Immunochemical and kinetic evidence for two different prostaglandin H-prostaglandin E isomerases in sheep vesicular gland microsomes.

Splenic lymphocytes from mice immunized with a partially purified prostaglandin (PG) H-PGE isomerase from sheep vesicular glands were fused with SP2/0-Ag14 myeloma cells. Two spleen cell-myeloma hybrids (hei-7 and hei-26) were selected and cloned. The mouse antibodies secreted by the two hybrids, IgG1 (hei-7) and IgG1 (hei-26), caused immunoprecipitation of a maximum of 45 and 22%, respectively, of the solubilized PGH-PGE isomerase activity of sheep vesicular gland; immunoprecipitation of activity by the two antibodies was additive. The antigens reactive with IgG1 (hei-7) and IgG1 (hei-26) were identified as proteins with Mr = 17,500 and 180,000, respectively, by Western transfer blotting or sodium dodecyl sulfate-polyacrylamide gel electrophoresis of immunoprecipitated 125I-labeled microsomes. The PGH-PGE isomerase activities precipitated by IgG1 (hei-7) and IgG1 (hei-26) exhibited different kinetic properties with respect to time course, Km for PGH2, and concentration dependence for GSH. No significant GSH-S-transferase activity was present in these immunoprecipitates. These data indicate that there are at least two different proteins in sheep vesicular gland microsomes capable of catalyzing GSH-dependent PGH-PGE isomerase reactions. IgG1 (hei-7), but not IgG1 (hei-26), caused coprecipitation of PGH synthase and PGH-PGE isomerase activities when incubated with intact right-side-out vesicular gland microsomes. Thus, the epitope for IgG1 (hei-7) is located on the cytoplasmic surface of those microsomal spheres which contain PGH synthase. This latter finding suggests that the isomerase reactive with IgG1 (hei-7) is involved in PGE synthesis in sheep vesicular glands.


Immunochemical and Kinetic Evidence for Two Different Prostaglandin H-Prostaglandin E Isomerases in Sheep
Vesicular Gland Microsomes* (Received for publication, June 16,1986) Yasuhito Tanaka, Sherry L. Ward, and William L. Smith From the Department of Biochemistry, Michigan State University, East L a n s i n g , Michigan 48824 Splenic lymphocytes from mice immunized with a partially purified prostaglandin (PG) H-PGE isomerase from sheep vesicular glands were fused with SP2/ 0-Ag14 myeloma cells. Two spleen cell-myeloma hybrids (hei-7 and hei-26) were selected and cloned. The mouse antibodies secreted by the two hybrids, IgGl(hei-7) and IgGl(hei-26), caused immunoprecipitation of a maximum of 45 and 2290, respectively, of the solubilized PGH-PGE isomerase activity of sheep vesicular gland; immunoprecipitation of activity by the two antibodies was additive. The antigens reactive with IgGl(hei-7) and IgGl(hei-26) were identified as proteins with M, = 17,500 and 180,000, respectively, by Western transfer blotting or sodium dodecyl sulfatepolyacrylamide gel electrophoresis of immunoprecipitated 12'I-labeled microsomes. The PGH-PGE isomerase activities precipitated by IgG,(hei-7) and IgGl(hei-26) exhibited different kinetic properties with respect to time course, K,,, for PGH2, and concentration dependence for GSH. No significant GSH-S-transferase activity was present in these immunoprecipitates. These data indicate that there are at least two different proteins in sheep vesicular gland microsomes capable of catalyzing GSH-dependent PGH-PGE isomerase reactions.
