enzyme The IC50 for benzydamine against thromboxane systems. synthetase was 100 ug/ml and 250 us/ml against RSV. Indomethacin shcwed a greater degree of selectivity with an IC5C of 5 us/ml for the ram seminal vesicle cyclooxygenase compared to 100 pg/ml for thromboxane synthetase. ACKNCWLEDGEMENT Part of this work was supported by grants HL-17646 and We would like to thank Sharon RCDA HL-19586 from NIH. L. Kay and Millie Parsons for excellent technical assistance.
SEPTEMBER
1976
VOL. 12 NO. 3
323
PROSTAGLANDINS
Prostaglandin endoperoxides (PGG2 and PGH2) exert They cause contraction pronounced biological effects. of strips of rabbit aorta (80 - 200 times more potent than PGE2), guinea-pig trachea (8 times more potent than PGF2,) and rat stomach (l/3 to l/2 as potent as PGE2). They also increase airway resistance in guinea-pigs and aggregate It has, therefore, been suggested human platelets (1,2). that endoperoxides are of importance in various physiopathological processes such as anaphylaxis and thrombosis. Some or all of these actions of the endoperoxides may be due to enzymic formation of thromboxane A2 (TXA2). We have recently identified an enzyme in the microsomal fraction of human and horse platelets which converts prostaglandin Sheep vesicular gland microendoperoxides into TXA2 (3). somes produce a rabbit aorta contracting substance (4,5) which is a mixture of the prostaglandin endoperoxides PGG and PGH2. We have used both enzyme systems separately ani coupled together further to differentiate the products of We have also studied the susceptibility of each reaction. thromboxane synthetase to selective inhibition. Lyophilized microsomes from horse platelets (100,000 x g pellet; 77% protein) (6) were prepared as described earlier (3) and used as the source of thromboxane synthetase. Lyophilized microsomes from ram seminal vesicles (RSV microsomes; 50% protein) were prepared according to the method of Takeguchi -et al (7) and used as the source of prostaglandin synthetase (cycle-oxygenase). Prostaglandin endoperoxides were generated by incubating RSV microsomes with arachidonic acid at room temperature They were extracted and purified (220 C) without cofactors. by column chromatography (Ubatuba, Moncada and Vane, unpublished). Biological activity was assayed on a rat stomach strip (8) and a rabbit aorta (9) superfused at 10 ml/min with Krebs' solution at 370 C containing a mixture of antagonists (10) and indomethacin (1 pg/ml) to inhibit synthesis of prostaglandins by the assay tissues (11) and to make the assay more specific for prostaglandins. Rapid ether extraction and stability of thromboxane A2 Rapid ether extraction was carried out after incubating The PGG2 with platelet microsomes for two minutes on ice. reaction was stopped by shaking vigorously for 5-10 set with an aliquot of ether (dried by addition of sodium wire) at Both phases were allowed to separate and the upper oo c. (ether phase) was then aspirated and dried under a flow of The dried extract was resuspended in 0.5 ml nitrogen.
SEPTEMBER 1976
VOL. 12 NO. 3
PROSTAGLANDINS
of Tris buffer (50 mM pH 7.8) and tested for activity on the rabbit aorta. The whole procedure took 60-75 seconds. Endoperoxide and thromboxane generating systems Incubation of arachidonate with RSVmicrosomes at 22O C produced an aorta contracting substance (Fig. 1). The reaction reached a peak at 2-5 min and declined slowly, Short term activity disappearing at about 60 min. incubation of RSV microsomes and arachidonate, in the absence of cofactors, favours endoperoxide formation over PGE2 or PGF2o (5). Incubation of PGG2 (or PGH2) with platelet microsomes at Oo C generated thromboxane A2 (Fig. l), which was characterized by the increase in contractile activity of the incubate on the rabbit aorta (3). A simple, rapid thromboxane A2 generating system was provided by coupling the two enzymic reactions together as The RSV enzyme generated endoperoxdescribed in Fig. 1. ides which then formed the substrate for the thromboxane synthetase of the platelet microsames. There is substantial evidence that the product of the RSV enzyme was a prostaglandin endoperoxide rather than TXA2. First, the RSV enzyme did not convert PGG2 into the Secondly, the product of the more active compound. incubation of RSV microsomes and arachidonate was readily converted by horse platelet microscxnesinto a substance which was much more potent in causing contractions of the Thirdly, the dose-response curve rabbit aorta (Fig. 1). for arachidonic acid plus RSV microsomes (Fig. 2) was parallel to the one observed with pure PGG2 or PGH2 in our Fourthly, horse platelet microsomes previous work (3). caused a comparable increase in potency