Brain Research, 543 (1991) 15-24 Elsevier
15
BRES 16372
Localization of prostaglandin endoperoxide synthase in neurons and glia in monkey brain Shogo Tsubokura 1'2, Yasuyoshi Watanabe ~, Hiroaki Ehara 4, Kazuyuki Imamura 1, Osamu Sugimoto 2, Hiroyuki Kagamiyama 3, Shozo Yamamoto 4 and Osamu Hayaishi 1 1Department of Neuroscience, Osaka Bioscience Institute, Osaka (Japan), Departments of 2obstetrics and Gynecology and 3 Medical Chemistry, Osaka Medical College, Osaka (Japan) and 4Departmentof Biochemistry, Tokushima University School of Medicine, Tokushima (Japan) (Accepted 18 September 1990) Key words: Prostaglandin endoperoxide synthase; Monoclonal antibody; Immunohistochemisty; Monkey; Brain; Neuron; Glia
The localization of prostaglandin endoperoxide synthase in monkey brain was investigated by the immunoperoxidase method using the monoclonal antibody (PES-7) raised against the enzyme purified from bovine seminal vesicle. The frozen sections with 30-gm thickness were employed after the brain was fixed with perfusion of 2% paraformaldehyde in phosphate-buffered saline. The immunoreactivity was most intense in the neurons of cerebral cortex and hippocampus, and was moderate in the neurons of caudate nucleus, putamen, globus pallidus and amygdala, while it was relatively weak in gliai cells in the whole brain regions including the white matter. The majority of neurons showed the immunoreactivity in the somata and proximal dendrites, but exceptionally in the pyramidal cells of the hippocampus, positive staining was also observed in the apical dendrites. In the cerebellum, the immunoreactivity in both neurons and glia was rather faint as compared with that in other regions. Positive staining was not significantly observed in the vasculatures and arachnoid membranes. These findings indicate that most of neuronal and glial cells in monkey brain contain the enzyme of the rate-limiting and initial step of the biosynthesis of prostaglandins which regulate a variety of neural functions.
INTRODUCTION The initial and c o m m o n step in the biosynthesis of prostaglandins (PGs) is the fatty acid cyclo-oxygenase reaction (arachidonic acid - , PGG2) followed by the PG hydroperoxidase reaction ( P G G 2 - , PGH2) , both of which are catalyzed by the same enzyme referred to as prostag!andin endoperoxide synthase (PES: EC 1.14.99.1; also known as cyclo-oxygenase) 12. Biosyntheses of PGD2, PGE2, PGF2a , PGI2, and Thromboxane A 2 (TXA2) require this unstable PG endoperoxide, P G H 2, as a c o m m o n substrate for each PG and TX synthase. PES was purified from microsomes of bovine 12 and ovine 4'28 vesicular gland, and its molecular and catalytic properties have been investigated 2,4,5,1°.12,17, 19,28,29,36
Recently, PGD2, PGE2, and PGF2a have been identified as major prostaglandins formed in the brain of several mammalian species, including rat 14 and human 18. A variety of central actions of PGs have been demonstrated, such as the regulation of sleep-wake 3, body temperature 11'26, luteinizing hormone-releasing hormone secretion 7'2°, and modification of olfactory sensation 32. Furthermore, the binding protein specific for each P G was found in the synaptic membrane of rat 9'23 and human
brain 31. The localization of these P G receptors by using in vitro autoradiography 3°'33-38 revealed the specific and distinct localization of each P G receptor with close relation to the respective central actions 34. The biosynthetic sites of each P G should also be elucidated for total understanding of the functional sites of PGs. Urade et al. 27 showed the localization of P G D synthase in the rat brain. However, little is known about PES in the central nervous system (CNS) because of a very low content of the enzyme and occasionally of the possible occurrence of certain endogenous enzyme inhibitor(s) 37. The localization of PES is of great importance for determining the sites of PG biosynthesis in CNS. Indeed, in some cell types, PES rather than each PG synthase has been shown to be the rate-limiting step in each PG production 13. We therefore examined the localization of PES in the m o n k e y brain by indirect immunoperoxidase staining using a monoclonal antibody. This is the first report of the existence of PES in the neurons.
MATERIALS AND METHODS Materials Monoclonal antibody against PES (PES-7) was prepared against
Correspondence: Y. Watanabe, Osaka Bioscience Institute, 6-2-4 Furuedai, Suita-shi, Osaka 565, Japan. 0006-8993/91/$03.50 © 1991 Elsevier Science Publishers B.V. (Biomedical Division)
16 the purified PES from bovine vesicular gland as described previously 37. Normal mouse IgG was purchased from Cappel Lab. Inc.; Vectastain ABC kit, from Vector Lab. Inc.; [114C]Arachidonic acid (59.6 mCi/mmol), from Amersham; Protein A-Sepharose CL-4B, from Pharmacia; Silver Stain Kit and 4chloro-l-naphthol, from Wako (Tokyo); precoated silica gel 60F 254 glass plates for TLC, from E. Merck; molecular weight standard mixture, Protein Assay Kit, and nitrocellulose membrane filters, from Bio-Rad; and bovine serum albumin (Fraction V), from Sigma.
Preparation of solubilized enzyme Solubilized enzyme was prepared as described previously 12. The monkey was anesthetized with ketamine hydrochloride followed by pentobarbital and was perfused with 10 mM sodium phosphate buffered saline (pH 7.4) (PBS) via the left ventricle of the heart. The monkey brain was rapidly removed and homogenized in 5 vols. of 20 mM potassium phosphate buffer (pH 7.3) containing 0.1 mM phenylmethylsulfonyl fluoride using a Potter-Elvehjem glass homogenizer with a Teflon pestle. The homogenate was centrifuged at 10,000 g for 10 rain and the resulting supernatant was further centrifuged at 164,000 g for 60 min. The pellet was dispersed in a half of the original volume of 20 mM potassium phosphate buffer (pH 7.3) containing 1% Tween 20. After standing at 4 °C for 30 min, the mixture was centrifuged at 164,000 g for 60 min, and the resulting supernatant was used as the solubilized enzyme.
Immunoaffinity column analysis The solubilized enzyme (4.8 mg protein/ml, 5 ml) was incubated with PES-737 (660/ag protein/ml, 1 ml) at 4 °C for 2 h, then the mixture was applied to Protein A-Sepharose CL-4B column (bed size = 7 mm i.d. x 4 cm). After the column was washed with 3 bed volumes of 0.1 M borate buffer (pH 8.2) containing 0.14 M sodium chloride, PES-7-PES complex was eluted with 5 mi of 0.1 M acetate containing 0.14 M sodium chloride. The pH of the eluting solution was immediately adjusted to approximately 7.0 using 1.0 M Tris-base. Enzyme activity in both passed through and eluted fractions was assayed according to the method described by Miyamoto et al. 12. Protein was determined by the method of Bradford I with bovine serum albumin (BSA) as standard.
lmmunoblotting analysis Sodium dodecyi sulfate-polyacrylamide gel electrophoresis (SDSPAGE) was carried out in 4-20% gradient acrylamide by the method of Laemmli s. The solubilized enzyme from monkey brain was subjected to SDS-PAGE at 40 mA for 2 h and then the proteins were transferred electrophoretically at 2 mA/cm 2 for 2 h to a nitrocellulose membrane by Trans Blot (Bio-Rad) according to the method of Towbin et al. 2s, in the presence of 192 mM glycine, 25 mM Tris-base, and 20% methanol. The nitrocellulose membrane was washed with PBS containing 0.05% Tween 20. Then, the membrane was blocked overnight at 4 °C with 3% (wt/vol) BSA and 0.5% (wt/vol) non-fat dry milk21 dissolved in PBS containing 0.05% Tween 20 to prevent non-specific binding, and was incubated at room temperature with PES-7 (20/ag/ml in PBS containing 0.05% Tween 20) in the presence of 1% BSA for 2 h. Thereafter the membrane was washed as above, and then incubated with the horse biotinylated anti-mouse IgG antibody in PBS at room temperature for 30 rain. The membrane was washed and treated with Vectastatin ABC kit (Vector Lab.). After being washed, the nitrocellulose membrane was subjected to the peroxidase reaction using 4chloro-l-naphthol as substrate for 3 rain. Tissue preparation
Macaca fuscata fuscata (male, 6.0-8.0 kg) was anesthetized with ketamine hydrochloride (Ketarai) and pentobarbitai and perfused via the left heart ventricle with cold PBS (5 1), then 2% paraformaldehyde (PFA) in PBS (3 I), followed by PBS containing 5% sucrose (3 !). The brain was rapidly removed, cut into 5-mm thick coronal blocks using microtome blades (Feather $35 type) and
immersed in 30% sucrose in PBS. After the specimens were sunk in the solution at 4 °C, they were frozen on dry-ice. Frozen sections (30/zm thickness) were cut in a microtome (YAMATO, TU-213) consecutively and thaw-mounted onto ovalbumin-coated glass slides (76 x 52 mm).