IgGl(hei-7), but not IgGl(hei-26), caused coprecipitation of PGH synthase and PGH-PGE isomerase activities when incubated with intact right-side-out vesicular gland microsomes. Thus, the epitope for IgGl(hei-7) is located on the cytoplasmic surface of those microsomal spheres which contain PGH synthase. This latter finding suggests that the isomerase reactive with IgGl(hei-7) is involved in PGE synthesis in sheep vesicular glands. PGE,' was first described in 1961 as a major prostaglandin metabolite of sheep vesicular gland (1). PGE, is derived * This work was supported in part by United States Public Health Service Grant AM22042, United States Public Health Service Predoctoral Traineeship HL07404, and by an Established Investigatorship from The American Heart Association (to W. L. s.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "aduertisernent" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
'The abbreviations used are: PG, prostaglandin; HT medium, hypoxanthine-thymidine-containing medium; HAT medium, hypoxanthine-aminopterin-thymidine-containing medium (see Ref. 21 for composition of culture media); PBS, phosphate-buffered saline; SDS, sodium dodecyl sulfate. biosynthetically from 8-cis,ll-cis,l4-cis-eicosatrienoic acid. Quantitatively, however, the most important prostaglandin of the E series is PGE, which is formed from arachidonic acid. PGEz is the major prostaglandin released by a number of cell types, including endothelial cells derived from the microvasculature (2), interstitial cells of the renal medulla (3), and epithelial cells of the collecting tubule (4-6). PGE2 is a vasodilator in most vascular beds (7, E)), inhibits Na' resorption in the renal collecting tubule (9) and C1resorption in the medullary thick limb of Henle's loop (lo), and inhibits vasopressin-induced CAMP formation in the cortical collecting tubule (11,12) and the medullary thick limb (11). In many cells which synthesize prostaglandins, PGE2 also causes the activation of adenylate cyclase (13, 14).
The biosynthesis of PGE, from arachidonic acid (or PGEl from 8,11,14-eicosatrienoic acid) involves two successive reactions which are catalyzed by PGH synthase and PGH-PGE isomerase, respectively. There is now a considerable body of information on the kinetic and physical characteristics of PGH synthase and biochemical and pharmacological regulation of this enzyme (see Ref. 15 for review). In contrast, little is known about the properties of PGH-PGE isomerase. PGH-PGE isomerase activities from vesicular gland have been solubilized and partially purified by Ogino et al. (16) and by Moonen et al. (17). The enzyme has been shown to require GSH for activity and to be inactivated by treatment with sulfhydryl reagents. GSH is not oxidized stoichiometrically during enzyme-catalyzed formation of PGE, from PGH2, and a mechanism consistent with this observation and analogous to the role of GSH in the glyoxylase I reaction has been proposed (18). According to this mechanism GSH interacts with C-9 of PGH, to facilitate a 1,2-hydride shift resulting in the 9-keto,ll-hydroxy structure characteristic of PGE derivatives.
As part of our studies on the regulation of PGE, metabolism by renal collecting tubules (4, 6), we prepared monoclonal antibodies against what was expected to be a single PGH-PGE isomerase. We used sheep vesicular glands as the source of the enzyme since extracts of this gland contain high levels of GSH-dependent PGH-PGE isomerase activity (16,17). Surprisingly, we obtained two different monoclonal antibodies, each of which causes significant, but incomplete, precipitation of GSH-dependent PGH-PGE isomerase activity. This report documents our studies on these activities, as well as related work, to determine if the proteins reactive with the monoclonal antibodies are localized in the same cells and on the same membrane systems as PGH synthase, the enzyme which catalyzes formation of the prostaglandin endoperoxide substrate for PGH-PGE isomerase. containing 2 mM GSH and 0.5% Triton X-100, and the samples were incubated on ice for 5 min. Rabbit antimouse IgG serum (40 pl; 92 pg of protein) was then added to each sample. After standing for 2 min on ice, the samples were centrifuged for 10 min at 1500 X g at 4 "C. The resulting pellets were washed with 1 ml of the same buffer solution and then suspended with the same buffer (55 pl). PGH-PGE isomerase activity was assayed as described under "Experimental Procedures." The ratio of IgGl(hei-7) to IgG1(hei-26) in D was 1:3. PGE, formation without enzyme (0.26 nmol/l5 s) was subtracted from the observed rates to give the values presented in the figure.

EXPERIMENTAL PROCEDURES AND RESULTS~
The methods used in the studies reported in this paper are detailed in the Miniprint. Also presented in the Miniprint is a description of the immunocytochemical staining patterns obtained with monoclonal antibodies to PGH-PGE isomerases.
Solubilization of PGH-PGE Isomerase Actiuity-The specific PGH synthase (cyclooxygenase) activity of sheep vesicular gland microsomes was determined to be 5400 nmol of PGH2 formed per min per mg of protein a t 37 "C at saturating concentrations of O2 (200 p~) and arachidonic acid (100 p~) ; the specific PGH-PGE isomerase activity of microsomes was 350 nmol of PGE2 formed per min per mg of protein at 24 "C and at 50 p~ PGH, (approximately the K,,,) at pH 7.0.
