A
polyclonal antibody to the rat oestrogen receptor expressed in Escherichia coli: characterization and application to immunohistochemistry H.
Okamura, K. Yamamoto, S. Hayashi, A. Kuroiwa and M. Muramatsu
Department of Anatomy and Embryology, Tokyo Metropolitan Institute for Neuroscience, 2-6 Musashidai, Fuchu City, Tokyo 183, Japan *Department of Biology, School of Education, Waseda University, Shinjuku-ku, Tokyo 169-50, Japan tDepartment of Cell Biology, Research Institute for Tuberculosis and Cancer, Tohoku University, Sendai, Miyagi 980, Japan |Department of Biochemistry, Faculty of Medicine, University of Tokyo, Bunkyo-ku, Tokyo 113, Japan (Requests for offprints should be addressed to S. Hayashi) received 28 February 1992 ABSTRACT
A rat oestrogen receptor-\g=b\-galactosidase fusion protein was expressed using a pEX2/rat oestrogen receptor cDNA construct. Scatchard analysis of [3H]oestradiol-17\g=b\binding to the cell lysate revealed that the fusion protein had functional binding sites specific for oestradiol with a dissociation constant of 1\m=.\49nmol/l. The relative molecular weight (Mr) of the fusion protein was determined as 180 000 by immunoblot analysis of the cell lysate employing a monoclonal antibody to the human oestrogen receptor. The protein was isolated by means of SDS-PAGE and subsequent electroblotting. By immunization with the purified materials on nitrocellulose membrane, a polyclonal antibody to the rat oestrogen receptor was raised in a rabbit. Binding of [3H]oestradiol to the oestrogen receptor from the rat uterus was inhibited by the antibody in a dose\x=req-\ dependent manner. The antibody was also able to recognize the oestrogen receptor occupied by
[3H]oestradiol. Thus, the antibody could react with both forms of the receptor molecule, either occupied or unoccupied by the hormone. In immunoblot analysis of the cytosol fraction of the rat uterus, a single band of Mr 67 000, the size of the oestrogen receptor, was detected by the antibody. Moreover, when the antibody was applied to immunohistochemical examination of paraffin-embedded pituitary and brain sections of the rat, immunostaining was observed in cells of the anterior pituitary and in neurones in specific regions of the brain. The immunoreactivity was restricted exclusively to cell nuclei in both tissues. These results demonstrate that the polyclonal antibody obtained in the present study was specific to the oestrogen receptor, and that it would be a powerful tool to detect and analyse the receptors in various target tissues for oestrogen. Journal of Endocrinology (1992) 135, 333\p=n-\341
INTRODUCTION
Classically the oestrogen receptor (ER) has been detected in target tissues by autoradiography using tritium-labelled oestradiol-17ß (OE2) (Pfaff & Keiner, 1973; Sheridan, Sar & Stumpf, 1974). Although this technique satisfies all of the criteria of receptor binding (Stumpf, Narbitz & Sar, 1980), it is a disad¬ vantage that the method requires a very long process¬ ing time. On the other hand, antibodies to the ER have been shown to offer direct and accurate localiza¬ tion of the receptors by an immunohistochemical method. Several groups have produced antibodies
Oestrogen is responsible for a variety of reproductive actions which are mediated by a specific intracellular receptor in target tissues. Following binding, the hor¬ mone-receptor complex functions as a nuclear transcription factor that alters the expression of spe¬ cific genes (Yamamoto, 1985). Therefore, in order to understand the mechanism of action of oestrogen, it is of primary importance to determine the function and localization of the receptor in detail.
directed to the receptor protein from human mam¬ mary tumours (Green, Sobel, King & Jensen, 1984) or calf uterus (Moncharmont, Su & Parikh, 1982), or antibodies directed to synthesized peptides which correspond to preselected regions in the amino acid sequence of the molecule (Traish, Kim & Wotiz, 1989 ; Furlow, Ahrens, Mueller & Gorski, 1990). Of these antibodies, the monoclonal antibody to the human ER, which is commercially available, has been widely used to detect the ERs in a variety of animals includ¬ ing man (King & Green, 1984; Walker, Bouzubar, Robertson et al. 1988), mice (Korach, Horigome, Tomooka et al. 1988; Koch, 1990), rabbits (Zaino, Clarke, Feil & Satyaswaroop, 1989) and birds (Balthazart, Gahr & Surlemont, 1989). This rat monoclonal antibody, however, may not be suitable to determine the localization of the ER in rat tissues, since the secondary antibody (anti-rat IgG) binds throughout the tissue to endogenous immunoglobulins and produces widespread background staining (Gee, Nicholson, Jasani et al. 1990). Thus, to study the precise localization of the ER in the rat, antibodies to the ER raised in any species other than the rat are desirable. Recent success in the cloning of a cDNA for the rat ER (Koike, Sakai & Muramatsu, 1987) enabled us to obtain the receptor protein in large amounts by its expression in a culture of Escherichia coli cells. The expressed protein contained a functional ER which bound specifically to OE2. The protein, purified by means of sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) and subsequent electroblotting, was injected into a rabbit and an antiserum was obtained. Using biochemical and immunohisto¬ chemical examinations, we proved that the antiserum contained antibodies directed specifically to the ER.