lmmunohistochemical staining The mounted sections were air-dried at room temperature for 90 rain. The dried sections were then fixed in acetone at 4 °C for 10 min and washed 3 times with PBS. The sections were covered with 20% normal horse serum, diluted with PBS containing 0.3% Triton X-100 (TPBS), as a blocking agent and kept at room temperature for 60 min. The sections were then incubated with PES-7 (20/zg/ml in TPBS containing 1% BSA) at 4 °C overnight, followed by 3 washings with TPBS. Then, the sections were incubated with biotinylated horse anti-mouse IgG diluted with TPBS followed by incubation with the avidine-biotin-peroxidase complex at room temperature for 30 rain. After each step, the sections were washed 3 times with TPBS. After the final wash, the sections were bufferized with 50 mM Tris-HCl (pH 7.2), stained by incubation with diaminobenzidine tetrahydrochloride (0.5 mg/ml) in 0.02% H20 2 for 4 min. The sections were then washed with water, dehydrated and mounted. The samples were examined and photographed using a VANOX-S microscope equipped with a camera (Olympus Co., Ltd.). Non-immune mouse IgG (Cappel) rechromatographed on a Protein A-Sepharose CL-4B column was used as a negative control. For assessment of the specificity of immunoreactivity for PES, PES-7 solution preabsorbed with the purified PES from sheep vesicular gland was used for staining as another negative control. Some of the sections were further fixed in 2% PFA at 4 °C overnight, then stained by 0.1% thionine (Nissl staining). Immunoreactivity with PES-7 was scored for evaluation of PES
A
B
200 K - ~ " ~
116K 97K
66K
43K--~llb
Fig. 1. Immunoblotting with monoclonal antibody against PES after SDS-polyacrylamide gel electrophoresis of the solubilized enzyme fraction from the monkey brain microsome. Anode and origin are at the top of the figure. The arrowhead indicates the position of the tracking dye. Lane A, silver staining of the monkey brain solubilized fraction; lane B, immunobiotting incubated with PES-7 as described under Materials and Methods.
17
Fig. 2. PES-immunoreactivity in the monkey temporal cortex. A and D: immunoperoxidase staining with the monoclonal antibody against PES in the coronal section of the temporal cortex. The section (30/zm thickness) was incubated with 20 ~ug/ml PES-7 as described under Materials and Methods. B and E: control staining in the consecutive section with non-immune mouse IgG. C and F: Nissl staining with 0.1% thionine in the consecutive sections~ Scale bar (A,B,C,) = 125/zm. D,E, and F higher magnification in layer III. Scale bar = 50 jum.
18
Fig. 3. PES-immunoreactivity in the monkey hippocampus. A, B, D and E: immunoperoxidase staining with the monoclonal antibody against PES in the coronal section of the hippocampus. C: control staining with non-immune mouse IgG in place of PES-7 in the consecutive section. Bar = 500/zm (A). Bar = 100/zm (B.C). D: higher magnification of the view in CA3 region. Bar ~ 50 #m (D). Bar = 12.5/~m (E).
19 content by (a) the percentage of positive stained cells and (b) the staining intensity (+ to ++ +) visually under the light microscopic examination. The intensity of specific staining was expressed as weak but detectable above control (+), intense (+ +), and markedly intense (+ + +). RESULTS
Specificity of the monoclonal antibody Specificity of the monoclonal antibody (PES-7) was evaluated by SDS-PAGE and immunoblotting. As shown in Fig. 1, PES-7 gave a single line of molecular weight of 75 kDa with the solubilized fraction of monkey brain microsomes. No other proteins cross-reacted with this antibody. When the homogenate of monkey brain prepared as described under Materials and Methods was transblotted after SDS-PAGE, no immunoreactive bands were observed because of the relatively low and undetectable amounts of PES in the homogenate. Furthermore, when the solubilized enzyme fraction incubated with PES-7 was applied to a Protein A-
Sepharose CL-4B column, about 62% of the original PES activity was recovered in the eluted fraction as an immune complex. No PES activity was detected in pass through or washing fraction. Then the eluted fraction was subjected to SDS-PAGE; only PES and PES-7 could be detected by Coomassie brilliant blue staining (data not shown). These results indicate that PES-7, the monoclohal antibody employed in this immunohistochemical study, is highly specific to PES in the monkey brain.
Immunohistochemical brain
localization of PES in monkey
When PES-7 was utilized for immunoperoxidase staining to reveal the cellular localization of PES in the monkey brain, intense staining was observed in a number of neurons and glias. In contrast, no positive staining was detected with non-immune IgG. The positive immunoreactivity in the hippocampus and cerebral cortices was titrated by using purified PES from sheep vesicular gland and 124 /zg/ml of the purified PES could completely
Fig. 4. PES-immunoreactivity in the monkey cerebellum. The contrast of these photographs is graded up compared with the photos representing the immunoreactivity in other brain regions (Figs. 2 and 3). A and C: immunoperoxidase staining with the monoclonal antibody against PES. B and D: control staining with non-immune IgG in the consecutive section. Bar = 200 ~m (A,B); bar = 50 ~m (C,D).
21 diminish the positive staining (molar ratio of PES/PES-7 = 12). The intensity of the PES-immunoreactivity was somewhat different among various brain regions. Fig. 2A shows a typical pattern of PES-immunoreactivity in the temporal cortex. Intensely positive staining was found throughout a n u m e r o u s n u m b e r of nerve cells, while glial cells both in the white matter and gray matter showed
TABLE I
Percentage of positive cell# and relative intensity of P ES-immunoreactivity in monkey brain regions The population of immunoreactive cells from more than 300 cells, good-shaped (avoided from the counting error because of the irregular sectioning), in each structure were counted under light microscope, and relative intensity of immunoreactivity was graded from + (weak) to + + + (intense).
weaker and fewer staining. Especially, the nerve cells in layer II showed most intense immunoreactivity. In layers I I - V I , the immunoreactive neurons constituted more than 95% of the n e u r o n s when calculated in comparison with an adjacent Nissl-stained section (Fig. 2C and F), but less than 25% of n e u r o n s were immunoreactive in layer I. A t higher magnification, only n e u r o n a l somata and proximal dendrites showed the immunoreactivity (Fig. 2D). Distal dendrites showed slightly positive, but the cell nuclei did not show any positive staining. In contrast, when n o n - i m m u n e IgG was used in place of
PES-7 in the consecutive sections, the specific staining
Frontal cortex
II III IV V VI w* Temporal cortex I II III IV
staining between PES-7 and control IgG was observed in the vasculatures, especially microvessels and arachnoid m e m b r a n e s . No immunoreactivity was seen in the terminal structure. This pattern and density of the immunoreactivity were similar to those in other cortical regions, such as frontal, parietal, and occipital cortices. The immunohistochemical staining in the hippocampal
formation is shown in Fig. 3. In this region, the granule cells of dentate gyrus and pyramidal cells of A m m o n ' s
horn were strongly immunoreactive (Fig. 3A), and some of the glial cells were less immunoreactive. All of the granule cell bodies showed intense immunoreactivity (Fig. 3B). The pyramidal cells at higher magnification showed the PES-immunoreactivity not only in the neuronal somata and proximal dendrites, but also in apical dendrites of a similar intensity (Fig. 3D and E). The positive staining in the apical dendrites was only seen in C A 2 and CA3 regions. Fig. 4 shows the pattern of staining in the m o n k e y cerebellum. The immunoreactivity (Fig. 4A and C) was close to the control level (Fig. 4B and D) with noni m m u n e mouse IgG in place of PES-7 in both neurons and glial cells. Moreover, the elongation of the time of D A B reaction could not extend the difference of the densities between PES-7 and control stainings. In the molecular layer, the stellate cells and a small n u m b e r of glia showed positive immunoreactivity as compared with
Parietal cortex
Occipital cortex
Glial cell
%
Intensity
%
23 100 95 100 97 100
+ +++ ++ ++ ++ ++
-
VI
13 100 100 100 100 100
w*
-
v
was not observed in either nerve cells or glial cells (Fig.
2B and E). No significant difference in the intensity of
I
Nerve cell
I II III IV v vI w* I II III
IV V VI W* Globuspallidus Putamen Caudate Amygdala Thalamus Hypothalamus Hippocampus granular cell layer CA4 CA3 CA2 CA1 dentategyrus Cerebellum molecular layer Purkinje cell layer granularlayer w* Pons Medulla oblongata Spinalcord
20 100 100 100 100 100 -
-
+ +++ ++ ++ ++ ++ -
+ +++ ++ ++ ++ ++ -
Intensity
22 19 13 11 10 11
+ + + + + +
54
+
22 7 5 22 14 7
++ + + + + +
48
+
46 18 8 11 10 19
+ + + + + +
41
+
14 100 100 99 100 99
+ +++ ++ ++ ++ ++
20 65 92 78 75 100
++ ++ ++ ++ ++ +++
11 19 28 17 30 15
+ + + + + +
92 87 84 80 100
+++ +++ +++ +++ +++
21 30 17 29 n.d.
