MSU LIBRARIES \' RETURNING MATERIALS: Place in book drop to remove this checkout from your record. FINES wiII be charged if book is returned after the date stamped be10w. THE PURIFICATION, QUANTITATION AND IMMUNOFLUORESCENT LOCALIZATION OF THE PGIZ FORMING ENZYMES, PGH SYNTHASE AND PGIZ SYNTHASE, WITH MONOCLONAL ANTIBODIES By David Lee Dewitt A DISSERTATION Submitted to Michigan State University in partial fulfuillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Biochemistry 1982 ABSTRACT THE PURIFICATION, QUANTITATION AND IMMUNOFLUORESCENT LOCALIZATION OF THE PGIZ FORMING ENZYMES, PGH SYNTHASE AND PGIZ SYNTHASE, WITH MONOCLONAL ANTIBODIES By David Lee Dewitt Four cloned hybridoma cell lines (designated 519:1,3,5, and 7) have been isolated which secrete antibodies against the PGH synthase enzyme. The antibodies, when coupled with Protein-A bearing Staphlococcus aureus cells, precipitate PGH synthase activity and a single protein with a monomer molecular weight identical to the purified PGH synthase. Two of the antibodies which bind different antigenic sites on the PGH synthase have been used in a immunoradiometric assay for the enzyme which is 102-104 times more sensitive than the most sensitive polarographic assay for PGH synthase. Two cloned hybridoma cell lines (designated 153-1 and jsgr3) which secrete antibodies against the P612 synthase have also been isolated. These antibodies precipitate PGIZ synthase activity and a single protein with a monomer molecular weight of 52,000 daltons. Active PGIZ synthase was purified using an Ig61 (153:3) Affigel-ID immunoaffinity column and found to contain a protoporphyrin IX heme prosthetic group. The purified enzyme is inactivated by PGHZ and 13-hydroperoxy-linoleic acid. The P612 synthase heme spectrum is bleached during inactivation. Azo Analogue I, a substrate analogue of PGHZ prevents inactivation and bleaching of the heme spectrum caused by PGHZ and 13-hydroperoxy-linoleic acid, but only at concentrations which inhibit P612 synthase activity. An imnunoradiometric assay, analogous to the PGH synthase assay, has also been developed for the P612 synthase. Using the two immunoradiometric assays, the concentrations of PGH synthase and P612 synthase in the cell layers of bovine aorta were determined. Smooth muscle and endothelial cells contain roughly equal amounts of P612 synthase; endothelial cells, however, contain 20 fold more PGH synthase than smooth muscle cells. Immunofluorescent staining with the isg antibodies indicates that PGIZ synthase is also present in most types of extra-vascular smooth muscle. TABLE OF CONTENTS Page LISt Of Tablesooooooooooooooooooooooooooooooooooooooooooooooooooooii List Of Figures.0000000000.000000000000000cococoooooooooocooooooooiii AbbreV‘ationSOOOOOOO0.0.0.0....0.0.0.0000...OCOOOOOOOOOOOOOOOOOOOCV]. Chapter I. Literature ReVieWOOOOOOOOO00.00.000.000...OOOOOOOOOOOOOOOI II. Isolation and Characterization of Monoclonal Antibodies Against PGH Synthase: An Immunoradiometric Assay.................................19 Materials and Methods...................................20 Resu‘tSOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO0.0.00.00000000031 Discussion........................... ....... ............50 III. Monoclonal Antibodies Against PGIZ Synthase: An Immunoradiometric Assay for Quantitating the Enzymes................. ....... .........................52 Materials and Methods...................................53 Results.................................................64 Discuss1OHCCOOOOOOOOOOOO ...... O. ...... O ....... 0.00.00.0075 Page IV. Purification and Characterization of the P612 Synthase..................................... ...... .....77 Materials and Methods...................................78 Results..... ...... ........ ...... . ....... . ............... 82 Discussion. .......... .. ........... . ........... . ...... ..110 V. Quantitation and Localization of PGH Synthase and P612 Synthase........... ........ . ....... ...............114 Materials and Methods..................................115 Results............... .......... .......................119 Discussion...... .......... .................... ......... 132 Bib]109raphy...OOOOOOOOOOOOOOOOOOOOOOOOOOOIOOOOOOOOOO0.00.00.00.0137 LIST OF TABLES Table Page 1 Analysis of monoclonal antibodies against PGH synthase.......................................... ...... 37 2 Factors influencing immunoradiometric assay for PGH synthase from detergent solubilized sheep vesicular gland microsomes........................................49 3 Partial purification of P612 Synthase...................57 4 Competitive binding between antibodies secreted by jsgrl and isgf3.........................................67 5 P612 synthase activity recovery from 153:3 and non-specific antibody Affigel-lo affinity columns.......83 6 A20 Analogue I (9,11-Azo-prosta-5,13-dienoic acid) protects P612 synthase from lipid peroxide inactivation...........................................108 7 A20 Analogue I (9,11-Azo-prosta-5,13-dinoic acid) protects PGIZ synthase from substrate inactivation...........................................109 8 Distribution of PGH synthase and P612 synthase immunoreactivity in muscle.............................120 9 Concentrations of PGH synthase and P612 synthase in cell layers of bovine aorta.........................127 ii LIST OF FIGURES Figure Page 1 An overview of Prostaglandin Biosynthesis.... ..... .......3 2 Immunoprecipitation of PGH synthase from 125I-sheep vesicular gland microsomes...................34 3 Precipitation of PGH synthase from intact and solubolized sheep vesicular gland microsomes using monoclonal antibody - S, aggggs complexes...............36 4 Illustration of the interactions involved in the immunoradiometric assay for PGH synthase................42 S Immunoradiometric assay using purified sheep VESiCUIar gland PGH synthaseOOOOOOOOOOOOOO0.0.0.0000000044 6 Comparisons of the activities of the PGH synthase from different species in the immunoradiometric assaYOOOOOOOOOOOOOOOO0.0...000.000.000.000...000.000000047 7 Precipitation of P612 synthase from 1251-m1crosome500000.000.000.000000000000....one...00.0066 8 Illustration of the interactions involved in the immunoradiometric assay for P612 synthase..... ..... .....71 9 Immunoradiometric assay of P612 synthase from SO1Ub0112ed bov1ne aortaOOOOOOOOOOOOOOOO0.00.00.00.0000073 Figure 10 11 12 13 14 15 16 17 18 19 Page SDS-Polyacrylamide gel electrophoresis of eluates from 1961 (_1°__'5_n-3)-Affigel-10............................85 SDS-Polyacrylamide gel electrophoresis of eluates from 1961 (control)-Affigel-10..........................87 Specific Activities of P612 synthase eluted from the 1961 (jggy3)-Affigel Column. ........................ 89 Protein elution profiles for 1961 (isn-3)-Affigel-10 and 1961 (control)-Affigel-10 columns.. ..... ............................... ........... 92 Coelution of P612 synthase activity and heme absorbance from 1961 (isnf3)-Affigel-IO column..........94 Carbon monoxide-dithionate reduced difference spectra of purified PGIZ synthase.......................98 Inhibition of purified P612 synthase by 13-hydroperoxy-linoleic acid...........................101 Effects of 13-hydroperoxy-linoleic acid on the absorbtion spectrum of purified PGIZ synthase: Protection from bleaching by Azo Analogue I............103 Effect of PGHZ on the absorbtion spectrum of P612 synthase: Protection from bleaching by Azo Analogue I.............................................105 Immunocytofluorescent localization of PGH synthase in an artery in a cross section of ovine myomEtriumooooooooo ...... ..OOOOOOOOOOOOOOOOOOOOO0.0.0.0123 iv Figure 20 21 22 Page Immunofluorescent localization of P612 synthase in an artery in a cross section of ovine myometriun............................... ............ ..125 Immunocytofluorescent localization of P612 synthase in rabbit urinary bladder.............. ....... 129 Control Immunofluorescent staining of rabbit urinary bladder with MOPC-Zl mouse 1961................131 ABBREVIATIONS The abbreviations used are: PGHZ, 15-hydroxy-9a,Ila-peroxido-prosta-S,13-dienoic acid; PGIZ, 9-deoxy-6,9e-epoxy-11a,IS-dihydroxy-prota-S,I3-dienoic acid; PGan, 9a,11e,15-trihydroxy-prosta-S,13-dienoic aicd; 6-keto-PGF17, 9a,11a,15-trihydroxy-G-keto-prosta-IB enoic acid; 505, sodium dodecyl sulfate; DMEM, Dulbecco's modified Eagle medium; HEPES, N-Z-hydroxy-ethylpeperazine-N'-2-ethanesulfonic acid; MES, 2-(N-morpholino)ethane-sulfonic acid; 13-HPLA, 13-hydroperoxy-9 gig, 11 trans-octadecadienocid acid. vi LITERATURE REVIEW Introduction In late 1976 Vane and coworkers discovered an enzymatic product of the action of pig aorta microsomes on protaglandin endoperoxides having pharmacological effects different from those of other known prostaglandins (Moncada gt_gl. 1976). This new compound, which they designated PGX, was inactivated by boiling for 15 sec, was more potent at inhibiting platelet aggregation than any other known substance, and caused relaxation of rabbit mesenteric and coeliac arteries. This compound was subsequently identified (Johnson gt_gl., 1976) as 9-deoxy-6,9 a epoxy-AS-PGFIQ (Figure 1) and was given the trivial name prostacyclin, now abbreviated P612. Biosynthesis of P612 Prostacyclin is formed from arachidonic acid by the pathway illustrated in Figure 1. The major control point in the synthesis of P812 and other prostaglandins is the hydrolysis of arachidonic acid from phospholipids. Release of arachidonic acid in cells that form P612 can be elicited by a variety of hormonal and proteolytic stimuli including trypsin, thrombin (Weksler gt al. 1978), bradykinin, angiotension II (Needleman gt 31., 1978; Hong, 1980) and platelet derived growth factor (Coughlin gt 31., 1980). Nonphysiological agents such as the calcium ionophore, A23187, or nitroglycerin can also cause Figure 1. An overview of prostaglandin biosynthesis. I'— 3901x033, j 904 I‘ 33'“39Axc°1a‘3 ZOOU n A q U.ZOQ.IU<¢< HUD Um<¢3 Uecuurdva «(Xh .9: 33:25 .3.» v gogllvv 2.5529. 33%.... l 03: So. I U Get 5o§l>\flr oz 3.8.. 3 #3505 1!. KO‘ 8000 $9 $9? \ z z ¢ 60‘ 800 o I 'Hsd 8.3m -23.. c O: as"mus 2000 2000 23430....wm 0.2mduruoczu :o :38 1 8:82.- omofw :w-¢ozoo.zu:.u< uo 0.01% 505 in assay enzyme buffer 3. 0.1 M or 1 M NaCl in assay 3. 2.2 fl_urea in assay buffer buffer 4. 