Localization of Arachidonate 12-Lipoxygenase in Parenchymal Cells of Porcine Anterior Pituitary*

12-Lipoxygenases oxygenate arachidonic ducing IBS-hydroperoxy


12-Lipoxygenases
oxygenate arachidonic acid producing its IBS-hydroperoxy derivative and are well known as platelet and leukocyte enzymes. When a peroxidase-linked immunoassay of the enzyme according to the avidin-biotin method was applied to the cytosol fractions from various parts of porcine brain, a considerable amount of the enzyme was found in the anterior pituitary.
The enzyme level (about 200 ng/mg cytosol protein) corresponded to about 6% of the enzyme content in porcine peripheral leukocytes. Posterior and intermediate lobes showed about one-tenth of the enzyme level of anterior pituitary.
Other parts of porcine brain contained the 12-lipoxygenase in amounts below 7 ng/mg cytosol protein.
Furthermore, various octadecapolyenoic acids were oxygenated almost as fast as the arachidonate 12-oxygenation. When anterior pituitary was investigated immunohistochemically with anti-12-lipoxygenase antibody, most of the immunostained cells were certain parenchymal cells with granules, which were not blood cells. These biochemical and immunohistochemical results provide a good reason for considering that l%-lipoxygenase does play an important role in pituitary function.
Arachidonate 12-lipoxygenase has been found in various mammalian tissues, especially in platelets (1,2) and leukocytes (3,4). As pointed out by Brash (5), "the pathways of arachidonic acid metabolism that are understood lead to the formation of potent biological mediators such as the prostaglandins and leukotrienes.
However, there is no obvious analogy to these mediators in the 12-lipoxygenase pathway." Na-  -keto-5,8,10,14-eicosatetraenoic acid (8) and 8-hydroxy-11,12-epoxy-5,9,14-eicosatrienoic acid (9). Furthermore, 12S-HETE (but not 12R-HETE) was proposed to participate in the expression or activation of a tumor cell glycoprotein receptor (10). Generalization of these intriguing observations to many other mammalian tissues has not yet been established. Thus, the general physiological function of the 12lipoxygenase pathway is still an open question.
Previously, we prepared monoclonal antibodies against the 12-lipoxygenase of porcine leukocytes and developed a peroxidase-linked immunoassay for determination of the enzyme amount in various porcine tissues (11). However, the occurrence of the 12-lipoxygenase in brain remained unclarified due to the insufficient sensitivity of the assay. As detected by transformation of exogenous arachidonic acid, the 12-lipoxygenase activity was found previously in pituitary (12-16), pineal gland (16), and other parts of the rat brain (17). These previous observations do not necessarily indicate the occurrence of 12-lipoxygenase in the parenchymal cells of these organs rather than contaminating or infiltrating blood cells. Recently, we improved the enzyme immunoassay of 12-lipoxygenase and increased its detectability by introducing the solid-phase and avidin-biotin methods and applied the modified sensitive assay to detect the 12-lipoxygenase in various parts of porcine brain. In addition, we performed immunohistochemical studies on the distribution and localization of 12lipoxygenase in brain tissues. was used for dilution of the purified 12-lipoxygenase, the cytosol fraction of various parts of porcine brain, the biotinylated Zox-2, and horseradish peroxidase-avidin. When the content in a well of immunoplate was discarded after each incubation described below, the well was washed three or four times with phosphate-buffered saline, pH 7.4, containing 0.01% bovine serum albumin. As the first step, the other antibody, lox-l (500 ng in 100 ~1 of 50 mM carbonate buffer at pH 9.5), was placed in a well of Immunoplate I (96 wells/plate), and kept at 4 "C overnight for its binding to the well. The second step was the addition of 400 ~1 of 0.5% bovine serum albumin (dissolved-in phosphate-buffered saline at pH 7.4) to the well, followed by incubation at 37 "C for 30 min. At the third step, the sample containing 12-lipoxygenase (100 ~1) was added to the well, and the mixture was kept at 37 "C for 2 h. As the fourth step, biotinvlated lox-2 (50 np in 100 ~1) was added. and the mixture was incubated at 37 "C for-l h. The fifth step was the addition of a peroxidase-avidin conjugate (100 ng in 100 ~1   1A). A minor peak (retention time, 16.4 min), which was coeluted with authentic SHETE, was also found. A peak corresponding to 5-HETE was hardly detected even in the presence of calcium ion (24) and ATP (25), which are known as 5-lipoxygenase activators. When arachidonic acid was incubated with the cytosol fraction of cerebellum, no significant amount of 12-HETE was detected in sharp contrast to the anterior pituitary (Fig. 1C). The specific 12-lipoxygenase activity of the cytosol fraction of anterior pituitary was calculated from the experimental result presented in Fig. 1B and was found to be 5.0 nmol/5 min/mg protein at 24 "C with arachidonic acid as a substrate. This value corresponded to about 7% of the activity of the leukocyte cytosol. The ratio was close to the above-mentioned value of 6% determined by the enzyme immunoassay.