on the rabbit aorta of either authentic PGG (or PGH2) or the product of incubation of seminal ves1 cles plus arachidonate (Fig. 1). The product of arachidonate incubated with RSV microsomes was isolated by rapid ether extraction. It decayed in aqueous solution at exactly the same rate (half life for loss of aorta contracting activity = 7 min) at room temperature as authentic PGG2 incubated without enzyme under the same conditions. This half life agrees with that previously reported (2). The product (TXA2) of the incubation of PGG2 with platelet microsomes or of the incubation of arachidonic acid with RSV microsomes followed by platelet microsomes had the same half life (2 30 set at 37o C) which was considerably shorter than that of the endoperoxides (Fig. 3). The half
SEPTEMBER 1976
VOL. 12 NO. 3
PROSTAGLANDINS
life in aqueous solution of TXA2 was previously reported to be 34 set (platelets) and 29 set (guinea-pig lung) (12). The product of the RSV cannot be transformed by the platelet microsomal fraction into a more potent aorta contracting substance after the RSV reaction is complete (>60 min incubaticm) (Fig. 4) or when it is carried out in the presence of glutathione (50 pg) which favours the breakdown of the endoperoxide to prostaglandin E (5). For all these reasons, we consider the mixture o3 R8V microsomes and arachidonate as an endoperoxide generating system (i.e. cycle-oxygenase) and the mixture of platelet microsomes and PGG as a TXA generating system (i.e. thromboxane synthez ase). T e use of these two enzyme systems either alone or in combination readily provides a source of the otherwise precious (i.e. extremely unstable and heat labile) materials.
2
Extraction of TKA, Ether extraction was carried out at neutral pH (7-7.5), for acid (pH 4-4.5) or alkali (pH 8-9) conditions, followed by extraction destroyed all rabbit aorta contracting activity; even reduction of pH to 5-5.6 reduced the recoverable activity. Resuspensicn of the ether extract in Krebs' solution at different temperatures shawed that the half life of the extracted thromboxane was 35.9 2 2.2 set at 370 C- 80 2 16 i s.e. of set at 22O C and 9.4 ,+1.2 minutes at Oo C (mean mean; three experiments) (Fig. 5). When the dried extract was dissolved in dry acetone it lost about 20% activity in 24 h at -200 C. Inhibition of thromboxane synthetase Platelet microsanes at 22O C were pre-incubated for 3 min with potential inhibitors, the tube was then placed in an ice bath, PGG2 was added and the formation of thromboxane A2 was assayed onthe rabbit aorta. When inhibition occurred the compound was also tested against the endoperoxide generating system. Benzydamine inhibited the conversion of PGG2 to TKA2 by horse platelet microsomes in a concentration dependent manner Benzydamine was somewhat less active against (Fig. 6). endoperoxide generation and thus showed some selectivity. For 50% inhibition, 100 ug/ml was needed against TKA2 synthetase and 250 pg/ml against the RSV cycle-oxygenase (Fig. 6). D2 or El (20 ug/ml) and chlorColchicine (400 pg/ml), PGD promazine (50 pg/ml) were ai'1 ineffective against thromboxane synthetase.
326
SEPTEMBER 1976
VOL. 12 NO. 3
PROSTAGLANDINS
Indomethacin showed a high degree of selectivity against the cycle-oxygenase (Fig. 6, right panel) : there was a 50% inhibition at 5 ug/ml, a concentration which did Indeed, for 50% inhibition not affect TEA*-synthetase. of the platelet microsome enzyme, indomethacin at 100 ug/ml Other compounds with differential selectivity was needed. were phenylbutazone, which produced 90% inhibition of the cycle-oxygenase at 500 ug/ml without affecting TXA2synthetase, and naproxen which produced >50% inhibition of the cycle-oxygenase at 20 ug/ml without activity on the TXA2-synthetase. Benzydamine was tested because when incubated with rabbit kidney microscnnes and arachidonate at a concentration that caused a 50% inhibition of PGF2 and PGD2 formation, In the same there was an increase in PGE2 format?on (13). system, phenylbutazone did not inhibit PGD2 formation at concentrations that inhibited PGE2 and PGF2u synthesis. Thus, these drugs appeared differentially to inhibit enzymes which transformed PGG2 to other substances such as D2 or E2. The products which arise from prostaglandin endoperoxide transformation have actions which may be beneficial or detriInhibition of cycle-oxygenase (e.g. mental to the organism. with indomethacin) blocks the formation of endoperoxides and Since some of them therefore of all these products also. may be useful to the organism, the methods we have described should allow the discovery of agents with an even greater selectivity for the inhibition of thromboxane synthetase than benzydamine.