0, 0.1, 1% Tween 20 in assay buffer 5. Allowing enzyme to stand 48 h at 4° 50 DISCUSSION PGH synthase catalyzes the formation of the prostaglandin endoperoxide, PGHZ, from arachidonic acid and oxygen. Fluctuations in tissue concentrations of PGH synthase have been shown to occur in rat Graafian follicles in response to luteinizing hormone (Clark gt 31., 1978), in the ovine (Huslig gt al., 1979) and guinea pig uterus (Poyser, 1979) during the estrous cycle and in the hydronephrotic rabbit kidney in response to perfusion ex 1119 (Morrison gt_gl. 1978). The diminished production of P612 by arteries of rabbits subjected to high fat diets (Dembinska-Kiec gt_gl., 1977) and the increased PGEZ formation by kidneys of rabbits fed low salt diets (Stahl gt 31., 1979) could also be due to alterations in the levels of PGH synthase. These observations from several laboratories indicate that changes in PGH synthase levels can play a role in regulating the rates of prosta- glandin formation in both normal and pathological situations. There are a number of methods for assaying PGH synthase enzyme activity in- cluding the polarographic assay, the use of radioactive fatty acid sub- strates and measurement of labeled products and the use of unlabeled fatty acids and measurement of products by radioimmunoassays. There does, however, exist a need for methods to quantitate changes in enzyme protein levels. In this chapter, we have described the preparation of four different monoclonal antibodies against the PGH synthase and the use of iodinated monoclonal antibodies for quantitating PGH synthase El protein concentrations in tissue extracts using an immunoradiometric assay. This immunoradiometric assay is 102-104 times as sensitive as the most accurate and sensitive polarographic assay for enzyme activity and should be useful for further studies on the regulation of PGH synthase synthesis. CHAPTER III MONOCLONAL ANTIBODIES AGAINST PGIZ SYNTHASE: AN IMMUNORADIOMETRIC ASSAY FOR QUANTITATING THE ENZYME In this chapter I describe the preparation of two hybridoma lines producing monoclonal antibodies against two different antigenic sites on the P612 synthase enzyme molecule. The hybridoma lines were derived from mice immunized with a partially purified preparation of P612 synthase from bovine aorta. Using these antibodies I have developed an immunoradiometric assay with which to quantitate PGIZ synthase protein concentrations. This assay is 50-100 times more sensitive than conventional radiochromatographic enzyme activity assays. 52 53 METHODS Materials Triton-X-100 was from Calbiochem. DEAF-cellulose (DE-52) was from Whatman. Thin layer chromatography plates (Silica Gel 60, 0.25 mm) were from Analtech. PGan and 6-keto-PGF1G standards were purchased from Upjohn Diagnostics. Cellophane gel backing and molecular weight standards for polyacrylamide gel electrophoresis was obtained from BioRad Laboratories, Inc. All other materials were obtained from the sources listed in chapter II. Animals and Tissues Female ICR Swiss white mice, 4-6 weeks old used for immunizations and for isolation of splenocytes were obtained from Harlan Laboratories. Bovine aorta was obtained fresh at slaughter from Michigan State University Meats Laboratory and stored at -80°. Preparation of L3H]PGH7 [3HJPGH2 was synthesized from [5,6,8,9,11,12,14,153-H-(N)] arachidonic acid (100 Ci/mmole; diluted as necessary to a specific activity of 10-20 Ci/mole (ca. 10,000-20,000 cpm/nmole) with unlabeled arachidonic acid) using a modification of the procedure of Hamberg gt .31. (1974); 85 ml of 02 saturated 0.1 M_sodium phosphate, pH 7.4 containing 500 BE phenol was mixed with 15 ml of freshly prepared sheep vesicular gland microsomes (10 mg protein/ml) and the sample equilibrated to 25°. [3H] arachidonic acid (1-2.5 mg) in 0.4 ml of 54 ethanol was added, and the sample was incubated for 2 min and then acidified to pH 3.0 with 6 M_HCl. The sample was extracted sequentially with 2-300 ml portions of ether cooled to 0° in 500 ml separatory funnels. The combined ether extracts were cooled to -78° and the ice was removed by filtration through glass wool. The resulting ether layer was evaporated to dryness on a rotary evaporator at 4° and the residue was dissolved in 20 ml of hexane/ether (8/2; v/v). Silicic acid chromatography was performed at 4° as described by Hamberg gt fl. (1974). [3H]PGH2, which was eluted with hexane/ether (4/6; v/v), was evaporated to dryness, redissolved in anhydrous acetone and stored in 100 pl (1,000 nmole) aliquots at -80°. Purity of [3HJPGH2 was routinely >90% when tested by thin-layer chromatography on Silica Gel G chromatography plates in ether/acetic acid (loo/0.2; v/v; 4°); PGHZ migrates just slightly ahead of PGBZ in this system. Enzyme ActivityiAssay of PGIZ_synthase Aliquots of [3H]PGH2 (50 nmoles in 5 pl of acetone) were pipetted into individual samples of enzyme in 0.1 ml of Tris Buffer (0.1 M tris-chloride), pH 8.0, containing 0.5% Triton X-100 and 10‘4‘M_Flurbiprofen). Incubations were performed at 24° for one minute and the reactions stopped by adding 50 pl of a freshly prepared solution of FeClz (6 mg/ml). The protein was removed by addition of 0.3 ml of acetone and centrifugation at 1,500 g for 5 min (Salmon and Flower, 1982). The supernatants were then transferred to new tubes, acidified with 50 pl of 0.2 M HCl and 1 ml of CHCl3 was added. After mixing thoroughly, the tubes were centrifuged to separate the phases. The aqueous layer was aspirated and the organic layer dried under N2. 55 The residue was redissolved in 75 pl CHCl3 and spotted on Silica Gel 60 250 pM thin layer chromatography plates which were then developed twice in the organic phase of ethyl acetateztrimethylpentane:acetic acidzH20 (110:50:20:100; v/v/v/v). The region corresponding to standard 6-keto-PGF1a, as well as the rest of each vertical lane, was scraped into scintillation vials and counted. P612 formation was calculated as the fraction of total radioactivity chromatographing with the 6-keto-PGF16 standard multiplied by the total substrate (5 nmoles) added to each assay mixture. PGIZ synthase does not exhibit standard Michaelis-Menten kinetics. The assay is performed as a matter of convenience for one minute, although at the 50 pM substrate concentration, the P612 synthase is always inactivated before one minute (Watanabe 1979). Thus, the rates reported below are not initial rates. Nevertheless, these rates are valid for comparing enzyme activities when the assays are run at concentrations of PGIZ synthase that are linearly related to the amount of P612 formation per min. One unit of activity is defined as the amount of enzyme that will catalyze the formation of 1 nmole of P612 per min under the standard assay conditions. PGIZ Synthase Purification Bovine aorta obtained fresh at slaughter was frozen immediately on dry ice, then stored at -80°. The abdominal region of the aorta beginning 15-20 cm from the heart is easiest to homogenize. Aorta was frozen in liquid nitrogen, shattered into small pieces (ca. 2 cm3) with a hammer and homogenized in 2-3 volumes of ice cold 0.1 M Tris-chloride, pH 8.0, containing 10‘4 M Flurbiprofen with a Polytron (Brinkman) homogenizer. Care was taken to maintain the buffer 56 temperature below 5° during homogenization. The homogenate was centrifuged at 10,000 x g for 10 min, and the resulting supernatant was centrifuged for 35 min at 200,000 x g to collect the microsomal pellet. When stored at -80° the pellets retain their PGIZ synthase activity for 2-3 months. Further purification of PGIZ synthase was performed using a modification of the method of Wlodawer and Hammarstrom (1979). Microsomal pellets (0.5 g) from 25-30 g of tissue were resuspended with a glass homogenizer in 10 ml of Tris Buffer. This homogenate was centrifuged at 200,000 x g for 35 min. The washed pellet was resuspended in 10 ml of 10 mM sodium phosphate, pH 7.4 containing 0.5% Triton-X-IOO and again centrifuged at 200,000 x g for 35 min. The supernatant was removed and the solubilized PGIZ synthase was applied to a DE-SZ cellulose column (2 x 8 cm) equilibrated with 10 mM sodium phosphate, pH 7.4 containing 0.1% Triton-X-100; the column was then washed with 60 ml of the equilibration buffer. P612 synthase was eluted with 0.2 M_sodium phosphate, pH 7.4, containing 0.1% Triton-X-100. The specific activity of the P612 synthase at this step ranged from 100-225 with an average of 150 units/mg protein/min. This represents a purification of approximately 10 fold. A summary of a typical purification is presented in Table 3. Immunization Protocol PGIZ synthase purified through DE-52 chromatography was used to immunize outbred 4-6 week old female ICR Swiss white mice. Approximately 250 pg of protein (average specific activity of 150 units/mg protein) in 0.2 ml of 0.2 M_phosphate, pH 7.4 containing 0.1% Triton-X-IOO was emulsified by sonication with 200 pl of complete 57 caspoo ~.o flaw cHH me. Nm-mo sac» peas.“ Ac mmEom nocope om~_ppn:_om kc m.“ aae so mo.H Seeeaecdasm a x oo.oo~ Am 65 In .ae_co_ed-m_su z H.9V N awe mm o.m :o_m:mam:m PQEOmocu_z AN wagon uo~_:mmoeoz mo H goes NH m.¢ Seeueecaaam a x ooo.o~ AH =o_uaa_cwc=a aww>_ud< co e_euecmlme\=_s A_e\mev wwwm copped_e_c=a upon xcm>ouom a \aamuauoamxum =o_umcacmu:oo mo mm—osc cpmuocm »u_>eud< a_c_daam emaeuexm NHua Lo ee_uea_esc=a _e_uces .m udm5 min) before centrifugation increase the degree of nonspecific precipitation of 125I to unacceptable levels. After centrifugation, the supernatant was removed by aspiration and the 63 pellets were washed once in 0.5 ml of the solubilization buffer. The washed cell pellets present in the 6 x 50 mm test tubes were placed in vials and 1251 determined by counting on a Beckman Biogamma counter. 64 RESULTS Preparation and characterization of monoclonal antibodies against P612 synthase. Outbred white mice were immunized with a crude preparation of bovine aorta microsomes (containing P612 synthase activity) which had been solubilized with 0.5% Triton X-100. Splenic lymphocytes from immunized mice were fused with the SP2/0-Agl4 myeloma cell line. Two hybrid lines (13371 and 153:3) that secrete monoclonal antibodies capable of causing immunoprecipitation of P612 synthase activity were selected and cloned from a total of about 1200 hybridomas tested. Immunoglobulins secreted by both iSflfl and igg—B are of the mouse I961 subclass as determined by Ouchterlony double diffusion analysis against allotype specific antisera. Antibodies secreted by both 1§gel and 1§3r3 react with PGIZ synthases from mouse, rat, rabbit, sheep, dog and cow as determined either by immunoprecipitation of enzyme activity from solubilized aortic microsomes or immunocytochemical staining of arterial smooth muscle in sections of renal cortex. To define the antigen precipitated by antibodies secreted by 133-1 and j§373, antibody secreted into culture media from iéflfl, iSflf3, and £§g¢4 (a control line secreting a nonspecific 1961) were coupled to attenuated S. ggrggg cells. Each antibodygS, Sprggg complex was then incubated with a mixture of 125I-labeled proteins prepared from bovine aortic microsomes. Immunoprecipitates were analyzed by SDS Figure 7. 65 Specificity of monoclonal antibodies to P612 synthase. Aortic microsomes were solubilized and then iodinated. .§- aureus cells complexed with 1965 secreted by 13341,.133-3 or two non-specific mouse hybrid lines were incubated with 1251-iabeied microsomes from bovine aorta (Lane a). The finnunoprecipitates obtained were analyzed by SDS-polyacrylamide gel electrophoresis: 1961 (isn-1)-S. aureus (Lane b); IgGl (isn-3)-S. aureus (fine c); IEGI—TE o-3)-S, aureus (Eafie d); 1961 (1&274)<§- aureus Lane e) and rabbit anti-mouse IgGfiS. aureus (Lane f). 