12-Lipoxygenase of Porcine Anterior Pituitary
The 12-HETE synthesized by the cytosol fraction of anterior pituitary, was purified by reversephase HPLC and then derivatized to its methyl ester. Absolute configuration of the 12-HETE methyl ester was determined by chiral-phase HPLC as described in Ref. 26. The ratio of the optical isomers (S/R) was approximately 96:4. As examined by reverse-phase HPLC monitoring at 235 nm for a conjugated diene of oxygenated products, the cytosol fraction of anterior pituitary was active with the following three octadecapolyenoic acids: linoleic acid, 5.5 nmol/5 min/ mg protein at 24 "C (110% of the activity with arachidonic acid); a-linolenic acid, 4.6 (92%); y-linolenic acid, 5.7 (114%). The products from these three fatty acids were chromatographically indistinguishable from the oxygenated products obtained by the purified enzyme of porcine leukocytes (18).

Localization of 12-Lipoxygenase in the Parenchymal Cells of Porcine Anterior
Pituitary-The specificity of the polyclonal anti-12-lipoxygenase antibody to be utilized in the following immunohistochemical studies was examined by immunoblotting. The cytosol fraction of porcine anterior pituitary showed a major colored band (Fig. 2) acid. Arachidonic acid (25 pM) was allowed to react at 24 "C for 5 min with the cytosol fractions of porcine anterior pituitary (3.5 mg of protein) (B) and cerebellum (4.7 mg protein) (C). The reaction mixture (1.0 ml) contained 50 mM Tris-HCl buffer at pH 7.4 and 2 fiM 13-hydroperoxy-9,11-octadecadienoic acid (added to abolish a lag phase of the 12-lipoxygenase reaction). The borohydride-reduced products were analyzed by reverse-phase HPLC on a Tosoh ODS-120T column equipped with a Waters HPLC system. The solvent system was a mixture of methanol/water/acetic acid (80:20:0.01) at a flow rate of 1.0 ml/min. A, authentic compounds were applied: 13hydroxy-9,11-octadecadienoic acid (peak I) derived from the 13hydroperoxy acid, 15-HETE (peak 2), 12-HETE (peak 3), 5-HETE (peak 4), and 15-hydroxy-11,13-eicosadienoic acid (peak 5) included as an internal standard.