As observed previously (16,17), PGH-PGE isomerase activity present in the postnuclear supernatant fraction of sheep vesicular glands homogenates was found almost exclusively (90%) in the 6 x lo6 g x min microsomal pellet. The microsomal enzyme activity could be partially solubilized in 50% yield with 1.5% Triton X-100. Higher concentrations of Triton X-100 (>1.5%) led to decreased recovery of activity. Lubrol PX and Tween 20 were less effective than Triton X-100 in solubilizing PGH-PGE isomerase activity.
Monoclonal Antibodies to PGH-PGE Isomerase Activities-Mice were immunized with samples of a partially purified PGH-PGE isomerase prepared from sheep vesicular gland microsomes essentially as described by Moonen et al. (17).
Splenic lymphocytes from the immunized mice were fused with cells from the SP2/0-Ag14 myeloma line (19,20). Three mouse spleen cell-plasmacytoma hybrids were found to secrete antibodies, which when adsorbed to attenuated Staphylococcus aureus cells to which rabbit anti-mouse IgG had been * Portions of this paper (including "Experimental Procedures," part of "Results," part of "Discussion," and Table IS) are presented in miniprint at the end of this paper. Miniprint is easily read with the aid of a standard magnifying glass. Full size photocopies are available from the Journal of Biological Chemistry, 9650 Rockville Pike, Bethesda, MD 20814. Request Document No. 86M-1887, cite the authors, and include a check or money order for $4.00 per set of photocopies. Full size photocopies are also included in the microfilm edition of the Journal that is available from Waverly Press, conjugated, would cause the precipitation of PGH-PGE isomerase activity. The three hybridomas secreting antibodies which precipitated PGH-PGE isomerase activity were cloned and designated hi-2, hi-7, and hi-26. The antibodies secreted by these cell lines are all of the mouse IgGl subclass as determined by Ouchterlony double diffusion analyses against commercial anti-allotype specific sera.
3The hybridomas designated day-1 and tsn-1 are lines which secrete mouse immunoglobulins (IgG1) which do not interact with any sheep vesicular gland protein as determined (a) by analysis of immunoprecipitates of solubilized sheep vesicular gland microsomes by SDS-polyacrylamide gel electrophoresis and ( b ) by indirect immunocytofluorescent staining of cryotome sections of sheep vesicular gland. A series of experiments were performed to identify the antigens precipitated by IgGl(hei-2), IgGl(hei-7), and IgGl(hei-26). In the first experiment, each antibody was incubated with a sample of '251-labeled solubilized microsomal proteins from sheep vesicular gland and an anti-mouse second antibody, and the resulting immunoprecipitates were analyzed by SDS-polyacrylamide gel electrophoresis and autoradiography. As shown in Fig. 3, IgGl(hei-7)  corresponding to an immunoglobulin heavy chain was found in all the immunoprecipitates. Between eight and eleven lZ5Ilabeled proteins were present in immunoprecipitates of IgGl(hei-2) (data not shown), and no further analyses were performed with this antibody. Immunoprecipitation of radioactivity using IgGl(hei-26) occasionally yielded a faint band ( M , = 120,000), but in most instances no radioactive proteins were precipitated. However, when we analyzed IgGl(hi-26) by Western transfer blotting of solubilized microsomes, a single band ( M , = 180,000) was routinely detected (Fig. 4); no protein-staining bands were seen in Western transfer blotting with IgGl(hei-2), IgGl(hei-7), or IgGl(cyo-3). Together, these experiments indicate that IgG1(hei-7) and IgGl(hei-26) are sheep vesicular gland microsomes with IgG,(hei-7) and IgG1(cyo-3). '2sI-Labeled microsomes ( l a n e 1 ) were prepared as described in the text and were incubated with 100 gg of IgG,(hei-7) ([anes 2 and 3) or IgGl(cyo-3) ([ane 4 ) . The complexes were then precipitated by the addition of 550 pg of rabbit anti-mouse IgG. The precipitates were subjected to SDS-polyacrylamide gel electrophoresis on a 15% SDS-polyacrylamide gel which was then dried and exposed to x-ray film to obtain the autoradiogram. Details of the methodology are presented in the text. directed against unique antigens having subunit molecular weights of 17,500 and 180,000, respectively. Further evidence supporting this concept was obtained in a second immunoblotting experiment. Solubilized microsomes (1.1 mg, 4.5 mg of protein/ml) were incubated with an excess (500-700 pg) of IgG,(cyo-S), 1gG1(hei-7), or IgG1(hei-26) linked to Protein Aagarose (-50 pl), and aliquots (25 pl) of the three resulting supernatants were analyzed by Western transfer blotting for reactivity with IgG1(hei-7), IgGl(hi-26), and IgGl(cyo-3). Protein bands staining with equal intensities with IgGl(hei-26) (Mr = 180,000) remained in the supernatant following immunoprecipitation with either IgGl(cyo-3) or IgG,( hi-7); in contrast, no IgGl(hei-26)-reactive protein was found in the supernatant following precipitation with the IgGl( hei-26)-Protein A-agarose complex (data not shown).