666 ' , 92,500 '- 93:. .2 66,200 -‘ 45,000 - 5n 3|,000- TABLE 4. PGIZ Synthase Activity IngS. aureus Microsome pmole 6-keto- Experiment Conju§3t3__ Pretreatment PGFlaLEEll§E__ a isn-1 isn—l 0 b " isn-3 750 c " cyo-l 1500 d " none 2550 e isn-3 isn-l 0 f " isn-3 0 g " cyo-l 1200 h " none 3250 i cyo-3 isn-l 0 j “ isn-3 k " cyo-1 0 l " none 0 Competitive Binding Between Antibodies Secreted by isn-l and isn-3 .§- aureus coupled to the indicated antibodies were mixed with microsomes pretreated with the indicated antibodies. .§- aureus cells were precipitated by centrifugation, resuspended and the P612 synthase activity measured. 68 polyacrylamide gel electrophoresis and autoradiography (Figure 7). Although a large number of radioiodinated proteins were present in the 125I-labeled solubilized aortic microsomes, only one of these iodinated species was precipitated by mouse immunoglobulins isolated from culture media from both 153:1 and jgge3; moreover, the iodinated proteins precipitated by each of the antibodies had the same electrophoretic mobilities (ca. 52,000 daltons). No 125I-labeled proteins were detected in the control (Eggr4 and 23S) precipitates. The results indicate that iggrl and 152:3 each secrete an 1961 which when bound to S. gggggg cells will precipitate (a) PGIZ synthase activity and (b) a single, 52,000 dalton protein. This is strong correlative evidence that the 52,000 dalton protein is a subunit of P012 synthase particularly because the antibodies secreted by iggfl and j§2e3 are directed against nonidentical antigenic determinants (i.e. binding of 1251-iabeied 1961 (i_s_g-3) to bovine aortic microsomes is blocked by preincubation with unlabeled IgGl secreted by iéfl‘3 but not by 1961 secreted by iéflfl (see below)). Antibodies Secreted by isn-l and isn-3 Bind Different Antigenic Sites Competitive binding experiments were performed to determine whether antibodies secreted by iggrl and j§g73 bind the same antigenic determinant. Solubilized microsomes were pretreated with 196 secreted by either igg:1 or 1§373 and each sample then incubated with 1961 (iggrl or 153:3)1S..ggggg§ complexes. As expected, incubation of microsomes with 1961 (iggel) or (SSE-3) completely inhibited precipitation of P612 synthase activity by S. Eggggs to which the same antibody was conjugated (Table 4; a,f). Moreover, microsomes pretreated with Ing (15373) S, aureus complexes (Table 4b) were 69 precipitated by 1961 (ignrl) S, gggggg complexes. However, microsomes pretreated with 1961 (iSflrl) were no precipitated by IgGl (iggg3) S, gprgg§_complexes (Table 4e). These results indicate that 1961 secreted by iéflf3 cannot bind the microsomal PGIZ synthase if 1961 secreted by iggel binds the enzyme first. Similar results were obtained when the antibodies secreted by jégrl and igg—3 were used to quantitate solubilized P612 synthase activity in the immunoradiometric assay (see below). There are several possible explanations for this anamolous binding behavior. The simplest is that although IgGl secreted by iSfl-l and 1§2f3 bind non-overlapping determinants, and binding of 1961 (iéfl'l) interferes sterically with the access of 1961 (133-3) to its epitope. Immunoradiometric Assay of P012 Synthase The elements of the immunoradiometric assay for P612 synthase are diagrammed in Figure 8. A positive linear relationship between precipitated 125I-IgG1 (jgge3) and added PGIZ synthase activity exists over the range of 0.005-0.05 units of activity when using I901 (i§g¢1)eS,‘ggggg§ cells as the precipitating complex (Figure 9); the slope is equal to 300,000 cpm precipitated per unit of P612 synthase. This assay is 50-100 times more sensitive than the enzyme activity assays. When the 1961 (isg-3)1S..ggggg§ complex is substituted as a control for 1961 (jggyl)-S, Sprggg cells, no 1251 above background is precipitated. A similar lack of precipitation of 125I is observed when non-specific mouse immunoglobulins are conjugated to S, ggrggg cells and substituted for Ing (igg:1)1S. aureus. 70 Figure 8. Illustration of the interaction involved in the immunoradiometric assay for P612 synthase. 7l 4.. \ mica / Eon e_ .3. 8.5.28 > as. con .-ee /2 =8 m \ < 2995 Figure 9. 72 Immunoradiometric assay of PGIZ synthase using solubilized bovine aortic microsones. PGI synthase was incubated with 1251-1961 (igg—B) for 3 min, then S. aureus cells conjugated to 1961 secreted by either iggrl (|———l) or igge3 (0———0) were added. The S. aureus cells were pelleted by centrifugation and washed and precipitated 251 quantitated. Results are averages of triplicates (0——-0); error bars t 3.0. 73 30. x mes mmm Nae m v m N _ _ _ O O # _ I 20:83 WEAK Om BLVlldIOEHd SflBHfiV '8 Nl I93. 74 This immunoradiometric assay provides a simple, sensitive and highly specific method for quantitating PGIZ synthase. The method should be useful for measuring changes in PGIZ synthase protein concentrations in tissues during physiological stresses such as aging and the development of atherosclerosis. 75 DISCUSSION Upon isolation of hybridoma clones producing antibody that precipitated PGIZ synthase activity, I was faced with the problem of determining the selectivity of the antibodies with no information abou t the physical properties of the P612 synthase enzyme. The results presented in this chapter provide evidence that immunoglobulins secreted by iggel and 152:3 both precipitate PGIZ synthase activity and the same, single 52,000 dalton protein monomer. Moreover, the competitive binding data suggest that the two antibodies interact with different determinants. Isolation of active PGIZ synthase by immunoaffinity chromatography (Chapter 4) has also indicated that a 52,000 dalton protein monomer can be co-eluted with P012 synthase enzyme activity. These data provide strong evidence that the antibodies secreted by 135-1 and 153:3 are directed against PGIZ synthase, and that this protein has a subunit molecular weight of 52,000 daltons. The immunoradiometric assay for P612 synthase may prove to be more versatile than the immunoradiometric assay for PGH synthase because the antibodies secreted by iSfl-l and igg-3 cross react with P012 synthase from a broad range of species. However, this assay is still in a developmental stage. For example, I currently standardize the assay using impure microsomal P612 synthase from bovine aorta. Once enough pure protein has been isolated, the assay can be standardized to the mass of P612 synthase measured. 76 Immunoradiometric assays such as those for the PGH synthase and the PGIZ synthase are clearly superior to single antibody radioimmunoassays. The reason for this is their increased specificity. The interference by substances that cross react with the antibodies in the entire assay is equal to the product of the percentage cross reaction of one antibody multiplied by the percentage cross reaction of the second antibody. In practice, the specificity of single monoclonal antibodies is so high that a double monoclonal antibody assay should be essentially free from interference. CHAPTER IV PURIFICATION AND CHARACTERIZATION OF THE PGIz SYNTHASE In this chapter I describe the purification of P612 synthase by immunoaffinity chromatography and report some spectral and kinetic properties of the purified enzyme. 77 78 MATERIALS Materials Affigel-IO was obtained from BioRad Laboratories, Inc. Linoleic acid was from NuChek Preps. Type IV lipoxygenase was purchased from Sigma Chemical Co. All other materials were procured from sources described in Chapters II and III. Preparation of anti-P619 synthase immunoaffinity columns The 1961 (153:3) AffiGel-ID column used in the purification of P612 synthase was prepared as follows (Staeheln gt g1. 1981): 3.0 mg of 1961 (purified from j§g73 culture media by chromatography on Protein A-Sepharose as described in Chapter III) was dissolved, in 3.0 ml of 0.1 M HEPES, pH 7.5, and incubated at 4° for 12 hr with 6 ml of AffiGel-IO. The gel was collected by centrifugation and 6 ml of 0.1 M ethanolamine was added. After 1 hr, the gel was packed in a small chromatography column prepared in a 10 ml plastic syringe and washed exhaustively with Tris Buffer. We have used the columns for several months with no apparent loss of binding capacity. Columns are usually stored in Tris Buffer. Purification of PGIz_synthase Microsomal pellets (0.3 g) were prepared from bovine aorta as described in Chapter III and were resuspended by homogenization in 15 ml of Tris Buffer, (.1 M_tris-chloride, pH 8.0, containing, .5% 79 Triton-X-100 and 1 x 10‘4 M_flurbiprotein) containing Z‘mfl 2-mercaptoethanol. This suspension was centrifuged at 48,000 x g for 30 min, and the supernatant containing the solubilized enzyme was applied to an I961 (igge3)-Affi6el-10 column equilibrated with Tris Buffer. Because the flow rate is rapid (5 ml/min) the enzyme solution was rechromatographed 5 times to maximize binding of PGI2 synthase activity. The column was then washed with 10 volumes of Tris Buffer. PGIZ synthase activity was eluted with 0.1 M MES-(Z-(N-morpholino)ethane-sulfonic acid), pH 6.0, containing 0.5% Triton-X—IOO, 1.M_NaCl and 2.mM_2-mercaptoethanol (pH 6.0 MES-NaCl). Seven 5 ml fractions were collected into 0.75 ml of 1.0 M Tris-chloride, pH 8.0; this adjusted the pH of the eluate to approximately 8.0. A control AffiGel-IO column consisting of 6 ml of resin to which was coupled 3 mg of non-specific mouse IgG1 was run in parallel with the 1661 (jége3)-AffiGel-10 column. Solid phase assay of P012 synthase For certain inactivation and protection studies involving enzyme inhibitors, PGIZ synthase was bound to S. ggrggg cells, via IgG (iggeB). The 1961(iéflf3) S, ggrgggePGIz synthase complex was incubated with the inhibitors and quickly separated from these agents by centrifugation. By measuring the PGIz synthase activity of the washed S. ggrggg pellets, it was possible to determine whether the inhibitors caused irreversible inactivation of the enzyme. The enzyme was immobilized by attachment to S. gggggg using the following protocol. A 10% suspension of rabbit anti-mouse IgGgS, Sggggg cells (1 ml) prepared as described in Chapter III were incubated with 100 pg of 1961 (igg—3) for 12 hrs and the cells collected by centrifugation. 