SEPTEMBER
1976 VOL. 12 NO. 3
327
PROSTAGLANDINS
REFERENCES 1.
Hamberg, M., Hedqvist, P., Strandberg, K., Svensson, J. and Samuelsson, B. Prostaglandin endoperoxides IV. Effects on smooth muscle, Life Sci. %:451-462, 1975.
2.
Hamberg, M., Svensson, J., Wakabayashi, T. and Samuelsson, 8. Isolation and structure of two prostaglandin endoperoxides that cause platelet aggregation, Proc. Natl. Acad. Sci. USA -71: 345349, 1974.
3.
Needleman, P., Moncada. S., Bunting. S., Vane, J.R., Hamberg, M. and Samuelsson, B. Identification of an enzyme in platelet microsomes which generates thromboxane A2 from prostaglandin endoperoxides, Nature (London) 261: 558-560, 1976. Hamberg, M. and Samuelsson, B. Detection and isolation of an endoperoxide intermediate in prostaglandin biosynthesis, Proc. Natl. Acad. Sci. USA 3: 899-903, 1973.
4.
5.
Nugteren, D.H. and Hazelhof, E. Isolation and properties of intermediates in prostaglandin biosynthesis, Biochim. Biophys. Acta. 1973.
6.
Gornall, A.G., Bardawill, C.J. and David, M.M. Determination of serum proteins by means of the biuret reaction, J. Biol. Chem. 177: 751-766, 1949.
7.
Takeguchi, C., Kohno, E. and Sih, C.J. Mechanism of prostaglandin biosynthesis. I. Characterization and assay of bovine prostaglandin synthetase, Biochemistry, -10: 2372-2376, 1971.
8.
Vane, J.R. A sensitive method for the assay of 5hydroxytryptamine, Br. J. Pharmacol. 12: 344-349, 1957.
9.
Piper, P.J. and Vane, J.R. Release of additional factors in anaphylaxis and its antagonism by antiinflammatory drugs, Nature (London) 223: 29-35, 1969.
10.
Gilmore, N., Vane, J.R. and Wyllie, H.J. Prostaglandins released by the spleen, Nature (London) 218: 1135-1140, 1968.
11.
Eckenfels, A. and Vane, J.R. Prostaglandins, oxygen tension and smooth muscle tone, Br. J. Pharmacol. -45: 451-462, 1972.
328
SEPTEMBER
1976
VOL. 12 NO. 3
PROSTAGLANDINS
12.
Svensson, J., Hamberg, M. and Samuelsson, B. Prostaglandin endoperoxides IX. Characterization of rabbit aorta contracting substance (RCS) from guinea pig lung and human platelets, Acta Physiol. Stand. 2: 222-228 1975.
13.
Blackwell, G.J., Flower, R.J. and Vane, J.R. Some characteristics of the prostaglandin synthesizing system in rabbit kidney microsomes, Biophys. Acta. 398: 178-190 1975.
SEPTEMBER
1976 VOL. 12 NO. 3
329
PROSTAGLANDINS
PG.G2 2fO MIC sv
+
10 M’C
The enzymic generation of endoperoxides (SV + AA) g 1. 0":ihromboxane A2 (AA + SV MIC). The abbreviations used are: SV = ram seminal vesicle microsanes; AA = arachidonic acid; MIC = horse platelet microsomes. The RSV microsomes (300 ug protein) were incubated in 500 ~1 of Tris buffer (50 mM, pH 7.8) with various concentrations of arachidonate for 3 min at room temperature. An aliquot (100 - 200 I.rl) of the reaction was tested on rabbit aorta. Horse platelet microsomes were then added to the reaction mixture for 30 set and a further aliquot was tested on the rabbit aorta. Higher concentrations of arachidonate (100 - 200 ng) produced greater effects and the gain of the amplifier was halEd. Numbers refer to nanograms used.
330
SEPTEMBER 1976
VOL. 12 NO. 3
PROSTAGLANDINS
1
30
00
100
ARACHIDONATE
200 ng
The generation of endoperoxides by RSV microsomes
epended on the concentration of arachidonate used. M-2 The generation of TXAZ by platelet microsomes (SWMIC) was
For each of these experiments, the similarly dependent. gain of the amplifier was adjusted to give equal responses Each point also shows the to a standard dose of PGG2. standard error of the mean (n = 3).