+ + + +
40 63 100
+ + +
44 + n.d. n.d.
-
41 74 63
-
18
+
8
+
5 12 10 10
+ + + +
50
+
-
67
+
+ + +
13 33 76
+ + +
* White matter; n.d.: not detected.
Fig. 5. PES-immunoreactivity in the C A 2 region of hippocampus (A), a part of caudate nucleus (B), layers III and IV of striate cortex (C), N. basalis Meynerti (D, shown by an arrow), the paraventricular, lateral superior central, and dorsomedial n. (ME)) of thalamus (E), and substantia nigra pars compacta and pars diffusa (F). × 5 0 (A, B and C), ×25 (D and E), and × 1 0 (F). Fig. 6. PES-immunoreactivity in the glial cells. A: comparison of stainability between neurons and glia in the C A 3 region of hippocampus (oriens layer and a part of alveus). × 150. B: glial cells in the corpus callosum. ×300.
22 the control staining. The Purkinje cells and Bergmann's glia were not positive. In the granule cell layer, almost all granule cells showed very weak immunoreactivity. Fig. 5 shows the PES-immunoreactivity in various brain regions with special reference to neurotransmitter systems. The immunoreactivity was abundant in various neurotransmitter systems. In the caudate nucleus (Fig. 5B), the intense staining was observed in the cytosol of neurons but not in the distal dendrites in contrast to the immunoreactivity in the pyramidal cells in CA2 and CA3 regions of the hippocampus (Fig. 5A). The immunoreactivity was also positive in the glutamatergic (in the striate cortex, Fig. 5C), cholinergic (N. basalis Meynerti, Fig. 5D), serotonergic (dorsomediai N. of thalamus, Fig. 5E), and dopaminergic (substantia nigra pars compacta and pars diffusa, Fig. 5F) neurons. However, the density of the immunoreactivity in the neurons in the same brain regions was different among the nuclei; for example, the ventromedial nucleus of hypothalamus showed less immunoreactivity than the surrounding hypothalamic nuclei (data not shown). Table I shows the summary of the distribution of PES immunoreactivity. In the whole brain, the hippocampal formation, cerebral cortex and caudate nucleus showed intense immunoreactivity, whereas midbrain, cerebellum, spinal cord, and white matter were weakly stained. The percentage of the immunopositive cells compared with the adjacent Nissl-stained sections was high in the case of neurons but was relatively low in the case of glias. Relatively low intensity of the immunoreactivity was seen in the glias (Fig. 6A), but still it was positive and clear from the surrounding tissues. Oligodendrocytes (Fig. 6B) seemed to contain greater amounts of PES-immunoreactivity than astrocytes, judged from the cell contour and the regions of their presence. We employed 5 monkey brains in this study. The results showed little heterogeneity in the pattern and density of localization of PES immunoreactivity among the brain specimens. DISCUSSION In the present study, we demonstrated the immunohistochemical localization of PES in the monkey CNS. The specificity of the immunostaining was assessed by using the control with non-immune mouse IgG and by blocking the immunoreaction with the purified PES. The protein immunoblotted after SDS-PAGE of monkey cerebral cortex supernatant showed a single band of the molecular weight of around 75 kDa which is essentially similar to the molecular weight of PES purified from bovine or sheep vesicular gland (Fig. 1). In addition, the roughly estimated regional distribution of PES-immuno-
reactivity was close to that in the bovine brain reported by Yoshimoto et al. 3v by using enzyme-linked immunoassay. Since the monoclonal antibody (PES-7) employed in this study does not cross-react with rat brain PES and rabbit kidney PES, we initially used ox brain for immunostaining. However, the conditions of ox brain obtained from a local slaughterhouse were not satisfactory for constant staining and brain structure. The monkey brain sections were available because of the autoradiographic studies on PG receptors in our laboratory 33'34 and were positively stained by the monoclonal antibody. Essentially similar results were obtained using ox brain sections, especially in the hippocampus, caudate nucleus and cerebral cortices. This is the first direct evidence demonstrating the existence of PES in the neurons as well as glial cells. Since the content of PES in CNS and the sensitivity of measurement of PES using [14C]arachidonic acid are low, no one has so far succeeded in detecting PES activity after fractionation of cells by collagenase and pronase treatment. Concerning the immunostaining of PES in CNS, only one report by Smith et al. 24 is available. Their pioneering work using polyclonal antibody, however, focused on the cerebellar cortex of pig, guinea pig, rat, mouse, cow, rabbit and sheep brains and failed to detect an adequate PES-immunoreactivity in neurons. They reported the positive immunoreactivity only in the vasculatures and Bergmann glial cells surrounding Purkinje cells in the porcine, ovine and bovine cerebellar cortices. Our results of immunostaining also showed relatively low density (or close to the control level) in the cerebellar neurons and glias (Fig. 4). In addition, another monoclonal antibody for PES (PES-5) employed in our preliminary work did not significantly bind any structure of ox and monkey brains after no fixation or 2-4% paraformaldehyde fixation. Furthermore, when we tested a variety of fixatives using ox brain and PES-7 such as no fixation, Bouin, Zamboni, various concentrations of paraformaldehyde and glutaraldehyde, intensely positive staining was obtained only by using 1-6% paraformaldehyde. Furthermore, only when using the present fixative, the positive staining in the apical dendrites was observed in the hippocampus. Thus the discrepancy between their results and ours could be interpreted by the species difference, the difference of the antibody employed, and especially, the difference of fixation condition, in addition to the fact that they tried to stain only cerebellar sections. The lack of positive PES-immunoreactivity in vasculatures in the present study might indicate the sensitivity of the staining, namely, below the threshold level of positive staining by this method. However, another
23 possibility lies in the use of perfused fixation of the brain tissues, since the vasculature is too much fixed by such a type of fixation as compared with the extent of fixation on the substantial tissues. In fact, in our experiments, too much fixation caused the low stainability for PES even in the neurons and glia in the case of 6% paraformaldehyde and/or 1% glutaraldehyde fixation. Ubiquitous but heterogenous distribution of PESimmunoreactivity in the neurons and glia was obtained in this study. Although the cell volume should be carefully estimated, the PES-immunoreactivity seemed to be more dense in neurons than that in glia. Also, the immunoreactivity was not limited either in the specific neurons of some brain structures or in the specific neurotransmitter systems. This finding is opposite to the limited and distinct localization of P G D 2 and P G E 2 receptors in the m o n k e y brain 33'34. Therefore, the localization of each PG synthase following this PES reaction should be the key for determining the responsibility of each PG for the specific neuronal functions. Urade et al. 27 showed the immunoreactivity of P G D synthase in the rat brain with dramatic developmental changes; almost all neurons in the 1-week-old brain and a limited number of neurons in the adult brain. The major site of the immunoreactivity changed from neurons to oligodendrocytes. This type of developmental change is also interesting, since the primary cultured astroglia produce much larger amounts of cyclo-oxygenase metabolites than the primary cultured neurons 6'22. However, a relatively low PES-immunore-
activity was seen in the astrocytes in whole brain regions in the present study. Such a type of developmental study and accurate identification of type 1 and type 2 astrocytes by using antiglial fibrillary acidic protein antibody and other markers is currently under investigation in our laboratory. High density of PES-immunoreactivity in the pyramidal cells in the hippocampus may be supported by the experimental results by Ogawa et al. 15'16. They showed the PGF2a-immunoreactivity in the pyramidal cells of hippocampus after reoxygenation in anoxic rats and after experimental subarachnoid hemorrhage in rats. At that time, blood vessels were positively stained for PGF2a but other cells such as glial cells were all negative. P G E 2immunoreactivity was also demonstrated in the pyramidal cells of rat hippocampus and was remarkably enhanced by administration of convulsive drugs (Fujimoto et al., unpublished results). If one of the rate-limiting steps of prostaglandin synthesis is actually the PES step in the brain, the regulatory mechanisms of PES under such pathological conditions are also the target of future research.