80 After a single wash with Tris Buffer the resulting 10% suspension of 1961 (igg—3)4S, ggrggg cells (1 ml) was mixed with an equal volume of solubilized bovine aortic microsomes (80 units/ml) and incubated at 4° for 15 min. 1961 (jggr3)gS,Igg§gg§ cojugates will quantitatively bind small amounts of P612 synthase, but excess enzyme was used in these experiments maximize binding of P612 synthase to the S. ggggus. The S. Iggrggg cells were then pelleted by centrifugation, the pellet washed once with Tris Buffer, and the cells resuspended in 10 volumes of Tris Buffer. Under these conditions, approximately 12 units of P612 synthase activity were bound per ml of the 1961 (iggy3)gS, gggggg suspension. Incubations of immobilized PGIZ synthase with inhibitors were performed as follows. To 0.1 ml of P612 synthase S. Sggggs suspension was added 0.1 ml of Tris Buffer and 10 pl of a solution containing the effector. The samples were allowed to incubate at 24° for various times and the S. gggggg cells were then collected by centrifugation for 1 min and the supernatant removed. The cell pellets were washed with 0.2 ml of Tris Buffer to remove traces of inhibitors and then resuspended in 0.2 ml of Tris Buffer and assayed with 50 nmole of [3HJPGH2 (25 E!) to determine the amount of PGIZ synthase activity remaining. No more than four samples were tested at a single time to insure uniformity in the handling of the samples. Preparation of 13-hydroperoxy-linoleic acid (13-HP-linoleic acid) Arachidonic acid (10 mg) was suspended by sonication in 49 ml of 0.1 M sodium borate buffer, pH 9.0; 1 ml of lipoxygenase (Type IV, Sigma, 4.45 106 units) was added, and the reaction was allowed to proceed for 5 min at 4° and stopped by acidification to pH 4.0 with 5 N 8) HCl. The lipid was extracted at 24° with two 50 ml portions of benzene. The benzene extracts were pooled and evaporated to dryness under nitrogen. The residue was redissolved in hexane:ether (95:5; v/v) and applied to a 3-4 9 silicic acid column. The column was washed with 150 ml of hexane:ether (95:5) and the 13-HP linoleic acid was eluted with hexane:ether (85:15; v/v). Fractions containing the hydroperoxy acid were pooled and evaporated under reduced pressure at 4°. The residue was redissolved in methanol and stored at -80°. Purity of the 13-HP linoleic acid was determined by thin-layer chromatography on silica gel 60 in diethyl ether/petroleum ether/petroleum ether/acetic acid (50:50:5; v/v/v). 13-HP-linoleic acid has an Rf of .65 in this system. Purity was estimated to be greater than 95%. ,Spectrophotometric analysis All spectra were recorded on a Aminco DW-2 a dual beam/dual wavelength recording spectrophotometer interfaced with a Aminco Midan T analogue computer. Identification and quantitation of the heme content of the preparation was determined by reduced minus oxidized difference spectra of pyridine haemochrome derivatives (Falk, 1964). 82 Results Purification of PG12 Synthasegby,Immunoaffinity Chromatography Solubilized bovine aortic microsomes containing 1080 units of PGIZ synthase activity were applied to both an 1961 (ign-3) AffiGel-IO column and an 1901 (control)-AffiGel-10 column (Table 5). About 90% of the P012 synthase activity was bound by the IgGl (iggf3)-AffiGel-10 column and no PG12 synthase activity was detected in the Tris Buffer washes. Upon elution of the 1901 (133:3)-Affi6el 10 column with pH 6.0-MES-NaCl Buffer, 24% of the starting activity was recovered. In contrast, the Ing (control)-AffiGel-10 column bound less than 10% of the original P012 synthase activity; 100% of the enzyme activity was recovered in the initial elutate and in the Tris Buffer washes; moreover, no PGIZ synthase activity was eluted with pH 6.0-MES-NaCl Buffer. Samples of the microsomes before and after application to the different affinity columns, as well as fractions eluted with pH 6.0-MES-NaCl buffer were examined by SDS polyacrylamide gel electrophoresis (Figure 10-11). Silver staining of the gels revealed that fractions eluted from the IgGa1 (igg:3)-AffiGel 10 column with pH 6.0-MES-NaCl Buffer contained a single protein with an apparent molecular weight of 52,000 daltons identical to that of the protein precipitated from radioiodinated bovine aortic microsomes by IgGl (isn-l) and IgG1 (isn-3). A sample of the protein that did not bind 83 o o o.H om mH.m a Hoez.m H-o 1a .mw: H. .HH o o a.H ON mH.m a Hoez m.H-e Ia .mm: H. .OH 8 o H.~ 8N mH.m H Huez.m H-e :a .mm: H. .m o o H.m as mH.m a Hue: m.H-e Ia .mHz H. .m o o o.a ms mH.m u Huez.m H-e za .mm: H. .H o o e.m om mH.m m Huez m.H-e za .mm: H. .e o o m.~ mN mH.m < Huez m.H-e :a .mm: H. .m N MN o o m m ewe: coccsm nHLH .4 MH mmH o o m < ewe: coccsm WHcH .m we was a Hm mH HeesHoo coHcaH meomoLo_e um~_—_napom .N ooH omoH ooH omoH mH HaasHoo ococoev mmaomocuwe um~_P_n=~om .H cowpomcm caspou «xv umcm>oomm Amu_=:v any umco>oomm Amuwczv idaaql. sHH>HHo< oneeHesm NHua »HH>HHo< omegeesm NHoa ossHo> oH-Hochc<-HHacHeooHmeH oH-HouHcc<-Hm-mmmvHomH mmmgu:»m NHua mo xsqmcmouesocsu xuw=_mmoocs=a: m mHQm» 84 Figure 10. SDS-Polyacrylamide gel electrophoresis of eluates from 1961(133-3)-Affi6el-10. PGIZ synthase was purified by immunoaffinity as described in the text and Table 5. Fractions were subjected to SDS-polyacrylamide gel electrophoresis and silver stained. Lane a, solubilized microsomes before the application to immunoaffinity column; lane b, solubilized microsomes after passage through the 1901 (isn- 3) AffiGel- 10 column; Lanes c-i, fractions eluted— from the IgG isn 3) -AffiGel-10 column with pH 6. 0- MES- NaCl Buffer—(Column fractions 5- 11 of Table 5). 45,000 - 3|,000- 85 86 Figure 11. SDS-Polyacrylamide gel electrophoresis of eluates from IgGl (control)-AffiGel-10 column. P012 synthase was applied to an 1961 (control)-AffiGel-10 column, as described in the text and in Table 5. After washing to remove unbound protein, PGIZ synthase was eluted at pH 6.0. Fractions were subjected to SDS-polyacrylamide gel electrOphoresis and silver stained. lane a, solubilized microsomes before application to column; lane b, solubilized microsomes after passage through the column; lanes c-i, fractions eluted from the IgGl (control)-AffiGel-10 with pH 630-MES-NaCl Buffer (Column Fractions 5-11 of Table 5 . 87 m. 83 Figure 12. Specific activities of P61 synthase eluted from the 1961 (j§p¢B)-AffiGel-10 Co umn. 89 29.542“. ¢ n ‘ u u "N CON 00¢ 00m 00m 000. 90 the 1901 (isn-3)-AffiGel-10 column showed a reduction in the intensity of staining of the 52,000 dalton band; there was no apparent, staining change in the relative staining intensities of other proteins. No proteins was detected in samples eluted from the Ing (control)-AffiGel-10 colunn with pH 6.0-MES-NaCl Buffer (Figure 11). These results indicate that a single protein with a molecular weight of 52,000 daltons is selectively bound to the 1961 (ign-3)-AffiGel 10 column and that this protein is eluted under conditions in which PGIZ synthase activity is also eluted. The specific activity of the isolated PGIZ synthase varied among fractions; a maximum specific activity of ca 1000 nmole PGIzlmin/mg protein was found in fraction 4 (Figure 12). There is clearly one major protein eluted from the IgG1 (igge3) column, and no major protein eluted from the control column. However, a relatively high background level of protein is bound non-specifically to the control column (Figure 13), and this protein is eluted with pH 6.0-MES-NaCl Buffer Apparently, proteins other than P612 synthase are bound uniformly so that while the amount of contaminating protein is significant, no single contaminating protein can be detected on SOS-polyacrylamide gels (Figure 11). Spectral Characteristics of Purified PGIZ Synthase Fractions eluted from the IgGl (lgge3) column with pH 6.0-MES-NaCl Buffer exhibited a Soret-like band of absorbance (“max =- 418-420 nm) whose intensity varied in direct proportion to the P612 synthase activity (Figure 14). Only a single band of absorbance was detected between 350 and 650 nm. The absorbance of less intense a and a heme absorption bands in the visible region, are 91 Figure 13. Protein elution profile for I961 (ign—B)-AffiGel-10 and 1961 (control)-AffiGel-10 Columns. Solubilized bovine aortic microsomes were applied to the two columns as described in the text and Table 5. Proteins in each fraction eluted with pH 6.0-MES-NaCl Buffer were assayed by the Lowry procedure . IgGl (j§gr3)AffiGel-10, (0—___l); 1961 (control)-AffiGel-10 (0———_0). 92 L l l o 2 o In N — (I‘ll/5") N13108:! FRACTION 93 Figure 14. Coelution of PGI synthase activity and Heme Absorbance from T961 (isn- 3) -AffiGel- 10 with 0.1 M MES, pH 6. 0, containing 0. 5% Triton X- 100 and 1 M NaCl. 94 Hoig €3.an 8.0.3 £284 68% w w m w m m H 1 H H H H l 7 I. 6 1 5 .m I. 4 m H L 3 12 _ L 4 2 m m To ego? o§< 95 usually 10% of the Soret absorbance and would be below our detection limits with the protein concentrations obtained (< 5 pg/ml). No absorbance was detected between 350 and 650 nm in the samples eluted from the 1961 (control)-AffiGel-10 column with pH 6.-MES NaCl buffer suggesting that the heme-absorbing species is the P612 synthase. The quantity of heme in various preparations was determined using several approaches. First, a sample of inactive P612 synthase was prepared by eluting the 1961 (1gp:3)-AffiGel 10 column with 1 M acetic acid, pH 2.5, containing, 0.15 M NaCl and 0.5% Triton X-100. This procedure desorbs most of the bound protein into a single concentrated 5 ml fraction. Pyridine hemochromogen analysis was performed on this inactive enzyme preparation using a difference millimolar extinction coefficient of the absorbance difference between the peak of the a band at 557 nm and the minimum which occurs beween the a band and the B band at 44l (Falk, 1964). Using this value and values for total protein, I computed that the sample contained protOprophyrin IX heme in ratio of one heme per ten 52,000 dalton polypeptides. Estimation of the heme content of active P012 synthase isolated by elution from the (1gg:3)-AffiGel 10 column with pH 6.0-MES-NaCl Buffer yielded a ratio of one heme per two-52,000 dalton proteins. For this calculations the concentration of P612 synthase protein was determined as the difference in protein concentrations between the peak activity fraction isolated from the 1961 (15373)-Affi6el 10 column and the corresponding numbered fraction eluted from the control column, and the absorbance at 420 nm was used to calculate the concentration of heme. The millimolar extinction coefficient for the PGH synthase,E nM412 = 120 “'1 mM'I. 96 (Roth gt 31. 