position corresponding to a molecular weight of about 68,000, which was slightly lower reproducibly than that of the purified 12-lipoxygenase of porcine leukocytes (72,000) (18). The band was not detected in the absence of the antibody. The anterior pituitary cytosol was incubated with a monoclonal anti-12lipoxygenase antibody 10x-2, and 12-lipoxygenase was immunoprecipitated with the aid of protein A-bearing S. aureus. Immunoblotting of the 12-lipoxygenase-free cytosol fraction thus prepared failed to detect the above-mentioned positive band by the use of the polyclonal antibody. Disappearance of the major band in this experiment supported its identity with 12-lipoxygenase protein. It was possible that the 12-lipoxygenase contained in porcine pituitary was derived from infiltrating or contaminating leukocytes. Therefore, the localization of 12-lipoxygenase in porcine pituitary was investigated immunohistochemically using the polyclonal anti-12-lipoxygenase antibody. As examined by light microscopy, about 7% of anterior pituitary parenchymal cells found in Fig. 3A were positively stained. These cells were distinguishable from granulocytes in blood vessels, which were also positively stained. No stained cells were found in anterior pituitary in a control experiment with non-immunized rabbit IgG at the same concentration (Fig.  3B). For the immunoabsorption test, the IgG fraction of the anti-12-lipoxygenase antiserum was incubated with the purified leukocyte 12-lipoxygenase at 4 "C overnight, and the IgG fraction thus treated failed to stain the parenchymal cells of anterior pituitary.
As shown in Fig. 4 croscopy showed that the positive cells of anterior pituitary contained many cytoplasmic granules, and 12-lipoxygenase was mainly localized in the cytoplasm of the stained cells. In the middle lobe a small number of immunostained cells were found, and they were distinguishable from granulocytes. In contrast, the posterior lobe showed no stained parenchymal cells. Only intravascular granulocytes were immunostained.

DISCUSSION
Our improved peroxidase-linked immunoassay of 12-lipoxygenase allowed the determination of 12-lipoxygenase content in porcine brain, which was impossible by the previous less sensitive method (11). The highest enzyme content was found in anterior pituitary.
In agreement with this observation, incubation of the cytosol fraction of porcine anterior pituitary demonstrated the 12-lipoxygenase activity with exogenous arachidonic acid as a substrate. Several previous papers described the conversion of exogenous arachidonic acid to 12-HETE in rat pituitary. Pilote et al. (12) incubated 50 PM ['"C] arachidonic acid for 10 min with rat anterior pituitary as such. 12-HETE was produced as the major product from added radioactive arachidonic acid in a 0.1% conversion rate. When our group reported a 12-lipoxygenase activity in rat pineal gland, we also detected a very low rate of conversion from 10 FM arachidonic acid to 12-HETE (only less than 0.2%) in rat pituitary (16). Vanderhoek et al. (13) prepared a gonadotrophenriched cell fraction from rat pituitary.
Incubation of this cell fraction with radioactive arachidonic acid produced a variety of metabolites of the cyclooxygenase pathway and several lipoxygenase products including 12-HETE as a minor component. The cyclooxygenase products were predominant by at least 3-4-fold over the lipoxygenase products. In all these earlier works the 12-oxygenation of arachidonic acid was observed to a minor extent, and it was unclear whether such a minor conversion was attributed to the parenchymal cells of pituitary or to the contaminating platelets or leukocytes. Indeed, our recent immunohistochemical work demonstrated the distribution of 12-lipoxygenase in porcine alimentary tract and lymphatic organs in addition to leukocytes in peripheral blood. Only various types of resident leukocytes were positively stained in the former organs (19). By the use of cloned tumor cells of rat pituitary, Rabier et al. (15) mentioned the production of 12-HETE (in a yield of 14% of all the products) together with 11,9%,respectively) from exogenous arachidonic acid. The production of a mixture of HETEs suggests a lack of stereospecificity usually associated with non-enzymic oxygenation.
Alternatively, a small percentage of 12-HETE could arise as a minor product of a 15-lipoxygenase reaction as is known to occur with the reticulocyte enzyme (27).
Thus, these previous papers described a minor extent of 12-HETE production from a quantitative view and a heterogeneous composition of products from a qualitative view. In contrast, our present work demonstrated clearly the occurrence of 12-lipoxygenase in porcine pituitary on the basis of not only the arachidonate transformation but also the peroxidase-linked immunoassay. Furthermore, we demonstrated the localization of the I2-lipoxygenase in some parenchymal cells of anterior pituitary rather than contaminating or resident blood cells. We attempted to identify the 12-lipoxygenase-containing pituitary cells by the use of antibodies specific for luteinizing hormone, follicle-stimulating hormone, thyroid-stimulating hormone, and adrenocorticotropic hormone of porcine origin. We have not yet tested antibodies against growth hormone and prolactin.