REFERENCES
synaptosomes, Brain Research, 236 (1982) 227-233. 10 Merlie, J.P., Fagan, D., Mudd, J. and Needleman, P., Isolation and characterization of the complementary DNA for sheep seminal vesicle prostaglandin endoperoxidase synthase (cyclooxygenase), J. Biol. Chem., 263 (1988) 3550-3553. 11 Milton, A.S. and Wendlandt, S., A possible role for prostaglandin E~ as a modulator for temperature regulation in the central nervous system of the cat, J. Physiol., 207 (1970) 76-77. 12 Miyamoto, T., Ogino, N., Yamamoto, S. and Hayaishi, O., Purification of prostaglandin endoperoxide synthetase from bovine vesicular gland microsomes, J. Biol. Chem., 251 (1976) 2629-2636. 13 Murota, S., Chang, W.C., Koshihara, Y. and Morita, I., Importance of cyclo-oxygenase induction in the biosynthesis of prostacyclin, Adv. Prostaglandin Thromboxane Leukotriene Res., 11 (1983) 99-104. 14 Narumiya, S., Ogorochi, T., Nakao, K. and Hayaishi, O., Prostaglandin D 2 in rat brain, spinal cord and pituitary: basal level and regional distribution, Life Sci., 31 (1982) 2093-2103. 15 Ogawa, H., Kassell, N.E, Sasaki, T., Hongo, K., Tsukahara, T., Hudson, S.B., Asban, G.I., Tuan, H.L. and Turner, J.C., ImmunohistochemicaUy demonstrated increase of prostaglandin F2a in neurons after reoxygenation in anoxic rats, Prostaglandins, 36 (1988) 891-900. 16 Ogawa, H., Kassell, N.F., Sasaki, T., Nakagomi, T., Hongo, K and Tsukahara, T., Rapid increase of prostaglandin F~ in neurons after subarachnoid hemorrhage in rats: an immunohistochemical study, Acta Neuropathol., 78 (1989) 621-628. 17 Ogino, N., Ohki, S., Yamamoto, S. and Hayaishi, O., Prostaglandin endoperoxide synthetase from bovine vesicular gland
1 Bradford, M.M., A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding, Anal. Biochem., 72 (1976) 248-254. 2 DeWitt, D.L. and Smith, W.L., Primary structure of prostaglandin G/H synthase from sheep vesicular gland determinated from the complementary DNA sequence, Proc. Natl. Acad. Sci. U.S.A., 85 (1988) 1412-1416. 3 Hayaishi, O., Sleep-wake regulation by prostaglandins D 2 and E2, J. Biol. Chem., 263 (1988) 14593-14596. 4 Hemler, M., Lands, W.E.M. and Smith, W.L., Purification of the cyclooxygenase that forms prostaglandins: demonstration of two forms of iron in the holoenzyme, J. Biol. Chem., 251 (1976) 5575-5579. 5 Hemler, M.E. and Lands, W.E.M., Evidence for a peroxideinitiated free radical mechanism of prostaglandin biosynthesis, J. Biol. Chem., 255 (1980) 6253-6261. 6 Keller, M., Jackisch, R., Seregi, A. and Hertting, G., Comparison of prostanoid forming capacity of neuronal and astroglial cells in primary cultures, Neurochem. Int., 7 (1985) 655-665. 7 Kinoshita, E, Nakai, Y., Katakami, H., Imura, H., Shimizu, T. and Hayaishi, O., Suppressive effect of prostaglandin (PG) D z on pulsatile luteinizing hormone release in conscious castrated rats, Endocrinology, 110 (1982) 2207-2209. 8 Laemmli, U.K., Cleavage of structural proteins during the assembly of the head of bacteriophage T4, Nature, 227 (1970) 680-685. 9 Malet, C., Scherrer, H., Saavedra, J.M. and Dray, E, Specific binding of [3H]prostaglandin E 2 to rat brain membranes and
Acknowledgements. We thank Drs. K. Mori and S. Oka of our institute for useful technical advice and discussions, and Ms. M. Tanemura for assistance. This work was supported in part by a Grant-in-Aid (60580158) and a Grant-in-Aid for Special Project Research of Plasticity of Neural Circuits from the Japanese Ministry of Education, Science and Culture, and by grants from the Japanese Foundation of Metabolism and Diseases and Sankyo Co., Ltd.
24 microsomes, J. Biol. Chem., 253 (1978) 5061-5068. 18 Ogorochi, T., Narumiya, S., Mizuno, N., Yamashita, K., Miyazaki, H. and Hayaishi, O., Regional distribution of prostaglandins D2, E 2. and F2~ and related enzymes in postmortem human brain, J. Neurochem., 43 (1984) 71-82. 19 Ohki, S., Ogino, N., Yamamoto, S. and Hayaishi, O., Prostaglandin hydroperoxidase, an integral part of prostaglandin endorperoxide synthesis from bovine vesicular gland microsomes, J. Biol. Chem., 254 (1979) 829-836. 20 Ojeda, S.R. and Campbell, W.B., An increase in hypothalamic capacity to synthesize prostaglandin E 2 precedes the first preovulatory surge of gonadotropins, Endocrinology, 111 (1982) 1031-1037. 21 Plapp, F.V., Rachel, J.M. and Sinor, L.T., Dipsticks for determining ABO blood groups, Lancet, 28 (1986) 1465-1466. 22 Seregi, A., Keller, M. and Hertting, G,, Are cerebral prostanoids of astroglial origin? Studies on the prostanoid forming system in developing rat brain and primary cultures of rat astrocytes, Brain Research, 404 (1987) 113-120. 23 Shimizu, T., Yamashita, A. and Hayaishi, O., Specific binding of prostaglandin D 2 to rat brain synaptic membrane, J. Biol. Chem., 257 (1982) 13570-13575. 24 Smith, W.L., Gutekunst, D.I. and Lyons Jr., R.H., Immunocytochemical localization of the prostaglandin-forming cyclooxygenase in the cerebeUar cortex, Prostaglandins, 19 (1980) 61-69. 25 Towbin, H., Staehelin, T. and Gordon, J., Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications, Proc. Natl. Acad. Sci. U.S.A., 76 (1979) 4350-4354. 26 Ueno, R., Narumiya, S., Ogorochi, T., Nakayama, T., Ishikawa, Y. and Hayaishi, O., Role of prostaglandin D2 in the hypothermia of rats caused by bacterial lipopolysaccharide, Proc. Natl. Acad. Sci. U.S.A., 79 (1982) 6093-6097. 27 Urade, Y., Fujimoto, N., Kaneko, T., Konishi, A., Mizuno, N. and Hayaishi, O., Postnatal changes in the localization of prostaglandin D synthetase from neurons to oligodendrocytes in the rat brain, J. Biol. Chem., 262 (1987) 15132-15136. 28 Van der Ouderaa, F.J., Buytenhek, M., Nugteren, D.H. and Van Dorp, D.A., Purification and characterization of prosta-
29
30
31
32
33 34
35
36 37
glandin endoperoxide synthetase from sheep vesicular glands, Biochim. Biophys. Acta, 487 (1977) 315-331. Van der Ouderaa, EJ., Buytenhek, M., Slikkerveer, EJ. and Van Dorp, D.A., On the haemoprotein character of prostaglandin endoperoxide synthase, Biochim. Biophys. Acta, 572 (1979) 29-42. Watanabe, Y., Yamashita, A., Tokumoto, H. and Hayaishi, O., Localization of prostaglandin D2 binding protein and NADPlinked 15-hydroxyprostaglandin D2 dehydrogenase in the purkinje cells of miniature pig cerebellum, Proc. Natl. Acad. Sci. U.S.A., 80 (1983) 4542-4545. Watanabe, Y., Tokumoto, H., Yamashita, A., Narumiya, S., Mizuno, N. and Hayaishi, O., Specific bindings of prostaglandin D 2, E2, and F2,1 in postmortem human brain, Brain Research, 342 (1985) 110-116. Watanabe, Y., Mori, K., Imamura, K., Takagi, S.E and Hayaishi, O., Modulation by prostaglandin D2 of mitral cell responses to odor stimulation in rabbit olfactory bulb, Brain Research, 378 (1986) 216-222. Watanabe, Yu., Watanabe, Y. and Hayaishi, O., Quantitative autoradiographic localization of prostaglandin E 2 binding sites in monkey diencephalon, J. Neurosci., 8 (1988) 2003-2010. Watanabe, Y., Watanabe, Yu., Hamada, K., Bommelaer-Bayet, M.-C., Dray, F., Kaneko, T., Yumoto, M. and Hayaishi, O., Distinct localization of prostaglandin D2, E 2 and F2,~ binding sites in monkey brain, Brain Research, 478 (1989) 143-148. Yamashita, A., Watanabe, Y. and Hayaishi, O., Autoradiographic localization of a binding protein(s) specific for prostaglandin D2 in rat brain, Proc. Natl. Acad. Sci. U.S.A., 80 (1983) 6114-6118. Yokoyama, C., Takai, T. and Tanabe, T., Primary structure of sheep prostaglandin endoperoxidase synthase deduced from cDNA sequence, FEBS Lett., 231 (1988) 347-351. Yoshimoto, T., Magata, K., Ehara, H., Mizuno, K. and Yamamoto, S., Regional distribution of prostaglandin endoperoxide synthase studied by enzyme-linked immunoassay using monoclonal antibodies, Biochim. Biophys. Acta, 877 (1986) 141-150.