1981) was used; however, other protoporphyrin IX containing proteins have similar extinction coefficients (e.g. cytochrome P420: 414 = 124; Omura and Sato 1964). Results of the heme analysis suggest that the molar ratio of heme to protein subunit (0.1 to 0.5) can account for the recovered activity of the P612 synthase (ca 25%). Because the P612 synthase appears to be a heme-containing protein, several heme binding compounds as well as hematin were examined for their effects on enzyme activity. Neither sodium cyanide (1 M40 _mM), nor sodiun azide (1 pM-100 _mM) had any effects on the activity of either the microsomal or purified P612 synthase. Both of these nitrogen compounds usually bind ferriheme. Carbon monoxide, a ferrous binding anion, was also ineffective in inhibiting the enzyme. However C0 does bind the heme following reduction of the enzyme by dithionite (Figure 15). It was thought that addition of heme might stabilize the P012 synthase enzyme during purification or restore activity possibly lost by dissociation of the heme during purification. However addition of lo‘pM_hematin to the buffers used in the purification failed to improve the yield of enzyme isolated from the 1961 (lgge3)-AffiGel 10 column; moreover, hematin added after purification failed to increase the . activity of P012 synthase isolated in the absence of added heme. Inactivation of P619 Synthasegby 13-hydroperoxy-linoleic acid and £§M2;_ Several specific P012 synthase inhibitors were found to profoundly affect both the activity and the spectral characteristics of the purified PGIz synthase. One compound, 97 Figure 15. Carbon monoxide-dithionite reduced difference spectrun of purified PGIZ synthase. Purified PGI synthase in the sample cuvette was reduced by adding a few crystals of sodium dithionite. Carbon monoxide was then bubbled through the sample cuvette and the reference cuvette containing unreduced P012 synthase for 30 min. 98 15.... séo séo 430 440 4éo. 480nm 99 9,11-azo-prosta-5,13-dienoic acid (Azo Analog I; Gorman gt gl., 1978), a PGHZ analogue differing from PGHZ in the substitution of a dinitrogen bridge for a dioxygen bridge between carbons 9 and 11 and lacking an hydroxyl group at C-IS, caused a dose related decrease (0.3-30 H!) in the activity of P612 synthase (Table 6). As discussed previously, unsaturated hydroperoxy fatty acids, also inhibit the microsomal P612 synthase (Salmon gt 91., 1978). Addition of 13-HP-linoleic acid (20, 50 or 100 E!) to the purified PGIZ synthase caused a rapid, time-dependent inactivation of the purified PGIZ synthase; 100 E! 13-00H linoleic acid caused 100% inhibition of activity within 5 min (Figure 16). Accompanying the inhibition caused by both Azo Analogue I and 13-HP linoleic were significant changes in the spectra of purified PGIZ synthase preparations. Addition of A20 Analogue I at the concentrations studied (0.3 ng30 E!) caused a shift in the peak of Soret absorbance from 420 nm to 426 nm (Figures 17,18). Presumably, this shift in absorbance in results from coordination of the A20 Analogue I with the heme group of the P012 synthase. Treatment of P612 synthase with 13-HP-linoleic acid caused a different spectral change. A two minute incubation of purified PGIZ synthase with 100 pM 13-HP linoleic acid resulted in the loss of 85% of the P612 synthase (Table 6) and concomitant bleaching of the heme spectrum (Figure 17). The reduced absorbance is likely due to chemical modification of the heme. If the heme had simply been dissociated from the protein, one would have expected to see a shift in the absorption maximum to that of free heme. 100 Figure 16. Inhibition of purified PGIZ synthase by 13-hydroperoxy-linoleic aCld; 20, 50, and 100 pM 13-HP-linoleic acid was added to PGIZ synthase purified by immunoaffinity chromatography as described in the text. After the indicated incubation times, residual P612 synthase activity was determined as described in the text. For zero time incubations substrate (PGHZ, 50 pM) and inhibitor were added simultaneously. lOl P612 Synthase Aciiviiy 0\c— a_ \opm l3-OOH linoleic a” /+2opu I3-OOH linoleic O.— 2 Incubation Time (nip) l02 Figure 17. Effect of 13-hydroperoxy-linoleic acid on the absorbtion spectrum of purified P612 synthase: protection from bleaching by Azo Analogue I. Spectra were recorded on an Aminco spectrophotometer as discussed in the text. Spectra of P612 synthase purified by immunoaffinity chromatography were taken: (a) following no treatment; (b) after a 2 min incubation at 24° with 100 pM 13-HP-linoleic acid and (c) after adding 30 pM Azo Analog I and then treating for 2 min at 24° with 100 pM 13-HPolinoleic acid. The protein concentration of the PGIZ synthase preparation used in these experiments was 8 mg/ml. 103 HrENZYhE+3OpM Azo Amloquo I ENZYhE+30pfl Azo Analogs. 14400;” 13 I-P-LINCLEIC ACID ENZYIEHOOM l3 l-P-LINOLEIC ACID 46045440 WAVELEMITH (Mi) (“I 104 Figure 18. The effect of PGHZ on the absorbtion spectrum of PGI synthase: Protection from Bleaching by Azo AnaTogue I. Spectra of PGI synthase purified by immunoaffinity chromatograp6y were recorded: (a) following no treatment; (b) after a 2 min incubation at 24° with 25 pM PGHZ and (c) after adding 30 pM Azo Analog I and then treating for 2 min at 24° with 25 pM PGHZ. The protein concentration of the P612 synthase preparation used in these experiments was 8 ug/ml- 105 360380 420440460400 VWELENGTHM l06 Addition of 25 pM_PGH2 to purified enzyme caused a bleaching of the heme spectrum similar to that observed with 13-HP-linoleic acid (Figure 18). It has been shown that the PGIz synthase is inactivated (Watanabe gt 31. 1979) during the conversion of PGHZ to P012 and we have confirmed this finding with enzyme imnobilized by attachment to S. lgggggg cells (see below). Under conditions in which the heme spectrum becomes bleached in the presence of PGHZ, the P012 synthase is completely inactivated. The fact that PGH2 (an endoperoxide) and 13-HP linoleic acid both cause similar spectral changes and enzyme inactivation suggests that the chemical modifications caused by these agents is similar and perhaps proceeds via a similar mechanism. Moreover, the bleaching of the heme spectrum observed with by both PGHZ and 13-HP linoleic acid peroxide can be prevented when Azo Analogue I is included in the incubation mixture. As shown in Figure 18, addition of 30 pM_Azo Analogue I to the purified PGIZ synthase preparation before adding 25 pM’PGHz protected the enzyme from bleaching. The absorbance peak remained unchanged in intensity at 426 nm for at least 3 min after addition of the PGHZ. The same concentration of A20 Analogue I also prevented bleaching of the heme spectrum caused by 100 E! 13-HP linoleic acid (Figure 17). It was not possible to determine directly whether prevention of the bleaching phenomenon by A20 Analogue I was accompanied by protection of the enzyme activity since A20 Analog I is itself a PGIZ synthase inhibitor. To circumvent this problem P612 synthase was conjugated to 1961 (133:3) to S, gggggg; this provided a way to incubate the enzyme with PGHZ or 13-HP linoleic in the presence or lO7 absence of Analogue I and then to separate the immobilized enzyme from the effectors. Employing this immobilized P612 synthase synthase system I observed that Azo Analogue I protected the immobilized PGIZ synthase from inactivation by both PGHZ and 13-HP linoleic acid (Tables 6 and 7). Concentrations of A20 Analogue I from 0.3 to 30 pM caused a dose-dependent inhibition of immobilized P012 synthase as assayed {7 with 50 pM_PGH2 (Table 6). In the presence 30 pM,Azo Analogue I, 88% of the total S, aureus bound PGIZ synthase activity remained after a one min incubation with 25 pM_PGH2 (Table 6); in contrast only 22% n of the starting activity remained when P612 synthase was incubated b for one min Azo Analogue I. When PGIZ synthase bound to S, EEEEEE cells was incubated with 20-100.pM 13-00H linoleic for 5 min in the presence of A20 Analog I, 76-86% of the original enzyme activity remained (Table 7). In a parallel incubation without Azo Analogue I, but with 100 pM 13-HP linoleic acid, the PGIZ synthase activity was completely lost. 108 0;» o>oEmL op umsmmz 0cm: mppmo wouoH—ma ;0_;z mcoHumaau:H «mos» Ease ucmuccgmasm as» .xuw>_uoo omegpcxm «Hum Lo» toxemma can umecmamammg cog» .mgouuomwm .coppuauoca «Hog Loo vm~apoco Hugues» 0L0: Nzca umuspoc_ .cowuomampcucmo xn umumqu_umsn cog» :Hs H so» msouomewm nmuauHuc_ any gavz umumnzu:_ mo: me AmLHMMV meH uw> mamcza .w.:u_z umxopasoo mmmnucam «Hana oooH omH H osmoHee< a~<.ms o.om + qus as o.m~ cam cos H osmoHee< a~<.ms o.m + urea.ms o.m~ omm omm H osmaHae< a~< as m.o + Nzua.ms o.m~ emu oooH Nzos as o.m~ omm - H osmoHe=< o~< z: o.om omoH - H osmoHe=< one as o.m omHH - H osmoHae< o~<.ms m.o omHH ago: daHmoauouoxum mHosav Nammwmuoumxio mmposag acmsummcuwsa HHHsHHo< omeeoesm Nst HeeHH copumnzucpmcm acpgzo eoHHosoosa NHua amzwa Ha coHua>Huuacp soc» amasucam NHoa muumuogn H osmopm=< o~< o mpnmh 109 Table 7 A20 analogue I protects PGI synthase from inactivation by 13-Hydroperoxy-Linoleic Aci a PGI Synthase Activity Pretreatment (pmo e 6-keto-PGFlcl none 1290 20 pM 13-HP linoleic 275 50 pM 13-HP linoleic 0 100 pM 13-HP linoleic 0 30 pM A20 Analogue I 1125 30 pM_Azo Analogue I + 20 pM 13-HP-linoleic 1025 30 pM Azo Analogue I + 50 pM 13-HP-linoleic 950 30 pM,Azo Analogue I + 100 pM 13-HP-linoleic 950 aPGIz synthase complexed with S. aureus via IgG (iéflr3) IgG was incubated with the indicated effectors for 5 mIn then precipitated by centrifugation. The pelleted cells were washed to remove the effectors, then resuspended and assayed for P612 synthase activity. llO DISCUSSION The P612 synthase has been purified to electr0ph0retic homogeneity and found to be a 52,000 dalton protein that is associated with protoporphyrin IX heme. There is strong circumstantial evidence that the heme is an integral part of the enzyme and necessary for catalytic activity. The most obvious evidence is that the heme and PGIZ synthase activity both co-elute; no heme is isolated non-specificall y by absorbtion to control col unns. The molar ratio of protein to heme (0.1—0.5) while not stoichiometric is large enough to account for recovered activity. The most convincing evidence to show that the heme is a necessary prosthetic group of the P612 synthase are the results of the inhibitor experiments. Three effectors, fairly specific inhibitors of P612 synthase activity, were found to produce changes in the heme absorbance at concentrations at which they also caused enzyme inhibition. Azo Analogue I caused a shift in heme absorbance from 420 nm to 426 nm, and PGHZ and 13-HP-linoleic acid caused irreversible bleaching of heme absorbance. All of the spectral changes were accompanied by inhibition of P012 synthase activity. The permanent inactivation of P012 synthase by PGHZ and 13-00H-linoleic acid and concurrent permanent bleaching of heme absorbance could be prevented by Azo Analogue I, but this protection required inhibition of P012 synthase by Azo Analogue I. Low lll concentrations of A20 Analogue I (0.3 pM) did not inhibit PGIZ synthase and did not prevent enzyme inactivation or heme bleaching by PGHZ and 13-HDP-linoleic. Higher concentrations of A20 Analogue I (30 MM) inhibited PGIZ synthase activity and protected the enzyme from inactivation and heme bleaching. If the heme were not associated with the PGIZ synthase, it is unlikely that the protection from bleaching would be so tightly coupled with enzyme inhibition. Eh Other researchers have postulated that the P012 synthase is a heme containing enzyme. Ulrich gt 31. (1981) obtained cytochrome P-450 like difference Spectra from pig aorta microsomes and argued from a '7 mechanistic point of view that the absorbance was due to P612 synthase. Very recently Ulrich and coworkers (Graf g; 31., May 1982) reported in a poster, at the International Prostaglandin Conference, the purification of P012 Synthase to homogeneity from pig aorta. The protein purified was shown to possess a cytochrome P-450 like spectra. While CO-reduced difference spectra of our purified P612 synthase (Figure 15) shows only a broad absorbance near 440 nm, it is likely that the difference between our spectrum and that of Graf gt 31. is due a difference in conditions used to isolate the enzyme. This is the first report of the preparation of monoclonal antibodies to an impure enzyme and the subsequent isolation of the enzyme by immunoaffinity chromatography with those antibodies. Others, however, have produced monoclonal antibodies to purified proteins or cell surface antigens and used these antibodies for purification of enzymatically and biologically active proteins. In a particularly elegant study, Hanson and Beavo (1982) isolated anibodies that bind calmodulin associated nucleotide phosphodiesterase only. The 112 phosphodiesterase could be eluted with EDTA buffer which causes dissociation of calmodulin. Other protein purification reported include: HLA-A and -B antigens (Parham 1979), H-2K antigen (Herrman and Mescher, 1979), nicotinic acetylcholine receptors (Lennon gt 91., 1980), and coagulation Factor V (Katzman g; 31., 1981). All of these proteins were isolated, in a biologically active form. They illustrate the potential for use of monoclonal antibodies in purification procedures. Immunoaffinity purification with columns prepared with conventional antisera are notoriously difficult to elute (Pharmacia, 1979); almost always the conditions that will elute a protein will also denature it. This is probably due to the binding of multiple antigenic determinants on a single protein by the various idiotypic antibodies present in antisera.