SEPTEMBER
1976 VOL. 12 NO. 3
331
PROSTAGLANDINS
+MlCROSOMES
2
4
6
MINUTES
Spontaneous decay of rabbit aorta contracting Fig. 3. activity following ether extraction after enzyme incubations. Four incubations (n=3 for each) were carried out: a) 500 ng of PGG2 was incubated in 500 ~1 of Tris buffer (50 mM, pH 7.8) for 3 min (left panel.), b) RSV microsomes (340 pg protein) were incubated in 650 l.11 of Tris buffer with arachidonate (5 pg) at 22O C for 3 min (left panel.); c) horse platelet microscxnes (350 ug protein) were incubated in 500 ~1 of Tris buffer with 500 ng of PGG at 00 c for 2 min (right panelo) i d) same as b) WI*&l the addition at 3 min of horse platelet microsomes (350 pg protein) for 30 set before extraction The reactions were stopped by shaking (right panel*). with 2 ml of ice cold ether; the ether phase was removed and dried in a stream of nitrogen. The resultant residue was dissolved in 0.5 ml of Tris buffer (at 22O C) and 100 ~1 aliquots were tested after various times in order to measure the half life of the activity Each point also.shows the standard error on rabbit aorta. of the mean (n = 3).
332
SEPTEMBER
1976
VOL. 12 NO. 3
PROSTAGLANDINS
Fig.
4.
generation
Decay of endoperoxide activity with of prostaqlandin E2-like activity.
concomitant
The figure shows tracings from a rabbit aorta and rat stomach strip superfused in cascade with Krebs' solution at 10 ml/min. For ccnvenience, the tracings have been arranged in three panels. Arachidonate (2 pq) was incubated with RSV micl-osomes (300 ug of protein) in 500 ~1 of Tris buffer (50 mM pH 7.8) at 22O C for 3 min (AA + SV, top panel and right-hand side of bottom panel), 25 min (middle panel) and 60 min (left-hand side of bottom panel). 100 ~1 of the mixture was tested and gave the contractions shown. The remainder of the mixture was then incubated with horse platelet microsomes (380 )1g protein) at O" C for 2 min (+ MIC) and 100 ~1 were immediately tested on the assay tissues.
SEPTEMBER 1976
VOL. 12 NO. 3
333
PROSTAGLANDINS
Fig. 4. continued The endoperoxide activity (rabbit aorta) declined with time. There was a concomitant Increase in PGEq-like activity (rat stomach strip). As the endopero;ide activity declined, the ability to form TXA2 (increased activity on rabbit aorta) by addition of horse platelet microsomes also decreased. Calibrating doses of PGG2 and PGE2 are shown in the top two panels (right hand side).
I
I-
I
I
I
I
0
1
2
3’4
I
I
I
5 6 MINUTES
I
I
I
7
8
9
I
10
“J
/r 20
Half life of extracted TXA2 at different Fig. 5. PGG2 (50 ng) was incubated with horse temperatures. platelet microsomes (380 ug protein) for 2 mln in an ice bath. The reaction was then stopped by shaking with 1 ml The ether phase was removed and dried of ice cold ether. The dried extract was resuspended in a stream of nitrogen. in 500 ul of Tris buffer at O" C, 22O C or 37O C and an aliquot (100 ~1) was tested for activity on rabbit aorta at different times. The half life of TXA2 at each temperature (see text) was calculated by comparisons with standard volume/response curves for extracted TXA2.
334
SEPTEMBER 1976
VOL. 12 NO. 3
PROSTAGLANDINS
BENZYDAMINE
us/ml
INDOMETHACIN
Comparison of inhibitian of TXA2-synthetase and Fig. 6. cycle-oxygenase by benzydamine and indomethacin. Horse platelet microsomes (MIC) (154 ug/protein) were preincubated with inhibitor in 500 ul of Tris buffer (50 mM, The reaction mixture was pH 7.8) for 3 min at 220 C. transferred to an ice bath and PGG2 (100 ng) was added. An aliquot (50 ~1) was bioassayed after 2 min. RSV microsomes (300 ug) were inubated at 220 C with inhibitor Arachidonate (2 vg) for 3 min in Tris buffer (400 ~1). was then added and after a further 3 min 100 ~1 was tested for endoperoxide formation. Values are the means 2 standard error of mean (n = 4).
SEPTEMBER
1976
VOL. 12 NO. 3
335