15
BRES 16372
Localization of prostaglandin endoperoxide synthase in neurons and glia in monkey brain Shogo Tsubokura 1'2, Yasuyoshi Watanabe ~, Hiroaki Ehara 4, Kazuyuki Imamura 1, Osamu Sugimoto 2, Hiroyuki Kagamiyama 3, Shozo Yamamoto 4 and Osamu Hayaishi 1 1Department of Neuroscience, Osaka Bioscience Institute, Osaka (Japan), Departments of 2obstetrics and Gynecology and 3 Medical Chemistry, Osaka Medical College, Osaka (Japan) and 4Departmentof Biochemistry, Tokushima University School of Medicine, Tokushima (Japan) (Accepted 18 September 1990) Key words: Prostaglandin endoperoxide synthase; Monoclonal antibody; Immunohistochemisty; Monkey; Brain; Neuron; Glia
The localization of prostaglandin endoperoxide synthase in monkey brain was investigated by the immunoperoxidase method using the monoclonal antibody (PES-7) raised against the enzyme purified from bovine seminal vesicle. The frozen sections with 30-gm thickness were employed after the brain was fixed with perfusion of 2% paraformaldehyde in phosphate-buffered saline. The immunoreactivity was most intense in the neurons of cerebral cortex and hippocampus, and was moderate in the neurons of caudate nucleus, putamen, globus pallidus and amygdala, while it was relatively weak in gliai cells in the whole brain regions including the white matter. The majority of neurons showed the immunoreactivity in the somata and proximal dendrites, but exceptionally in the pyramidal cells of the hippocampus, positive staining was also observed in the apical dendrites. In the cerebellum, the immunoreactivity in both neurons and glia was rather faint as compared with that in other regions. Positive staining was not significantly observed in the vasculatures and arachnoid membranes. These findings indicate that most of neuronal and glial cells in monkey brain contain the enzyme of the rate-limiting and initial step of the biosynthesis of prostaglandins which regulate a variety of neural functions.
INTRODUCTION The initial and c o m m o n step in the biosynthesis of prostaglandins (PGs) is the fatty acid cyclo-oxygenase reaction (arachidonic acid - , PGG2) followed by the PG hydroperoxidase reaction ( P G G 2 - , PGH2) , both of which are catalyzed by the same enzyme referred to as prostag!andin endoperoxide synthase (PES: EC 1.14.99.1; also known as cyclo-oxygenase) 12. Biosyntheses of PGD2, PGE2, PGF2a , PGI2, and Thromboxane A 2 (TXA2) require this unstable PG endoperoxide, P G H 2, as a c o m m o n substrate for each PG and TX synthase. PES was purified from microsomes of bovine 12 and ovine 4'28 vesicular gland, and its molecular and catalytic properties have been investigated 2,4,5,1°.12,17, 19,28,29,36
Recently, PGD2, PGE2, and PGF2a have been identified as major prostaglandins formed in the brain of several mammalian species, including rat 14 and human 18. A variety of central actions of PGs have been demonstrated, such as the regulation of sleep-wake 3, body temperature 11'26, luteinizing hormone-releasing hormone secretion 7'2°, and modification of olfactory sensation 32. Furthermore, the binding protein specific for each P G was found in the synaptic membrane of rat 9'23 and human
brain 31. The localization of these P G receptors by using in vitro autoradiography 3°'33-38 revealed the specific and distinct localization of each P G receptor with close relation to the respective central actions 34. The biosynthetic sites of each P G should also be elucidated for total understanding of the functional sites of PGs. Urade et al. 27 showed the localization of P G D synthase in the rat brain. However, little is known about PES in the central nervous system (CNS) because of a very low content of the enzyme and occasionally of the possible occurrence of certain endogenous enzyme inhibitor(s) 37. The localization of PES is of great importance for determining the sites of PG biosynthesis in CNS. Indeed, in some cell types, PES rather than each PG synthase has been shown to be the rate-limiting step in each PG production 13. We therefore examined the localization of PES in the m o n k e y brain by indirect immunoperoxidase staining using a monoclonal antibody. This is the first report of the existence of PES in the neurons.
MATERIALS AND METHODS Materials Monoclonal antibody against PES (PES-7) was prepared against
Correspondence: Y. Watanabe, Osaka Bioscience Institute, 6-2-4 Furuedai, Suita-shi, Osaka 565, Japan. 0006-8993/91/$03.50 © 1991 Elsevier Science Publishers B.V. (Biomedical Division)
16 the purified PES from bovine vesicular gland as described previously 37. Normal mouse IgG was purchased from Cappel Lab. Inc.; Vectastain ABC kit, from Vector Lab. Inc.; [114C]Arachidonic acid (59.6 mCi/mmol), from Amersham; Protein A-Sepharose CL-4B, from Pharmacia; Silver Stain Kit and 4chloro-l-naphthol, from Wako (Tokyo); precoated silica gel 60F 254 glass plates for TLC, from E. Merck; molecular weight standard mixture, Protein Assay Kit, and nitrocellulose membrane filters, from Bio-Rad; and bovine serum albumin (Fraction V), from Sigma.
Preparation of solubilized enzyme Solubilized enzyme was prepared as described previously 12. The monkey was anesthetized with ketamine hydrochloride followed by pentobarbital and was perfused with 10 mM sodium phosphate buffered saline (pH 7.4) (PBS) via the left ventricle of the heart. The monkey brain was rapidly removed and homogenized in 5 vols. of 20 mM potassium phosphate buffer (pH 7.3) containing 0.1 mM phenylmethylsulfonyl fluoride using a Potter-Elvehjem glass homogenizer with a Teflon pestle. The homogenate was centrifuged at 10,000 g for 10 rain and the resulting supernatant was further centrifuged at 164,000 g for 60 min. The pellet was dispersed in a half of the original volume of 20 mM potassium phosphate buffer (pH 7.3) containing 1% Tween 20. After standing at 4 °C for 30 min, the mixture was centrifuged at 164,000 g for 60 min, and the resulting supernatant was used as the solubilized enzyme.
Immunoaffinity column analysis The solubilized enzyme (4.8 mg protein/ml, 5 ml) was incubated with PES-737 (660/ag protein/ml, 1 ml) at 4 °C for 2 h, then the mixture was applied to Protein A-Sepharose CL-4B column (bed size = 7 mm i.d. x 4 cm). After the column was washed with 3 bed volumes of 0.1 M borate buffer (pH 8.2) containing 0.14 M sodium chloride, PES-7-PES complex was eluted with 5 mi of 0.1 M acetate containing 0.14 M sodium chloride. The pH of the eluting solution was immediately adjusted to approximately 7.0 using 1.0 M Tris-base. Enzyme activity in both passed through and eluted fractions was assayed according to the method described by Miyamoto et al. 12. Protein was determined by the method of Bradford I with bovine serum albumin (BSA) as standard.
lmmunoblotting analysis Sodium dodecyi sulfate-polyacrylamide gel electrophoresis (SDSPAGE) was carried out in 4-20% gradient acrylamide by the method of Laemmli s. The solubilized enzyme from monkey brain was subjected to SDS-PAGE at 40 mA for 2 h and then the proteins were transferred electrophoretically at 2 mA/cm 2 for 2 h to a nitrocellulose membrane by Trans Blot (Bio-Rad) according to the method of Towbin et al. 2s, in the presence of 192 mM glycine, 25 mM Tris-base, and 20% methanol. The nitrocellulose membrane was washed with PBS containing 0.05% Tween 20. Then, the membrane was blocked overnight at 4 °C with 3% (wt/vol) BSA and 0.5% (wt/vol) non-fat dry milk21 dissolved in PBS containing 0.05% Tween 20 to prevent non-specific binding, and was incubated at room temperature with PES-7 (20/ag/ml in PBS containing 0.05% Tween 20) in the presence of 1% BSA for 2 h. Thereafter the membrane was washed as above, and then incubated with the horse biotinylated anti-mouse IgG antibody in PBS at room temperature for 30 rain. The membrane was washed and treated with Vectastatin ABC kit (Vector Lab.). After being washed, the nitrocellulose membrane was subjected to the peroxidase reaction using 4chloro-l-naphthol as substrate for 3 rain. Tissue preparation
Macaca fuscata fuscata (male, 6.0-8.0 kg) was anesthetized with ketamine hydrochloride (Ketarai) and pentobarbitai and perfused via the left heart ventricle with cold PBS (5 1), then 2% paraformaldehyde (PFA) in PBS (3 I), followed by PBS containing 5% sucrose (3 !). The brain was rapidly removed, cut into 5-mm thick coronal blocks using microtome blades (Feather $35 type) and
immersed in 30% sucrose in PBS. After the specimens were sunk in the solution at 4 °C, they were frozen on dry-ice. Frozen sections (30/zm thickness) were cut in a microtome (YAMATO, TU-213) consecutively and thaw-mounted onto ovalbumin-coated glass slides (76 x 52 mm).