‘ The difficulty in eluting active proteins, together with the low specificity of most antisera immuno-affinity columns, makes affinity-purification with polyclonal antisera useless for most purposes. Monoclonal antibodies, however, react with a single antigenic determinant, and the conditions necessary to disrupt a single interaction and effect elution from a column are much less harsh than those for multiple interaction. The P012 synthase was eluted at pH 6.0 with 1.0 M NaCl; acetylcholine receptors were eluted at pH 10.0, with 0.5 M NaCl (Lennon 3; 21., 1980); Factor V was eluted at pH 6.5 with 1.2 M NaCl (Katzman gt 31.); HLA-A and -8 antigen were eluted at pH 11.5, 0.5 M NaCl (Parham, 1979); and H2-K antigen was eluted at pH 8.0 with 0.5% deoxycholate (Herrman and Mescher, 1979). These mild conditions have allowed for recovery of native proteins in each case. 113 The ability to purify active proteins 1000-5000 fold (Katzman gt gl., 1981; Secher and Burker, 1980) in a single step makes the use of immunoaffinity chromatography with monoclonal antibodies an attractive purification technique. The P612 synthase in this purified form has been shown to be quite sensitive to inhibition by 13-HP-linolic acid. The mechanism of inhibition by lipid peroxides has been the subject of controversy for several years. Egan gt 31. (1976) reported that during the reduction of P602 to PGHZ, PGH synthase produces a nascent oxidizing species that might be responsible for the catalytically-induced inactivation of PGH synthase. They also hypothesized (Ham gt_gl., 1979) that reduction of lipid peroxides to alcohols by PGH synthase in microsomes may release the same oxidizing species, and speculated that this may be the mechanism of P612 synthase inhibition by lipid peroxides. Our work with 13-HP-linoleic shows that the purified enzyme can be inactivated by a direct action of the lipid peroxide on the P012 synthase. CHAPTER V QUANTITATION AND LOCALIZATION OF PGH SYNTHASE AND PGI2 SYNTHASE In this chapter I describe the quantitation of PGH synthase and P012 synthase in vascular smooth muscle and endothelium using immunoradiometric assays. I also report studies on the immunocytochemical localization of P612 synthase in non-vascular snooth muscle. 114 115 MATERIALS AND METHODS Materials Monoclonal antibodies secreted by the mouse hybridoma lines igg-l T (1961). 1511-3 (1961). 92-3 (1961) and 2172-5 (1962b) were prepared as described previously (Chapters 11, III). 1961 purified by (NH4)2504 precipitation and DEAE-cellulose chromatography from i MOPC-21 culture media was a gift from Dr. Paula Jardieu. Fluorescein isothiocyanate (FITC)-labeled rabbit anti-mouse IgG was purchased from Miles Laboratories, Inc. CLS II collagenase was obtained from Worthington Biochemicals. All other materials were obtained from sources documented in Chapters II, III and IV. Methods Immunocytofluorescence staining New Zealand white rabbits (2-3 kg) were sacrificed by intravenous injection of lethal doses of 5% Nembutal via the marginal ear vein. Tissues were removed immediately after sacrifice and placed on ice. Tissue samples were then embedded in 5% gum tragacanth on cork cylinders and quick frozen in either isopentane or hexane cooled to -78° in a dry ice-acetone bath. Tissue sections (10 pm) were cut at -25° using an Ames Lab-Tek cryotome. The sections were transferred to glass cover slips and placed in a desicator under water aspiration for 30 min. Each coverslip was then overlayed with one of five different 116 "first antibodies"; these included monoclonal antibodies present in complete HT culture media and secreted by lines iéflr1:.l§flr3» gygr3 or gyg-S; a MOPC-21 1961 was dissolved at a concentration of 5 pg IgG per ml in complete HT media and used as a control first antibody. After a 30 min incubation with first antibody, the coverslips were washed with phosphate-buffered saline (PBS) (composition in MM: 151 NaCl, 45 KH2P04 and 2.5 NaOH), pH 7.2. FITC-labeled rabbit anti-mouse IgG diluted 1:20 in PBS, pH 7.2, was then added to each coverslip. After 30 min, coverslips were again washed with PBS, pH 7.2, and then mounted in glycerol. Epi-fluorescence microscopy was performed using Leitz Orthoplan microscope equipped with a 150 watt Xenon Lamp and appropriate filters for examining FITC fluorescence. Photomicroscopy was performed with an Orthomat camera and Kodak Tri-X Pan film (ASA 400). Isolation of Bovine Aorta Endothelial Cells and Smooth Muscle Bovine aortic endothelial cells were isolated by the method of Ingerman-Wojenski gt 31. (1981). Bovine aorta 20-30 cm long were obtained fresh at slaughter from Michigan State University and immediately placed in 4° Krebs buffer (composition in mM, 118 NaCl, 4.1 KCl, 22 KH2P04, 25 NaHCOz, 2.5 CaClz, 1.8 mM M9504, 14 glucose), pH 7.2. At the laboratory, the aorta were cleaned of adipose tissue and the intercostal vessels were ligated with silk sutures. Aorta were then hung vertically from a ring stand and washed with addi- tional Krebs buffer. The aorta were sealed at the bottom by clamping with hemostats, and the aorta then were filled with Krebs buffer containing 0.5 mg/ml CLS II collagenase. The top of the aorta was covered with foil and the aorta were incubated at 37° for 30 min. The 117 collagen solution was then drained and discarded. The bottom of the aorta were sealed again and the aorta were carefully filled with Krebs, buffer, which was also drained and discarded. Next, the endothelial cells were harvested. The aorta were sealed at the bottom and the aorta were filled 2/3 full with Krebs buffer. The tap of the aorta was clamped shut and the aorta were removed from the ring stand and shaken vigorously. This wash was saved as were two additional washes collected in the same manner. The pooled washes from a single aorta were centrifuged at 1000 x g to collect the endothelial cells into a pellet, and the pellet was washed once with 5 ml of Krebs buffer. After the final centrifugation each cell pellet was resuspended in 1 ml of 0.1 M_Tris-chloride, pH 8.0 containing 0.5% Triton-X-IOO (v/v). The cell were resuspended by vortexing and then sonicated for 30 sec at setting 50 on a Bisonik sonicator. The homogenate was then centrifuged at 48,000 x g for 20 min to remove the insoluble material. The solubilized protein was used for determination of PGH synthase and P012 synthase protein concentrations. Aorta denuded of endothelial cells by the above procedure were slit open longitudinally. The luminal surface was scraped gently with a razor and washed with Krebs to assure that all endothelial cells were removed. Several small pieces of smooth muscle were cut from various locations along the length of the aorta. Pieces from an individual aorta were added to 15 ml of 0.1 M Tris-chloride, pH 8.0 and homogenized with a Teckmar homogenizer. Care was taken to maintain the temperature below 15°. Next, Triton-X-lOO was added to a final concentration 0.5% (v/v), and the sample was sonicated for 1 min at full power on a Biosonik sonicator. The homogenate was centrifuged at 118 48,000 x g for 20 min to remove insoluble material and the PGH synthase and P612 synthase concentrations in the solubilized protein was determined. Immunoradiometric assays for PGH synthase were as in Chapter 2. The assay was standardized using PGH synthase activity from solubilized sheep vesicular gland microsomes. PGH synthase activity was determined polarographically (Hemler gt 31., 1976). Immunoradiometric assays for P012 were performed as in chapter 3 using P612 synthase activity from bovine aortic microsomes, as a standard. 119 RESULTS Quantitation of PGI; synthase and PGH synthase in smooth muscle and endothelium Immunoradiometric assays for quantitating P612 synthase and PGH synthase were developed previously (Chapters II, III). We applied these assays to the measurement of the concentrations of these two proteins in isolated endothelial cells and smooth muscle tissue prepared from bovine aorta (Ingerman-Wojenski g§_gl. 1981). The results are summarized in Table 8. The concentrations of P012 synthase protein were found to be quite similar in both cell layers. The averaged results of the determinations from eleven aorta showed that endothelial cells contained 4.2 t 1.2 units and smooth muscle contain 5.8 t 1.0 units of P612 synthase activity per mg of solubilized protein. Surprisingly, the PGH synthase levels differed considerably between the smooth muscle and endothelium with the endothelium having approximately 20 times the PGH synthase concentrations of smooth muscle. Localization of P012 Synthase in Vascular Smooth Muscle and Endothelium The distribution of PGH synthase and P612 synthase as determined by immunoradiometric assay was visually corroborated by immunocytochemical localization. The patterns of fluorescent staining obtained with IgG(gyg) and IgG(1§M) are analogous to the quantitation 120 Table 8. Concentrations of PGH Synthase and P012 Synthase in Cell Layers of the Bovine Aortaa PGH Synthaseb P612 Synthasec Cell Layer (units/mg protein) (units/mg protein) Endothe1ium (7) 50 t 14 4.2 t 1.2 Smooth muscle (7) 2.8 t 1.1 5.8 t 1.0 a Samples from seven different bovine aorta were analyzed using imnunoradiometric assays for P612 and PGH synthases as described in the text. bOne unit activity is defined as that amount of enzyme which will