Kinetic Properties of PGH-PGE Isomerases-Some simple kinetic analyses of the PGH-PGE isomerases precipitated by the different anti-PGH-PGE isomerase antibodies were performed to determine if the activities exhibited different characteristics. The results of experiments to determine pH optima, K , values, and other parameters are summarized in Table I. The K,,, for PGH, of the activity precipitated by IgG1(hei-26) was 150 p~, 3-4 times the K,,, values observed with the solubilized microsomal enzyme or with the PGH-PGE isomerase activity precipitated by IgGl(hei-7). In addition, the enzyme precipitated by IgGl(hei-26) required a relatively lower concentration of GSH (50.5 mM) for maximal activity than the enzyme precipitated by IgGl(hei-7) (2 mM). The time course of the reaction catalyzed by the PGH-PGE  were separated on a 9% SDS-polyacrylamide gel and transferred electrophoretically to nitrocellulose. Nitrocellulose strips containing the microsomal proteins were incubated with IgGl(hei-26) ( l a n e I ) or a nonspecific antibody IgGl(day-l) ( l a n e 2) for 5 h a t room temperature, followed by incubation with horseradish peroxidase-affinitypurified goat anti-mouse I g G for 1 h. Details of the methodology are presented in the text.

TABLE I Kinetic properties of PGH-PGE isomerases precipitated by anti-PGH-PGE isomerase antibodies
Immunoprecipitates Immunoprecipitates were prepared as described in the legend to Fig. 1 using 4 pg of IgGl(hei-'l) or 16 pg of IgG1(hei-26).
isomerase precipitated by IgGl(hei-'7) was also significantly different from the time course observed with the activity precipitated by IgGl(hei-26) (Fig. 5). Maximal PGE, formation had occurred by 15 s with the enzyme precipitated by IgGl(hei-7), while PGEz formation had not yet plateaued at 60 s in the reaction catalyzed by the enzyme precipitated by IgGl(hei-26). The time course for the enzyme precipitated by IgGl(hei-7) resembles the time course seen with PGI, synthase where the enzyme activity is rapidly lost during the course of the reaction (21).
The PGH-PGE isomerases precipitated by the different antibodies resembled one another ( a ) in being refractory to inhibition by Ca2+ or M e at concentrations up to 10 mM (Table I) and ( b ) in being activated by phosphatidylcholine (maximal activation occurring with 3 mM phosphatidylcholine at 0.01% Triton X-100; data not shown).
None of the immunoprecipitates exhibited significant GSH-S-transferase activity ( Table I). The specific GSH-S-transferase activity of solubilized microsomes measured using 1chloro-2,4-dinitrophenol (which is a substrate active with most forms of this enzyme (22)) was 17 pmol/min/mg. The GSH-S-transferase activities found in the immunoprecipitates formed with IgG,(hei-7) and IgG1(hei-26) represented 0.7 and <0.1%, respectively, of the starting activity (as compared with 45 and 22%, respectively, of the starting PGH-PGE isomerase activity).
Topological Relationships between PGH Synthase and PGH-PGE Isomerases-PGH-PGE isomerases important in PGE synthesis in vivo would be expected to be located in the same cell and on the same membrane system as the PGH synthase. Therefore, we determined the cellular and subcellular location of the antigens reactive with the anti-PGH-PGE isomerase antibodies. As discussed in the Miniprint, immunocytochemical studies indicated that in sheep vesicular glands, the PGH synthase reactive with IgGl(CY0-3) and the PGH-PGE isomerases reactive with both IgGl(hei-7) and IgGl(hei-26) are all present in greatest relative abundance in glandular epithelial cells.