The use of these antibodies has not clarified as yet whether the 12-lipoxygenase-containing cells are composed of several subgroups each secreting different hormone or they are associated with a group of pituitary cells secreting a specific type of hormone.' Selective production of S-isomer of 12-HPETE is known as a common feature of mammalian 12-lipoxygenase of platelet (1) and leukocyte (26), whereas cytochrome P-450 was reported to produce predominantly 12R-HETE over 12Sepimer in a ratio of 81:19 (28). Since the 12-lipoxygenase of porcine anterior pituitary produced selectively 12S-HPETE, the enzyme is distinguishable from the cytochrome P-450 as a 12R-hydroxylase, which might be contained in the microsome of porcine anterior pituitary.
Capdevila and others (14) reported the production of 12-HETE without a mention of stereochemistry as one of various oxygenated products when the microsome of rat anterior pituitary was incubated with arachidonic acid. It was proposed that there were two types of 12lipoxygenase, a leukocyte type and a platelet type (4,26,29).
The high reactivities of anterior pituitary I2-lipoxygenase with linoleic and linolenic acids were similar, if not identical, to the activities of porcine leukocyte 12-lipoxygenase with these octadecapolyenoic acids (18). In contrast, platelet 12-lipoxygenase was almost inactive with these octadecapolyenoic acids (2,26,29).
Thus, the 12-lipoxygenase of porcine anterior pituitary is a leukocyte-type enzyme in terms of the catalytic properties. Incidentally, it should be noted that 12-lipoxygenase was not detected in porcine platelets by the enzyme immunoassay and the activity assay (11). It cannot be ruled out that the platelet-type 12-lipoxygenase is also present in the cytosol of porcine anterior pituitary in such a small amount that does not significantly affect the results of substrate specificity study. Incubation of arachidonic acid with the cytosol fraction of various parts of porcine brain except for pituitary did not produce 12-HPETE.
Thus, in these areas of porcine brain there was no significant amount of the platelet-type 12-lipoxygenase which was not detected by our enzyme immunoassay.
A minor peak of 15-HETE observed upon HPLC (Fig. lB), may indicate the presence of 15lipoxygenase. An alternative interpretation is the l&oxygenase activity of 12-lipoxygenase which was clearly demonstrated with 5-HETE and the porcine leukocyte 12-lipoxygenase (18). In our study on the tissue distribution of 12-and 5lipoxygenase (11, 30), we have not found a 15-lipoxygenase from various porcine tissues.
Several papers have reported stimulatory effects of arachidonate metabolites on the hormone release from anterior pituitary cells of rut: luteinizing hormone by 5,6-epoxyeicosatrienoic acid (31), leukotriene Cq (32) and 5-HETE (33), and prolactin by 5-HETE (34). The presence of 12-lipoxygenase in porcine anterior pituitary suggests a possible role of 12-HPETE or its metabolites in the hormone release from porcine pituitary cells.
A previous finding that 12-HETE stimulated the release of luteinizing hormone-releasing hormone from rut median eminence (35) prompted us to carefully examine the 12-lipoxygenase content in porcine median eminence. However, only a * N. Ueda et al., unpublished observation. very small amount of the enzyme was detected in the porcine hypothalamic area containing median eminence. Earlier we screened lipoxygenase activities in rut brain tissues using l"Clabeled arachidonic acid. The 12-lipoxygenase activity was by far the highest in pineal gland, and less than 5% of the activity was found in pituitary gland and hypothalamus (16). In relation to this observation, 12-HPETE, but not 12-HETE, stimulated melatonin synthesis in rat pineal gland (36). However, in the present work the enzyme immunoassay showed only a low enzyme content in porcine pineal gland. The cytosol of porcine pineal gland did not show the 12-lipoxygenase activity as examined by incubation with arachidonic acid. Thus, porcine pineal gland does not appear to contain significant amount of 12-lipoxygenase of both the leukocyte-type and the platelet-type.