utilize 1 nmole oxygen per min. cOne unit activity is defined as the amount of enzyme which will convert 1 nmole PGHZ to P012 per min. l21 results. In a fluorescent micrograph (Figure 19) of an artery of sheep myometrium, staining for PGH synthase with 19023 (21235) revealed intense fluorescence in the endothelial cell layer; comparatively little staining is seen in the smooth muscle. By contrast, a similar section (Figure 20), stained for PGIZ synthase with IgG1(j§M¢3) shows intense fluorescence throughout the smooth muscle and in the endothelium. Localization of PGIgsynthase in nonvascular smooth muscle The fact that PGIZ synthase concentrations in vascular smooth muscle were relatively high suggested to us that P012 synthesis might be characteristic of other types of muscle tissue. Accordingly, we examined the smooth muscle from a variety of rabbit tissues and organs for the presence of P612 synthase antigenic reactivity by indirect immunocytofluorescence using procedures similar to those employed earlier in studies on PGH synthase (Smith and Bell, 1978). PGIZ synthase immunoreactivity was found to be associated with nonvascular smooth muscle in lung, trachea, uterus, urinary bladder, ureter, ductus deferens, and seminal vesicles (Table 9). The longitudinal smooth muscle and muscularis mucosa, but not the circular smooth muscle of the gastrointestinal tract also stained positively for the PGIZ synthase antigen; in contrast to the results with smooth muscle, we found that neither striated cardiac nor skeletal muscle stained for P012 synthase. Our results indicate that PGIZ synthesis is characteristic of both vascular and nonvascular smooth muscle cells. Presunably, PGIZ has a general role as a regulator of the contractile response in smooth muscle. 122 Figure 19. Immunocytofluorescent localization of PGH synthase in a artery in a cross section of ovine myometrium. Cross sections of ovine myometrium were incubated sequentially with 196 b (gygeS) and fluorescein isothiocyanate-labele rabbit anti-mouse IgG and then examined by fluoresence microsc0py. Magnification, 200 X. 123 124 Figure 20. Immunocytofluorescent localization of P612 synthase in an artery in a cross section of ovine myometrium. Cross sections of ovine myometrium were incubated sequentially with 196 (isn-3) and fluorescein isothiocyanate-labele r366it anit-mouse IgG and then examined by fluorescence microsc0py. Magnification, ZOOX. 125 126 One curious observation from the immunocytochemical studies was that the most intense fluorescent staining in some, but not all, types of smooth muscle was associated wnth the periphery of the cell. Figure 21 provides an example of this unique pattern of anti-P012 synthase staining as it occurs in the rabbit urinary bladder. Figure 22 shows control staining with control MOPC-Zl for comparison. These results suggest that the P612 synthase can be associated with the cell surface in certain cell types. It is not yet clear whether in these instances, the enzyme is on the inside or outside of the plasma membrane. A plasmalemma location for P612 synthase contrasts with the PGH synthase which appears to be associated with the endoplasmic reticulum in most cells (Rollins and Smith, 1980). 127 . . uoHuLoo Heme: angam . . Hmumpwxm mHLouuma upnnmm + . zuceEm aHgoog» HHnnmm + . :aooEm .m.=omm_mm mcsH HHHHum + . swooEm Loewe—m Hpmc apnoea + + spooEm .cu—augpo magma: amogm 0.2mnpa nouxo Amouwczm _pmuv+ .gm0—0=:v+ guooEm ..ucpu:u_mco— magma: qmmsm + . spoosm .meaugru magma: HHnnoa + - 58.5 .HesoBHmeoH 3.3.. :23. Housmczm HH00V+ . space» mmH0Hmm> Hocpsmm anamm Amuumcam Hpmuw+ . geooEm wuopmoca HHAHnm Amummszm Hpmu + . geoOEm mcmsmwoo msaoao upnnoa + . spoosm .Lopzugpu Lupus: annaa + - 58...... .HeeHeSHaeoH .32: :83. 500mmg=m Hpouv+ . guoosm Locum—m ng:_g= apnoea + + :uoOEm .umouas chaH:um:s ocHammucH _pmsm apnoea Acme—0:: w Acmmpoac + :uooEm .gapzuspu ocHHmmucH —Ha5m “Hanna Asoopoac .ouuwczm _Ho0w+ Homo—u:=w+ guooEm .pacvuauwmco— mcpumoqu HpaEm HHnnem + :uoosm .umcuze mHLu—aumae ocwumoucH macuH “Home: u . swooEm .supzus_o 0=HHmmucH omcmH upnnea + . spoosm .Hucpuauwmco— 0:.ammucH maLuH HHHAua + Acoopuzcv+ cacoEm .nmouas mvcmpaumse Amauczmv consoum «Hanna 1 . zuoosm .Lu—aucpu mavcsm seasoum uvnnam + - 58...... .HeEHHBHHHeoH M323 .3283 :23. + . gpoosm .ga—=0m~> Hgoug< “Hanna amazucxm omnsucxm «Haw o—umzz mammvh mmpuomm NEH. =9. m—Umaz :_ xup>pauumsocassH omagucxw «Hum ecu omngucam :aa mo :ovasawgumwo .m opaah Figure 21. 128 Immunocytofluorescent localization of P612 synthase in rabbit urinary bladder. Cryotome sections of urinary bladder were incubated sequentially with 190 (isn- 3) and fluorescein isothiocyanate-labeled rab6it— anti-mouse IgG and then examined by fluorescence microscopy. Magnification, 200 X. 129 130 Figure 22. Control immunofluorescent staining of rabbit urinary bladder with MOPC- 21 mouse IgGl. Cryotome sections of urinary bladder were incubated sequentially with 196 (MOPC- 21) and fluorescein isthiocyanate-labeled rabbit anti-mouse IgG and then examined by fluorescence microsc0py. Magnification, 200 X. 131 132 DISCUSSION We have measured by immunoradiometric analysis the concentration of PGH synthase and P612 synthase in smooth muscle and endothelium. Both tissues were found to contain similar amounts of P012 synthase and thus probably have equal capacity to convert PGHZ to PGIZ. These results conflict with earlier assays for P612 synthase enzyme activity in rabbit aorta (Moncada ggngl., 1977) but are consistent with more recent observations suggesting that the denuded vasculature (Ts'ao gt 31., 1979) as well as isolated smooth muscle cells (Baezinger g; 31,, 1977, 1979) have the capacity to synthesize significant amounts of prostacyclin. Endothelial cells however were found to contain roughly 20 times the PGH synthase that smooth muscle contained, and are thus able to produce 20 times as much PGHZ from arachidonic acid. These results are consistent with the relative inability of newly deendothelialized vasculature to synthesize PGIZ from arachidonic acid (Eldor gt 21., 1981) and suggest that formation of a PGIz-forming, neointimal layer following vascular injury (Eldor gt 51., 1981) is associated with an increase in PGH synthase. The endothelial also appears to have a greater capacity the synthesize PGHZ than to convert PGHZ to P012. The endothelial cell layer has 50 units of PGH synthase activity per mg of protein, which means a mg of endothelial protein contains enough PGH synthase to 133 utilize 50 nmole/min 02, and thus can convert 25 nmoles arachidonic acid to PGHZ per min. The endothelial cells contained 4.2 units of P612 synthase activity per mg protein. Because the P612 synthase assay I have used does not measured initial rates the assay may under- estimate P012 synthase activity 2-3 fold. Even so, the endothelial layer has at least twice the capacity to form PGHZ as it does to convert PGHZ to PGIZ. It is tempting to speculate that PGHZ produced by the endothelium may be transported out of these cells, either to the smooth muscle to be converted to P0121 or to the blood to be converted to TxAz by platelets. A specific effector such as thrombin, which is known to elicit PGIZ production in endothelial cells in culture (Weksler g; l., 1978), could stimulate PGHZ production lg vivo. Depending on the amount of PGHz synthesized, it would either be converted in the endothelial cell to P612 or exported. If transported to smooth muscle the PGHZ would be converted to P012 and smooth muscle relaxation would occur. If the PGH were transported to the blood, platelets would produce TxAz and aggregation would be enhanced. Endothelial cells could thus be control cells for regulation of smooth muscle tone, or platelet aggregation. Endothelial cells have already been shown to mediate the responses of acetylcholine in smooth muscle (Furchgottlg§.gl., 1981). Acetycholine added to vascular strips with the endothelium intact cause relaxation, while acetylcholine added to vascular strips with the endothelium removed causes contraction. It can be shown that acetylcholine induces the endothelium to release an effector, which though not identified causes the relaxation. Thus, endothelial cells may mediate the effects of humoral stimuli such thrombin thereby 134 participating in the regulation of vascular tone. Further work will be necessary to determine whether release of PGH2 from the endothelium actually occurs. A series of monoclonal antibodies specific for either PGIZ synthase (secreted by jggrl and 1§373) 0r PGH synthase (secreted by nggS and Engs have been used to stain cryotome sections from a variety of tissues by indirect immunocytofluorescence. The two monoclonal antibodies used to detect the two antigens were selected on the basis of their known reactivities with different determinants on P612 and PGH synthases. Positive staining using a monoclonal antibody along with negative staining using nonimmune and enzyme-adsorbed immune controls indicates that the antigen is present. It should be noted, however, that a lack of positive staining with a monoclonal antibody should not be taken as evidence that an enzyme is not present. Rather the enzyme may be present at levels below the detection limit of the immunocytofluorescence procedure or may be in a physical state or isozymic form that renders the antigen unreactive with the monoclonal antibody. For example, PGH synthase in the smooth muscle layer of bovine aorta is not readily detectable by immunofluorescence; however, the enzyme in this smooth muscle layer can be measured by more sensitive immunoradiometric assays. The fact that the P012 synthase antigen is present in most smooth muscle layers but not in the striated muscles examined in this study (Table 8) suggests that P612 synthesis is a characteristic feature of smooth muscle metabolism. P012 normally causes relaxation of vascular smooth muscle. In this regard, the action of P612 may be 135 to elevate intracellular levels of cAMP which in turn results in relaxation of muscle tone (Miller g; 31., 1979). No P012 synthase was detected in circular smooth muscle in the gastrointestinal tract. Furthermore, there are differences in the subcellular distribution of PGIZ synthase immunofluorescence among smooth muscles from different tissues. These observations indicate that the synthesis of P612 must be regulated in different ways in different smooth muscle layers. One might therefore anticipate that PGIZ also plays different roles in modulating smooth muscle activity in different organs and tissues. In fact, longitudinal smooth muscle of the large intestine has been shown to contract rather than relax in response to PGIZ (Nakahata and Suzuki, 1981). The subcellular location of PGI2 synthesis is of interest in understanding the regulation of prostacyclin synthesis. Previous studies employing both differential centrifugation techniques (DeWitt g£_al., 1981) and light and immunoelectron microscopy (Rollins and Smith, 1980) have indicated