We next sought to determine if PGH synthase and the PGH-PGE isomerases were all associated with the same membrane system. Anti-PGH synthase, anti-PGH-PGE isomerase, anti-PGI, synthase (IgGl(isn-3) (21)), and control antibodies were complexed to S. aureus cells, and the different antibody3 aureus conjugates were tested for their abilities to coprecipitate PGH synthase and PGH-PGE isomerase activities from microsome preparations (Table 11). IgGl(cyo-3)-and 1&1(hei"I)-s. aureus complexes when incubated with TABLE I1 Immunoprecipitation of PGH synthase and PGH-PGE isomerase activities of intact sheep vesicular gland microsomes by various antibody-S.aureus complexes Rabbit anti-mouse IgG was conjugated with S. aureus cells as described in the Miniprint. Aliquots (0.4 ml) of rabbit anti-mouse IgG-S. aureus conjugate (10% w/v) were mixed with 0.4-ml aliquots of hybridoma medium containing the indicated monoclonal antibodies (SP2 denotes medium from growth of mouse myeloma SP2/0-Ag14 line). After incubation for 60 min at 24 "C, the cell pellets were collected by centrifugation. In experiments designed to measure immunoprecipitation of PGH-PGE isomerase activity, the mouse IgG-S. aureus complexes were suspended in 60 p1 of PGH-PGE isomerase assay buffer (without Triton X-100) and incubated for 3 min at 4 "C with 80 p1 (180 pg of protein) of a suspension of sheep vesicular gland microsomes. The pellets were collected by centrifugation and washed with 1 ml of assay buffer lacking detergent and resuspended in assay buffer. The resuspended pellet and the supernatant were assayed for PGH-PGE isomerase activity. The original PGH-PGE isomerase activity of intact microsomes was 290 nmol/mg/l5 s. In experiments designed to measure immunoprecipitation of PGH synthase activity, the mouse IgG-S. aweus complexes were suspended in 60 pl of PGH synthase assay buffer (20) and incubated for 3 min at 4 "C with 40 pl (90 pg of protein) of a suspension of sheep vesicular gland microsomes. Pellets and supernatants were collected as indicated above and assayed for PGH synthase activity. The original PGH synthase activity of intact microsomes was 5.8 pmol/mg/min. intact vesicular gland microsomes did cause coprecipitation of PGH-PGE isomerase and PGH synthase activities; in contrast, IgGI(hei-26)-, IgGl(isn-3)-, and IgGl(day-l)-S. aureus complexes precipitated only background levels of both activities. The anti-PGH synthase-S. aureu complex failed to precipitate PGH-PGE isomerase activity from solubilized membrane preparations but, of course, did precipitate solubilized PGH synthase activity (data not shown); similarly, anti-PGH-PGE isomerase-S. aureus complexes caused partial precipitation of detergent-solubilized PGH-PGE isomerase activity, but did not precipitate solubilized PGH synthase activity. These immunoprecipitation experiments were conducted under conditions developed previously to establish that PGH synthase is located on the cytoplasmic surface of intact right-side-out sheep vesicular gland microsomes (20). Thus, the present results indicate that antigenic determinants of both PGH synthase and the protein reactive with IgGI(hei-7) are associated with the same membrane systems and are on the same side of the membrane.4

DISCUSSION
The major prostaglandin products formed upon incubations of 8,11,14-eicosatrienoic acid or arachidonic acid with sheep vesicular gland microsomes are PGE, and PGEp, respectively (23). Small amounts of PG12 are also synthesized (24), probably by membranes derived from smooth muscle and endothelial cells present in these glands (25). Vesicular glands 'It should be noted that the anti-PGI? synthase (IgG1(isn-3)-S. aureus) complex failed to precipitate either PGH synthase or PGH-PGE isomerase activities when incubated with intact microsomes (Table 11); however, this anti-PGIz synthase complex did precipitate PGIp synthase activity from microsomal preparations (data not shown). Unlike PGH synthase and PGH-PGE isomerase, PGIp synthase is located primarily on membranes of smooth muscle cells of sheep vesicular gland (25). The fact that anti-PGIz synthase antibodies but not anti-PGH synthase or PGH-PGE isomerase antibodies precipitate microsomal PGI, synthase establishes that IgG1(cyo-3)and IgGl(hei-7)-S. aureus complexes do not simply cause nonspecific precipitation of all membranes present in microsomes prepared from sheep vesicular glands. have been shown by Ogino et al. (16) and Moonen et al. (17) to have a heat-labile activity capable of catalyzing a GSHdependent isomerization of prostaglandin endoperoxides to PGE derivatives; about 90% of this activity is membrane associated. Using Triton X-100, we were able to solubilize approximately 50% of the microsomal PGH-PGE isomerase activity and were able to account for 70-80% of this solubilized activity in immunoprecipitates obtained using IgGl(hei-2), IgGl(hei-7), and IgGl(hei-26).