lmmunohistochemical staining The mounted sections were air-dried at room temperature for 90 rain. The dried sections were then fixed in acetone at 4 °C for 10 min and washed 3 times with PBS. The sections were covered with 20% normal horse serum, diluted with PBS containing 0.3% Triton X-100 (TPBS), as a blocking agent and kept at room temperature for 60 min. The sections were then incubated with PES-7 (20/zg/ml in TPBS containing 1% BSA) at 4 °C overnight, followed by 3 washings with TPBS. Then, the sections were incubated with biotinylated horse anti-mouse IgG diluted with TPBS followed by incubation with the avidine-biotin-peroxidase complex at room temperature for 30 rain. After each step, the sections were washed 3 times with TPBS. After the final wash, the sections were bufferized with 50 mM Tris-HCl (pH 7.2), stained by incubation with diaminobenzidine tetrahydrochloride (0.5 mg/ml) in 0.02% H20 2 for 4 min. The sections were then washed with water, dehydrated and mounted. The samples were examined and photographed using a VANOX-S microscope equipped with a camera (Olympus Co., Ltd.). Non-immune mouse IgG (Cappel) rechromatographed on a Protein A-Sepharose CL-4B column was used as a negative control. For assessment of the specificity of immunoreactivity for PES, PES-7 solution preabsorbed with the purified PES from sheep vesicular gland was used for staining as another negative control. Some of the sections were further fixed in 2% PFA at 4 °C overnight, then stained by 0.1% thionine (Nissl staining). Immunoreactivity with PES-7 was scored for evaluation of PES
A
B
200 K - ~ " ~
116K 97K
66K
43K--~llb
Fig. 1. Immunoblotting with monoclonal antibody against PES after SDS-polyacrylamide gel electrophoresis of the solubilized enzyme fraction from the monkey brain microsome. Anode and origin are at the top of the figure. The arrowhead indicates the position of the tracking dye. Lane A, silver staining of the monkey brain solubilized fraction; lane B, immunobiotting incubated with PES-7 as described under Materials and Methods.
17
Fig. 2. PES-immunoreactivity in the monkey temporal cortex. A and D: immunoperoxidase staining with the monoclonal antibody against PES in the coronal section of the temporal cortex. The section (30/zm thickness) was incubated with 20 ~ug/ml PES-7 as described under Materials and Methods. B and E: control staining in the consecutive section with non-immune mouse IgG. C and F: Nissl staining with 0.1% thionine in the consecutive sections~ Scale bar (A,B,C,) = 125/zm. D,E, and F higher magnification in layer III. Scale bar = 50 jum.
18
Fig. 3. PES-immunoreactivity in the monkey hippocampus. A, B, D and E: immunoperoxidase staining with the monoclonal antibody against PES in the coronal section of the hippocampus. C: control staining with non-immune mouse IgG in place of PES-7 in the consecutive section. Bar = 500/zm (A). Bar = 100/zm (B.C). D: higher magnification of the view in CA3 region. Bar ~ 50 #m (D). Bar = 12.5/~m (E).
19 content by (a) the percentage of positive stained cells and (b) the staining intensity (+ to ++ +) visually under the light microscopic examination. The intensity of specific staining was expressed as weak but detectable above control (+), intense (+ +), and markedly intense (+ + +). RESULTS
Specificity of the monoclonal antibody Specificity of the monoclonal antibody (PES-7) was evaluated by SDS-PAGE and immunoblotting. As shown in Fig. 1, PES-7 gave a single line of molecular weight of 75 kDa with the solubilized fraction of monkey brain microsomes. No other proteins cross-reacted with this antibody. When the homogenate of monkey brain prepared as described under Materials and Methods was transblotted after SDS-PAGE, no immunoreactive bands were observed because of the relatively low and undetectable amounts of PES in the homogenate. Furthermore, when the solubilized enzyme fraction incubated with PES-7 was applied to a Protein A-
Sepharose CL-4B column, about 62% of the original PES activity was recovered in the eluted fraction as an immune complex. No PES activity was detected in pass through or washing fraction. Then the eluted fraction was subjected to SDS-PAGE; only PES and PES-7 could be detected by Coomassie brilliant blue staining (data not shown). These results indicate that PES-7, the monoclohal antibody employed in this immunohistochemical study, is highly specific to PES in the monkey brain.
Immunohistochemical brain
localization of PES in monkey
When PES-7 was utilized for immunoperoxidase staining to reveal the cellular localization of PES in the monkey brain, intense staining was observed in a number of neurons and glias. In contrast, no positive staining was detected with non-immune IgG. The positive immunoreactivity in the hippocampus and cerebral cortices was titrated by using purified PES from sheep vesicular gland and 124 /zg/ml of the purified PES could completely
Fig. 4. PES-immunoreactivity in the monkey cerebellum. The contrast of these photographs is graded up compared with the photos representing the immunoreactivity in other brain regions (Figs. 2 and 3). A and C: immunoperoxidase staining with the monoclonal antibody against PES. B and D: control staining with non-immune IgG in the consecutive section. Bar = 200 ~m (A,B); bar = 50 ~m (C,D).
21 diminish the positive staining (molar ratio of PES/PES-7 = 12). The intensity of the PES-immunoreactivity was somewhat different among various brain regions. Fig. 2A shows a typical pattern of PES-immunoreactivity in the temporal cortex. Intensely positive staining was found throughout a n u m e r o u s n u m b e r of nerve cells, while glial cells both in the white matter and gray matter showed
TABLE I
Percentage of positive cell# and relative intensity of P ES-immunoreactivity in monkey brain regions The population of immunoreactive cells from more than 300 cells, good-shaped (avoided from the counting error because of the irregular sectioning), in each structure were counted under light microscope, and relative intensity of immunoreactivity was graded from + (weak) to + + + (intense).
weaker and fewer staining. Especially, the nerve cells in layer II showed most intense immunoreactivity. In layers I I - V I , the immunoreactive neurons constituted more than 95% of the n e u r o n s when calculated in comparison with an adjacent Nissl-stained section (Fig. 2C and F), but less than 25% of n e u r o n s were immunoreactive in layer I. A t higher magnification, only n e u r o n a l somata and proximal dendrites showed the immunoreactivity (Fig. 2D). Distal dendrites showed slightly positive, but the cell nuclei did not show any positive staining. In contrast, when n o n - i m m u n e IgG was used in place of
PES-7 in the consecutive sections, the specific staining
Frontal cortex
II III IV V VI w* Temporal cortex I II III IV
staining between PES-7 and control IgG was observed in the vasculatures, especially microvessels and arachnoid m e m b r a n e s . No immunoreactivity was seen in the terminal structure. This pattern and density of the immunoreactivity were similar to those in other cortical regions, such as frontal, parietal, and occipital cortices. The immunohistochemical staining in the hippocampal
formation is shown in Fig. 3. In this region, the granule cells of dentate gyrus and pyramidal cells of A m m o n ' s
horn were strongly immunoreactive (Fig. 3A), and some of the glial cells were less immunoreactive. All of the granule cell bodies showed intense immunoreactivity (Fig. 3B). The pyramidal cells at higher magnification showed the PES-immunoreactivity not only in the neuronal somata and proximal dendrites, but also in apical dendrites of a similar intensity (Fig. 3D and E). The positive staining in the apical dendrites was only seen in C A 2 and CA3 regions. Fig. 4 shows the pattern of staining in the m o n k e y cerebellum. The immunoreactivity (Fig. 4A and C) was close to the control level (Fig. 4B and D) with noni m m u n e mouse IgG in place of PES-7 in both neurons and glial cells. Moreover, the elongation of the time of D A B reaction could not extend the difference of the densities between PES-7 and control stainings. In the molecular layer, the stellate cells and a small n u m b e r of glia showed positive immunoreactivity as compared with
Parietal cortex
Occipital cortex
Glial cell
%
Intensity
%
23 100 95 100 97 100
+ +++ ++ ++ ++ ++
-
VI
13 100 100 100 100 100
w*
-
v
was not observed in either nerve cells or glial cells (Fig.
2B and E). No significant difference in the intensity of
I
Nerve cell
I II III IV v vI w* I II III
IV V VI W* Globuspallidus Putamen Caudate Amygdala Thalamus Hypothalamus Hippocampus granular cell layer CA4 CA3 CA2 CA1 dentategyrus Cerebellum molecular layer Purkinje cell layer granularlayer w* Pons Medulla oblongata Spinalcord
20 100 100 100 100 100 -
-
+ +++ ++ ++ ++ ++ -
+ +++ ++ ++ ++ ++ -
Intensity
22 19 13 11 10 11
+ + + + + +
54
+
22 7 5 22 14 7
++ + + + + +
48
+
46 18 8 11 10 19
+ + + + + +
41
+
14 100 100 99 100 99
+ +++ ++ ++ ++ ++
20 65 92 78 75 100
++ ++ ++ ++ ++ +++
11 19 28 17 30 15
+ + + + + +
92 87 84 80 100
+++ +++ +++ +++ +++
21 30 17 29 n.d.