that the prostaglandin endoperoxide, PGHZ, is formed on the cytoplasmic surface of the endoplasmic reticulum and on the nuclear membrane. However, the subcellular locations of P612 synthase, TxAz synthase and PGH-PGE isomerase have not been defined. The results of the present studies suggest that in many cells the P012 synthase can occur in association with the plasma membrane. Thus, the general concept that formation of PGH2 occurs proximal to the conversion of PGHZ to P612 and PGE2 may need revision. For example, PGHZ synthesized in association with the endoplasmic reticulum may require a specific carrier system for transport to the cell surface before final metabolism to PGIZ. 136 Synthesis of PGIz from PGHZ may then occur in conjunction with P612 exit from the cell, or alternatively PGHZ exit may precede enzymic transformation to P612. In order to answer these questions it will be necessary to determine the transverse orientation of the active site of P812 on the plasma membranes of smooth muscle cells such as those derived from the urinary bladder (Figure 21). REFERENCES Aiken, J.W., Shebuski, R.J., Miller, D.V. and Gorman, R.S. (1982): Pharmacol. Exp. Ther.; in press. Ali, A.E., Barrett, J.C. and Eling, T.E. (1980): Prostaglandins, gg:667-688. Baenziger, N.L., Dillender, M.J. and Majerus P.W. (1977): Biochem. Biophys. Res. Commun., 18:294-301. Baenziger, N.L., Becherer, P.R. and Majerus, P. (1979): Cell 1S:967-974. Bolton, A.E. (1976): In: Radioiodination Techniques, Amersham Corporation, Arlington Heights, Illinois. Bolton, A.E. and Hunter, W.M. (1973): Biochem. J., 1SS5529-539. Born, G.V.R. (1964): Nature, 125,927-292. Bradford, M.M. (1976): Analytical Biochemistry, 12:248-254. Bunting, S., Gryglewski, R., Moncada, M., and Vane, J.R. (1976):Prostaglandins,.lgz897-913. Burch, J.W., Baenziger, W.C., Stanford, N. and Majerus, P.W. (1978): Proc. Natl. Acad. Sci. USA, 1S55181-5184. Christ-Hazelhof, E. and Nugteren, D.H. (1981): Prostaglandins, gg;739-746. Clark, M.R., Marsh, J.M. and LeMaire, W.J. (1978): J. Biol. Chem., SSS:7757-7761. 137 l38 Clement, M.G., Triulzi, M.O., Maggi, G.C. and Aguggini, G. (1980): Prostaglandins and Medicine, S5323-335. Coughlin, S.R., Moskowitz, M.A., Zetter, B.R., Antoniades, H.N. and Levine, L. (1980): Nature, gS§:600-602. Czervionke, R.L., Hoad, J.C. and Fry, G.L. (1978): J. Clin. Inv., S2:847-856. Czervionke, R.L., Smith, J.B., Fry, G.L., Hoak, J.C. and Haycraft, O.L. (1979a): J. Clin. Inv., SS:1089-1092. Czervionke, R.L., Smith J.B., Hoak, J.C., Fry, G.L., and Haycraft, O.L. (1979b): Thrombosis Research, 14:781-786. Dembinska-kiec, Gryglewski, T., Zmuda, A. and Gryglewski, R.J. (1977): Prostaglandins, l551025. DeWitt, O.L., Rollins, T.E., Day, J.S., Gauger, J.A., and Smith, W.L. (1981): J. Biol. Chem., gSS:10375-10382. DeWitt, O.L. and Smith, W.L. (1982): In: Methods in Enzymology. edited by W.L. Lands and W.L. Smith, Academic Press, New York, SS in press. Dusting, G.J., Moncada, S. and Vane, J.R. (1977): Prostaglandins, 1333-15. Dusting, G.J., Moncada, S. and Vane, J.R. (1978): Br. J. Pharmacol., §§:414. Edlund, A., Bomfinn, W., Kaizser, L., Olin, C., Patrono, C., Pinca, E. and Wennmalm, R. (1981): Prostaglandins, 22 323-331. Egan, R.W., Paxton J. and Kuehl, F.A. (1976): J. Biol. Chem., SSS:7329-7335. Eldor, A., Falcone, P.J., Hajjar, D.P., Minick, C.R. and Weksler, 8.8. (1981): J. Clin. Invest., SZ:735-741. l39 Ey, P.L., Prowse, S.J. and Jenkin, C.R. (1978): Immunochemistry lS:429-435. Fitzgerald, G.A., Brash, A.R., Falardeau, P. and Oates, J. (1981): J. Clin. Invest., SS51272-1276. Freedman, M.M., Gossberg, A.L. and Pressman, D. (1968): Biochemistry, 151941-1950. Furchgoot, R.F., Zawdzki, J.V. and Cherry, P.D. (1981): in: Vasodialation, ed. by P.M. Vahoutte and I. Leusen, Raven Press, New York, pp. 49-66. Galfre, G., How, S.C., Milstein, C., Butcher, G.W. and Howard J.C. (1977): Nature, 2SS;550-552. Garcia-Perez, A. and Smith W.L. (1982): Am. J. Phys., in press. Gorman, R.R., Bunting, S. and Miller, D.V. (1977): Prostaglandins, SS:377-381. Gorman, R.R., Hamilton, R.O. and Hopkins, N.K. (1978): J. Biol. Chem., gSg:1671-1676. Gorman, R.R. (1979): Fed. Proc.,ngz79-88. Graf, H., Castles, L. and Ullrich, V. (1982): Abstract from poster presented International Conference on Prostaglandins, Florence Italy, May 18-21. Gryglewski, R.J., Bunting, S., Moncada, 5., Flowers, R.J. and Vane, J.R. (1976): Prostaglandins, lg:685-713. Gryglewski, R.J., Korbut, R. and Ocetkiewicz, A. (1978): Nature, 21Sg765-767. Gryglewski, R.J., Kobut, R., Ocetkiewicz, A. (1978): Prostaglandins, lSz637-644. 140 Hall, E.R. and Tai, H.H. (1981): Biochemica et Biophysica Acta, _6_6_S:498-503. Ham, H.A., Egan, R.W., Soderman, 0.0., Gale, P.H., and Kuelhl, E.A. (1979): J. Biol. Chem., 2S5 2191-2194. Hamberg, M. and Samuelsson, B. (1974): Proc. Nat. Acad. Sci. USA, 21,3824-3828. Hamberg, M., Svensson, J., Wakabayashi, T. and Samuelsson, B. (1974): Proc. Nat. Acad. Sci. USA, 115345-349. Hamburg, M., Svensson, J., Wakabayashi, T. and Samuelsson, B. (1974): Proc. Nat. Acad. Sci. USA, 11:3824-3828. Hansen, R.S. and Beavo, S.A. (1982): Proc. Nat. Acad. Sci. USA, 12,2788-2792. Harada, Y., Tanaka, K., Katori, M. (1980): Proc. Jap. Acad., SS3394. Hawkins, H.J., Smith, J.B., Nicolaou, K. and Eling, T.E. (1978): Prostaglandins,‘;§:871-884. Hemler, M., Lands, W.E. and Smith W.L. (1976): J. Biol. Chem., g§;;5575. Hemler, M.E., Cook, H.W. and Lands, W.E. (1979): Arch. of Biochem. and Biophysics,.;SS:340-345. Higgs, G.A., Moncada, S. and Vane J.R. (1977): Br. J. Pharmacology, §l3137. Hong, S.C. (1980): Thrombosis Research, lS:787-795. Huslig, R.L., Fogwell, R.L. and Smith, W.L. (1979): Biol. Reprod. 21:589. Ingerman-Wojenski, C., Silver, M.J., Smith, J.B. and Maracarak R. (1981): J. Clin. Invest., S151292-1296. 141 Johnson, R.A., Morton D.R., Kinner, J.H., Gorman, R.A., McGuire, J.C., Sun, F.F., Wittaker, N., Bunting, S., Salmon, J., Moncada, S. and Vane, J.R. (1976): Prostaglandins, 12:915. Katzman, J.A., Nesheim, M.E., Hibbard, L.S. and K.G. Mann (1981): Proc. Nat. Acad., Sci. 18:162-166. Kent, R.S., Kitchell, B.B., Shand, 0.0. and Whorton, A.R. (1981): Prostaglandins,‘2l:483. Kessler, S.W. (1975): Journal Immunolgoy, 117:1617-1625. Laemmli, V.K. (1970): Nature, 221:680-685. Lennon, V.A., Thompson, M. and Chan, J. (1980) J. Biol. Chem., 2SS:4395-4398. Levin, R.F., Jaffe, E.A., Weksler, 8.8. and Tackgoldman, K. (1981): J. Clin. Invest. SZ:762-769. Marcus, A.J., Weksler, B.B., Jaffe, E.A. and Broaekman, M.S. (1980): J. Clin. Invest., SS:979-986. Miller, 0.V., Aiken, J.H., Hemker, D.P., Shebuski, R.J. and Gorman, R.R. (1979): Prostaglandins, lS:915-925. Moncada, S., Gryglewski, R.J., Bunting, S. and Vane, J.R. (1976a): Nature, 29S:663-665. Moncada, S., Gryglewski, R.J., Bunting, S., and Vane, J.R. (1976b): Prostaglandins, $2,715-733. Moncada, S., Herman, A.G., Higgs, E.A. and Vane, J.R. (1977): Thrombosis Research, SS, 323. Moncada, S. and Vane, J.R. (1977b): In: Biochemical Aspects of Prostaglandins and Thromboxanes, eds. by N. Kharasch and J. Fried, Academic Press, New York, pp. 157-177. 142 Moncada, S., Korbut, R., Bunting, S. and Vane, J.R. (1978): Nature, mam-768. Moncada, S. and Vane, J.R. (1979) New England Journal of Medicine, SQQ:1142-1147. Moonen, P., Buytennek, M. and Nugteren D.H. (1981): In: Methods in Enzymology, edited by W.E. Lands and W.L. Smith, Academic Press, New York SS, in press. Morrison, A.R., Montz, H. and Needleman, P. (1978): J. Biol. Chem., SSS:8210-8212. Nakahata, N. and Suzuki, T. (1981): Prostaglandins, 22:159-165. Needleman, P., Moncada, S., Bunting, S., Vane, J.R., Hamburg, M. and Samuelsson, B. (1976): Nature g§l5558-56O. Needleman, P., Bronson, 5.0. and Wyche, A. (1978): J. Clin. Invest., S1:839-849. Needleman, P., Wyche, A. and Raz, A. (1979): J. Clin. Invest., SS:345-349. O'Grady, R., Harrington, S., Moti, M.J., Bunting, 5., Flower, R., Fowle, A.S.E., Higgs, E.A. and Moncada, S. (1980): Prostaglandins, 12:319-332. Omura, T. and Sato, R. (1967): J. Biol. Chem., gSS:2379-2385. Pace-Asciak, C. and Wolfe, L.S. (1971): Biochemistry, l9:3657-3667. Parham, P. (1979): J. Biol. Chem., SS458709-8712. Pharmacia (1981): In: Affinity Chromatography. Pharmacia Upsula, Sweden, pp. 92-95. Phife, 0.0., Lauren, M.C. and Chesney, C. (1981) Prostaglandins, 2;,165-175. Poyser, N.L. (1979): J. Reprod. Fert., SS1,559. 143 Ro1lins, T.E. and Smith. N.L. (1980): J. Biol. Chem., 55:4877-4875. Roth, G.J., Stanford, N.0. and Majerus, P.W. (1976): Proc. Nat. Acad. Sci. USA, _7_2:3073-3076. Roth, G.J., Machuga, E.T. and Strittmatter, P. (1981): J. Biol. Chem., gS§:10018-10022. Salmon, J.A., Smith, D.R., Flower, R.J., Moncada, S. and Vane, J.R. (1978): Biochemica et Biophysica Acta, S2S:250-262. Salmon, J.A. and Flowers, R.J. (1982): In: Methods in Enzymology, edited by W.E. Lands and W.L. Smith, Academic Press, New York, SS: in press. Schafer, A.J., Cooper, 8., Uhara, P. and Hadin, R.I. (1979): J. Biol. Chem., 2S4:2914. Secher, D.S. and Burke, 0.0. (1980): Nature, SSS3446-450. Smith, J.B., Ogletree, M.L., Lefer, A.M. and Nicolaou, R.C. (1978): Nature, 215:64-65. Smith, W.L. and Lands, W.E. (1972) Biochemistry, 1S33276. Smith, W.L. and 8e11, T.G. (1978): Am. J. Physiol., §S§:F451-457. Smith, W.L. and Wilken, G.P. (1977): Prostaglandins, lS;873-892. Smith, W.L. (1981): Mineral Electrolyte Metabolism, S310. Staeheln, 0., Hobbs, S., Kung, H., Lai, C. and Pestka, S. (1981): J. Biol. Chem., SSS 9750-9754. Stahl, R.A.K., Altallah, A.A., Bloch, O.L. and Lee, J.B. (1979): Am. J. Physiol., ggle334. Steer, E., MacIntyre, 0., Levine, L. and Salzman, E.W. (1980): Nature, .QSS:194-195. Sun, F.F., Chapman, J.P. and Maguire, J.C. (1977): Prostaglandins, l4:1055-1074. l44 Tateson, J.E., Moncada, S. and Vane, J.R. (1977): Prostaglandins, lS:389-399. Titus, B.G., Kucacz, R.J. and Lands, W.E. (1982): Archives of Biochemistry and Biophysics, 214:824-828. Ts'AO, C., Holly, C.M., Serieno, M.A. and Galluzzo, T.S. (1979): Thrombosis and Haemostasis, 42:873-884. Ubatuba, F.B., Moncada, S. and Vane, J.R. (1979): Thrombosis Research, 4;,425-435. Ullrich, V., Castle, L. and Weber, P. (1981): Biochem. Pharm., Sg:2033-2036. VanDerOuderaa, F.J., Buytennek, M., Nugeren, D.H. and VanDorp, D.A (1977): Biochemica et Biophysica Acta, fi§1:315-331. Vane, J.R. (1971) Nature (New Biology), gSl:232-235. Watanabe, K., Yamamotom, S. and Hayaishi, O. (1979): Biochem. Biophy. Res. Comm., SZ:192-199. Weksler, B.B., Ley, C.W. and Jaffe, E.A. (1978) J. Clin. Invest., 92:923-930. Wlodawer, P. and Hammarstrom, S. (1978): Biochem. Biophys. Res. Comm., SQ;525-532. Wlodawer, P. and Hammarstrom, S. (1979): FEBS Letters, 21,32. Wlodawer, P. and Hammarstrom, S. (1980): Prostaglandins, 125969-976. Wray, W., Bollikas, T., Wray, V.P. and Hancock, R. (1981): Analytical Biochemistry, SlS:197-203. Wu, K., Chen, Y., Fordham, E., Ts'ao, C., Rayudu, G. and Matyoshi, D. (1981): J. Clin. Invest., SS3362-382. Yoshimoto, T., Yamamoto, S., Okuma, M. and Hayaishi, 0. (1977): J. Biol. Chem., g§255871-5874.