Our data suggest that there are at least three different proteins in sheep vesicular gland microsomes which can catalyze GSH-dependent PGH-PGE isomerase reactions at significant rates. Two of these isomerases are precipitated by IgG,(hei-7) and IgG,(hei-26), respectively. In addition, since 20-30% of the total solubilized activity could not be precipitated by the antibodies we have developed, there is likely at least one other isomerase activity. IgG,(hei-7) and IgGl(hei-26) precipitate different amounts of the total PGH-PGE isomerase activity and precipitation is additive. The protein antigens precipitated by IgGl(hei-7) and IgG,(hei-26) are different, having subunit molecular weights of 17,500 and 180,000, respectively. Moreover, the PGH-PGE isomerase activities precipitated by IgGl(hei-7) and IgGl(hei-26) differ substantially in their kinetic profiles.
The PGH-PGE isomerase reactive with IgG,(hei-7) has several properties which would be expected for a biologically important PGH-PGE isomerase. Immunofluorescent staining indicated that the protein reactive with IgGl(hei-7) is present in greatest abundance in glandular epithelial cells of vesicular gland, the same cells which stain most intensely for PGH synthase. Moreover, IgGl(hei-7) and IgGl(cyo-3), when conjugated to s. aureus cells, cause the coprecipitation of both PGH synthase and PGH-PGE isomerase activities from intact right-side-out microsomes. The cyclooxygenase active site and the IgG1(cyo-3)-reactive epitope of PGH synthase are on the cytoplasmic surface of the endoplasmic reticulum (20). The present findings indicate that epitope of the antigen reactive with IgG,(hei-7) is also present on the cytoplasmic surface of the endoplasmic reticulum. Thus, the PGH-PGE isomerase precipitated by IgGl(hei-7) is present in the same membrane system as PGH synthase, the site of generation of prostaglandin endoperoxide. A similar correspondence between sites of endoperoxide generation and PGI, and TxAz formation has been observed in studies on PGI, synthase (25) and TxA synthase (26).
While IgGl(hei-7) is directed against a protein that likely catalyzes PGE, formation in vesicular gland, it is not possible to make the same case for the protein reactive with IgGl(hei-26). This latter antigen is also located in glandular epithelia of vesicular gland. However, IgGl(hei-26) precipitates PGH-PGE isomerase activity only from solubilized microsomes. Thus, it is unclear whether or not the IgG1(hei-26)-reactive PGH-PGE isomerase is present in the vicinity of PGHZ synthesis. Moreover, the K,,, for PGH, of the PGH-PGE isomerase precipitated by IgG,(hei-26) is 150 PM, which is 5-10 times the K , values for other PGHz-metabolizing enzymes (15,(27)(28)(29). It is possible that the PGH-PGE isomerase activity of the protein precipitated by IgGl(hei-26) is actually a spurious activity associated with a protein having another primary function. A precedent for this has been noted with serum albumin which will catalyze the synthesis of PGD, at an appreciable rate (30, 31).
A perplexing result of our studies is that the cellular distribution of the IgGl(hei-7)-reactive epitope, as determined by immunofluorescent staining (Miniprint), is inconsistent with the known distribution of sites of PGE, synthesis in tissues other than vesicular glands. The basis for these apparent discrepancies needs to be explored. At present, the most likely possibilities are that IgGl(hei-7) interacts with related epitopes of other proteins lacking PGH-PGE isomerase activity and/or that there may be tissue-specific PGH-PGE isomerase isozymes.