+ + + +
40 63 100
+ + +
44 + n.d. n.d.
-
41 74 63
-
18
+
8
+
5 12 10 10
+ + + +
50
+
-
67
+
+ + +
13 33 76
+ + +
* White matter; n.d.: not detected.
Fig. 5. PES-immunoreactivity in the C A 2 region of hippocampus (A), a part of caudate nucleus (B), layers III and IV of striate cortex (C), N. basalis Meynerti (D, shown by an arrow), the paraventricular, lateral superior central, and dorsomedial n. (ME)) of thalamus (E), and substantia nigra pars compacta and pars diffusa (F). × 5 0 (A, B and C), ×25 (D and E), and × 1 0 (F). Fig. 6. PES-immunoreactivity in the glial cells. A: comparison of stainability between neurons and glia in the C A 3 region of hippocampus (oriens layer and a part of alveus). × 150. B: glial cells in the corpus callosum. ×300.
22 the control staining. The Purkinje cells and Bergmann's glia were not positive. In the granule cell layer, almost all granule cells showed very weak immunoreactivity. Fig. 5 shows the PES-immunoreactivity in various brain regions with special reference to neurotransmitter systems. The immunoreactivity was abundant in various neurotransmitter systems. In the caudate nucleus (Fig. 5B), the intense staining was observed in the cytosol of neurons but not in the distal dendrites in contrast to the immunoreactivity in the pyramidal cells in CA2 and CA3 regions of the hippocampus (Fig. 5A). The immunoreactivity was also positive in the glutamatergic (in the striate cortex, Fig. 5C), cholinergic (N. basalis Meynerti, Fig. 5D), serotonergic (dorsomediai N. of thalamus, Fig. 5E), and dopaminergic (substantia nigra pars compacta and pars diffusa, Fig. 5F) neurons. However, the density of the immunoreactivity in the neurons in the same brain regions was different among the nuclei; for example, the ventromedial nucleus of hypothalamus showed less immunoreactivity than the surrounding hypothalamic nuclei (data not shown). Table I shows the summary of the distribution of PES immunoreactivity. In the whole brain, the hippocampal formation, cerebral cortex and caudate nucleus showed intense immunoreactivity, whereas midbrain, cerebellum, spinal cord, and white matter were weakly stained. The percentage of the immunopositive cells compared with the adjacent Nissl-stained sections was high in the case of neurons but was relatively low in the case of glias. Relatively low intensity of the immunoreactivity was seen in the glias (Fig. 6A), but still it was positive and clear from the surrounding tissues. Oligodendrocytes (Fig. 6B) seemed to contain greater amounts of PES-immunoreactivity than astrocytes, judged from the cell contour and the regions of their presence. We employed 5 monkey brains in this study. The results showed little heterogeneity in the pattern and density of localization of PES immunoreactivity among the brain specimens. DISCUSSION In the present study, we demonstrated the immunohistochemical localization of PES in the monkey CNS. The specificity of the immunostaining was assessed by using the control with non-immune mouse IgG and by blocking the immunoreaction with the purified PES. The protein immunoblotted after SDS-PAGE of monkey cerebral cortex supernatant showed a single band of the molecular weight of around 75 kDa which is essentially similar to the molecular weight of PES purified from bovine or sheep vesicular gland (Fig. 1). In addition, the roughly estimated regional distribution of PES-immuno-
reactivity was close to that in the bovine brain reported by Yoshimoto et al. 3v by using enzyme-linked immunoassay. Since the monoclonal antibody (PES-7) employed in this study does not cross-react with rat brain PES and rabbit kidney PES, we initially used ox brain for immunostaining. However, the conditions of ox brain obtained from a local slaughterhouse were not satisfactory for constant staining and brain structure. The monkey brain sections were available because of the autoradiographic studies on PG receptors in our laboratory 33'34 and were positively stained by the monoclonal antibody. Essentially similar results were obtained using ox brain sections, especially in the hippocampus, caudate nucleus and cerebral cortices. This is the first direct evidence demonstrating the existence of PES in the neurons as well as glial cells. Since the content of PES in CNS and the sensitivity of measurement of PES using [14C]arachidonic acid are low, no one has so far succeeded in detecting PES activity after fractionation of cells by collagenase and pronase treatment. Concerning the immunostaining of PES in CNS, only one report by Smith et al. 24 is available. Their pioneering work using polyclonal antibody, however, focused on the cerebellar cortex of pig, guinea pig, rat, mouse, cow, rabbit and sheep brains and failed to detect an adequate PES-immunoreactivity in neurons. They reported the positive immunoreactivity only in the vasculatures and Bergmann glial cells surrounding Purkinje cells in the porcine, ovine and bovine cerebellar cortices. Our results of immunostaining also showed relatively low density (or close to the control level) in the cerebellar neurons and glias (Fig. 4). In addition, another monoclonal antibody for PES (PES-5) employed in our preliminary work did not significantly bind any structure of ox and monkey brains after no fixation or 2-4% paraformaldehyde fixation. Furthermore, when we tested a variety of fixatives using ox brain and PES-7 such as no fixation, Bouin, Zamboni, various concentrations of paraformaldehyde and glutaraldehyde, intensely positive staining was obtained only by using 1-6% paraformaldehyde. Furthermore, only when using the present fixative, the positive staining in the apical dendrites was observed in the hippocampus. Thus the discrepancy between their results and ours could be interpreted by the species difference, the difference of the antibody employed, and especially, the difference of fixation condition, in addition to the fact that they tried to stain only cerebellar sections. The lack of positive PES-immunoreactivity in vasculatures in the present study might indicate the sensitivity of the staining, namely, below the threshold level of positive staining by this method. However, another
23 possibility lies in the use of perfused fixation of the brain tissues, since the vasculature is too much fixed by such a type of fixation as compared with the extent of fixation on the substantial tissues. In fact, in our experiments, too much fixation caused the low stainability for PES even in the neurons and glia in the case of 6% paraformaldehyde and/or 1% glutaraldehyde fixation. Ubiquitous but heterogenous distribution of PESimmunoreactivity in the neurons and glia was obtained in this study. Although the cell volume should be carefully estimated, the PES-immunoreactivity seemed to be more dense in neurons than that in glia. Also, the immunoreactivity was not limited either in the specific neurons of some brain structures or in the specific neurotransmitter systems. This finding is opposite to the limited and distinct localization of P G D 2 and P G E 2 receptors in the m o n k e y brain 33'34. Therefore, the localization of each PG synthase following this PES reaction should be the key for determining the responsibility of each PG for the specific neuronal functions. Urade et al. 27 showed the immunoreactivity of P G D synthase in the rat brain with dramatic developmental changes; almost all neurons in the 1-week-old brain and a limited number of neurons in the adult brain. The major site of the immunoreactivity changed from neurons to oligodendrocytes. This type of developmental change is also interesting, since the primary cultured astroglia produce much larger amounts of cyclo-oxygenase metabolites than the primary cultured neurons 6'22. However, a relatively low PES-immunore-
activity was seen in the astrocytes in whole brain regions in the present study. Such a type of developmental study and accurate identification of type 1 and type 2 astrocytes by using antiglial fibrillary acidic protein antibody and other markers is currently under investigation in our laboratory. High density of PES-immunoreactivity in the pyramidal cells in the hippocampus may be supported by the experimental results by Ogawa et al. 15'16. They showed the PGF2a-immunoreactivity in the pyramidal cells of hippocampus after reoxygenation in anoxic rats and after experimental subarachnoid hemorrhage in rats. At that time, blood vessels were positively stained for PGF2a but other cells such as glial cells were all negative. P G E 2immunoreactivity was also demonstrated in the pyramidal cells of rat hippocampus and was remarkably enhanced by administration of convulsive drugs (Fujimoto et al., unpublished results). If one of the rate-limiting steps of prostaglandin synthesis is actually the PES step in the brain, the regulatory mechanisms of PES under such pathological conditions are also the target of future research.
REFERENCES
synaptosomes, Brain Research, 236 (1982) 227-233. 10 Merlie, J.P., Fagan, D., Mudd, J. and Needleman, P., Isolation and characterization of the complementary DNA for sheep seminal vesicle prostaglandin endoperoxidase synthase (cyclooxygenase), J. Biol. Chem., 263 (1988) 3550-3553. 11 Milton, A.S. and Wendlandt, S., A possible role for prostaglandin E~ as a modulator for temperature regulation in the central nervous system of the cat, J. Physiol., 207 (1970) 76-77. 12 Miyamoto, T., Ogino, N., Yamamoto, S. and Hayaishi, O., Purification of prostaglandin endoperoxide synthetase from bovine vesicular gland microsomes, J. Biol. Chem., 251 (1976) 2629-2636. 13 Murota, S., Chang, W.C., Koshihara, Y. and Morita, I., Importance of cyclo-oxygenase induction in the biosynthesis of prostacyclin, Adv. Prostaglandin Thromboxane Leukotriene Res., 11 (1983) 99-104. 14 Narumiya, S., Ogorochi, T., Nakao, K. and Hayaishi, O., Prostaglandin D 2 in rat brain, spinal cord and pituitary: basal level and regional distribution, Life Sci., 31 (1982) 2093-2103. 15 Ogawa, H., Kassell, N.E, Sasaki, T., Hongo, K., Tsukahara, T., Hudson, S.B., Asban, G.I., Tuan, H.L. and Turner, J.C., ImmunohistochemicaUy demonstrated increase of prostaglandin F2a in neurons after reoxygenation in anoxic rats, Prostaglandins, 36 (1988) 891-900. 16 Ogawa, H., Kassell, N.F., Sasaki, T., Nakagomi, T., Hongo, K and Tsukahara, T., Rapid increase of prostaglandin F~ in neurons after subarachnoid hemorrhage in rats: an immunohistochemical study, Acta Neuropathol., 78 (1989) 621-628. 17 Ogino, N., Ohki, S., Yamamoto, S. and Hayaishi, O., Prostaglandin endoperoxide synthetase from bovine vesicular gland
1 Bradford, M.M., A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding, Anal. Biochem., 72 (1976) 248-254. 2 DeWitt, D.L. and Smith, W.L., Primary structure of prostaglandin G/H synthase from sheep vesicular gland determinated from the complementary DNA sequence, Proc. Natl. Acad. Sci. U.S.A., 85 (1988) 1412-1416. 3 Hayaishi, O., Sleep-wake regulation by prostaglandins D 2 and E2, J. Biol. Chem., 263 (1988) 14593-14596. 4 Hemler, M., Lands, W.E.M. and Smith, W.L., Purification of the cyclooxygenase that forms prostaglandins: demonstration of two forms of iron in the holoenzyme, J. Biol. Chem., 251 (1976) 5575-5579. 5 Hemler, M.E. and Lands, W.E.M., Evidence for a peroxideinitiated free radical mechanism of prostaglandin biosynthesis, J. Biol. Chem., 255 (1980) 6253-6261. 6 Keller, M., Jackisch, R., Seregi, A. and Hertting, G., Comparison of prostanoid forming capacity of neuronal and astroglial cells in primary cultures, Neurochem. Int., 7 (1985) 655-665. 7 Kinoshita, E, Nakai, Y., Katakami, H., Imura, H., Shimizu, T. and Hayaishi, O., Suppressive effect of prostaglandin (PG) D z on pulsatile luteinizing hormone release in conscious castrated rats, Endocrinology, 110 (1982) 2207-2209. 8 Laemmli, U.K., Cleavage of structural proteins during the assembly of the head of bacteriophage T4, Nature, 227 (1970) 680-685. 9 Malet, C., Scherrer, H., Saavedra, J.M. and Dray, E, Specific binding of [3H]prostaglandin E 2 to rat brain membranes and
Acknowledgements. We thank Drs. K. Mori and S. Oka of our institute for useful technical advice and discussions, and Ms. M. Tanemura for assistance. This work was supported in part by a Grant-in-Aid (60580158) and a Grant-in-Aid for Special Project Research of Plasticity of Neural Circuits from the Japanese Ministry of Education, Science and Culture, and by grants from the Japanese Foundation of Metabolism and Diseases and Sankyo Co., Ltd.
24 microsomes, J. Biol. Chem., 253 (1978) 5061-5068. 18 Ogorochi, T., Narumiya, S., Mizuno, N., Yamashita, K., Miyazaki, H. and Hayaishi, O., Regional distribution of prostaglandins D2, E 2. and F2~ and related enzymes in postmortem human brain, J. Neurochem., 43 (1984) 71-82. 19 Ohki, S., Ogino, N., Yamamoto, S. and Hayaishi, O., Prostaglandin hydroperoxidase, an integral part of prostaglandin endorperoxide synthesis from bovine vesicular gland microsomes, J. Biol. Chem., 254 (1979) 829-836. 20 Ojeda, S.R. and Campbell, W.B., An increase in hypothalamic capacity to synthesize prostaglandin E 2 precedes the first preovulatory surge of gonadotropins, Endocrinology, 111 (1982) 1031-1037. 21 Plapp, F.V., Rachel, J.M. and Sinor, L.T., Dipsticks for determining ABO blood groups, Lancet, 28 (1986) 1465-1466. 22 Seregi, A., Keller, M. and Hertting, G,, Are cerebral prostanoids of astroglial origin? Studies on the prostanoid forming system in developing rat brain and primary cultures of rat astrocytes, Brain Research, 404 (1987) 113-120. 23 Shimizu, T., Yamashita, A. and Hayaishi, O., Specific binding of prostaglandin D 2 to rat brain synaptic membrane, J. Biol. Chem., 257 (1982) 13570-13575. 24 Smith, W.L., Gutekunst, D.I. and Lyons Jr., R.H., Immunocytochemical localization of the prostaglandin-forming cyclooxygenase in the cerebeUar cortex, Prostaglandins, 19 (1980) 61-69. 25 Towbin, H., Staehelin, T. and Gordon, J., Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications, Proc. Natl. Acad. Sci. U.S.A., 76 (1979) 4350-4354. 26 Ueno, R., Narumiya, S., Ogorochi, T., Nakayama, T., Ishikawa, Y. and Hayaishi, O., Role of prostaglandin D2 in the hypothermia of rats caused by bacterial lipopolysaccharide, Proc. Natl. Acad. Sci. U.S.A., 79 (1982) 6093-6097. 27 Urade, Y., Fujimoto, N., Kaneko, T., Konishi, A., Mizuno, N. and Hayaishi, O., Postnatal changes in the localization of prostaglandin D synthetase from neurons to oligodendrocytes in the rat brain, J. Biol. Chem., 262 (1987) 15132-15136. 28 Van der Ouderaa, F.J., Buytenhek, M., Nugteren, D.H. and Van Dorp, D.A., Purification and characterization of prosta-
29
30
31
32
33 34
35
36 37
glandin endoperoxide synthetase from sheep vesicular glands, Biochim. Biophys. Acta, 487 (1977) 315-331. Van der Ouderaa, EJ., Buytenhek, M., Slikkerveer, EJ. and Van Dorp, D.A., On the haemoprotein character of prostaglandin endoperoxide synthase, Biochim. Biophys. Acta, 572 (1979) 29-42. Watanabe, Y., Yamashita, A., Tokumoto, H. and Hayaishi, O., Localization of prostaglandin D2 binding protein and NADPlinked 15-hydroxyprostaglandin D2 dehydrogenase in the purkinje cells of miniature pig cerebellum, Proc. Natl. Acad. Sci. U.S.A., 80 (1983) 4542-4545. Watanabe, Y., Tokumoto, H., Yamashita, A., Narumiya, S., Mizuno, N. and Hayaishi, O., Specific bindings of prostaglandin D 2, E2, and F2,1 in postmortem human brain, Brain Research, 342 (1985) 110-116. Watanabe, Y., Mori, K., Imamura, K., Takagi, S.E and Hayaishi, O., Modulation by prostaglandin D2 of mitral cell responses to odor stimulation in rabbit olfactory bulb, Brain Research, 378 (1986) 216-222. Watanabe, Yu., Watanabe, Y. and Hayaishi, O., Quantitative autoradiographic localization of prostaglandin E 2 binding sites in monkey diencephalon, J. Neurosci., 8 (1988) 2003-2010. Watanabe, Y., Watanabe, Yu., Hamada, K., Bommelaer-Bayet, M.-C., Dray, F., Kaneko, T., Yumoto, M. and Hayaishi, O., Distinct localization of prostaglandin D2, E 2 and F2,~ binding sites in monkey brain, Brain Research, 478 (1989) 143-148. Yamashita, A., Watanabe, Y. and Hayaishi, O., Autoradiographic localization of a binding protein(s) specific for prostaglandin D2 in rat brain, Proc. Natl. Acad. Sci. U.S.A., 80 (1983) 6114-6118. Yokoyama, C., Takai, T. and Tanabe, T., Primary structure of sheep prostaglandin endoperoxidase synthase deduced from cDNA sequence, FEBS Lett., 231 (1988) 347-351. Yoshimoto, T., Magata, K., Ehara, H., Mizuno, K. and Yamamoto, S., Regional distribution of prostaglandin endoperoxide synthase studied by enzyme-linked immunoassay using monoclonal antibodies, Biochim. Biophys. Acta, 877 (1986) 141-150.