MATERIALS AND METHODS
Expression The pEX2
of the rat ER in E. coli
(Stanley & Luzio, 1984) was pur¬ Boehringer Mannheim GmBH (Mannheim, Germany) and host strain N4830-1, a PI transductant of N4830 (Gottesman, Adhya & Das,
chased
vector
from
1980), was obtained from Pharmacia Fine Chemicals (Uppsala, Sweden). cDNA for the rat ER (Koike et al. 1987) was modified to introduce an EcoRI site at
the 5' terminus on the sequence for the construction produce the ß-galactosidase-rat ER fusion protein. The fragment encompassing from codon 68, through translational termination codon to some additional 3' non-coding sequence was excised by SacII and EcoRI double digestion. The EcoRI site at the 3' end of the fragment was artificially introduced during cDNA to
et al. 1987). To introduce another EcoRI site at the 5' end of the fragment, oligonucleotides corresponding to the region from 391 to 461 in the cDNA (5-AATTCAACGCCGCCGCCGCCGCCGCG and 5-GCGGCGGCGGCGGCGTTG) were synthesized with a single base substitution G at position 393 of the cDNA sequence to A. This substitution creates the EcoRI site at the 5' end of the oligonucleotide without changing the amino acid sequence of the rat ER portion. The oligonucleotides were annealed to form a double strand, ligated with the SacII-EcoRI fragment and the resulting chimeric fragment was cloned into the EcoRI site of pUC18. The sequence of the exchanged region was then deter¬ mined. The EcoRI fragment was then excised, trans¬ ferred to the EcoRI site of the pEX2 vector and a clone containing the cDNA with the sense orientation for ß-galactosidase was selected. The final construct was introduced into N4830-1 host cells for protein production. The rat ER portion in the fusion protein corresponds to amino acid 61 to carboxyl terminus of native rat ER and contains DNA binding and ligand binding domains. The protein was expressed in the host cells according to the method described by Sambrook, Fritsch & Maniatis (1989). Briefly, the cells were incubated in two x YT broth (bacto-trypton, 16 g/l ;bacto-y east extract, 10 g/l ; NaCl, 5 g/l; pH 7-0) containing 50 pg ampicillin/ml at 30 °C with constant agitation. Absorbance at 650 nm was monitored. When it reached 0-8, the culture was transferred to 42 °C and incubated for a further 2 h with constant agitation. The cell pellet was stored at —80 °C until further processing.
cloning (Koike
Biochemical characterization of the fusion protein The E. coli cells were sonicated and centrifuged at 10 000 g for 15 min. The pellet was washed three times with 10 mmol Tris-HCl/1, pH 7-4, containing 1 mmol EDTA/1,1 mmol dithiothreitol/1 and 10 mmol sodium molybdate/1 (TEDM buffer) and resuspended in the buffer. The suspension (crude fraction) containing 400 pg protein was incubated with increasing concen¬ trations of [3H]OE2 (New England Nuclear, MA, U.S.A.) in a total volume of 0-25 ml with or without a 100-fold excess of unlabelled OE2 for 20 min at 37 °C. The mixture was then cooled on ice and centri¬ fuged at 1000 g for 30 min. After washing the pellets three times with 3 ml TEDM buffer, radiolabelled materials were extracted with 1 ml absolute ethanol. Scintillation fluid (3 ml) was added to each extract and the radioactivity was measured. Specific binding was calculated as the difference between the radio¬ activity bound in the absence and that bound in the presence of unlabelled ligand. Binding data were cal¬ culated by the method of Scatchard (1949) and ana¬ lysed using a suitable computer program.
In order to examine the
cross-reactivity
of the
expressed ER with other steroid hormones, the crude fraction (400 pg protein) was incubated with 10 nmol
[3H]OE2/l in the presence of various concentrations of unlabelled OE2, progesterone, corticosterone or tes¬ tosterone. Binding of [3H]OE2 to the ER in each incu¬ bation was measured as described above. The relative molecular weight (Mr) and the quan¬ tity of the fusion protein in the crude fraction were determined by immunoblot analysis. The E. coli cells were sonicated, dialysed and then lyophilized. Sam¬ ples were resuspended in the SDS-PAGE sample buffer containing ß-mercaptoethanol and boiled to denature. Proteins were separated on discontinuous 7-5% SDS-PAGE (Laemmli, 1970), followed by electroblotting onto nitrocellulose membranes (Towbm, Staehelin & Gordon, 1979). The blot was immuno¬ stained employing the anti-human ER monoclonal antibody and reagents kit (Abbott Labs, IL, U.S.A.) according to the manufacturer's instructions. A band specific for the ER was visualized by incubating the membranes for 1 min with diaminobenzidine (DAB) solution. Alternatively, the membrane was stained with amido black and the staining density of the band corresponding to the fusion protein was compared with that of a standard protein in order to estimate the amount of the fusion protein in the crude fraction. Immunization The lyophilized E. coli cell crude fraction was sub¬ jected to SDS-PAGE and electroblotting as described above. The blot was briefly stained with amido black, destained with water, and the band corresponding to the Mr of the fusion protein, 180 000, was cut. The strips were pooled and kept at 4 °C until used for first immunization. The total amount of protein on the strips was roughly estimated at 180 pg from the silver staining density of the minigel (Phast system, Pharmacia Fine Chemicals, Uppsala, Sweden) (data not
shown).
Solubilized antigen was prepared for the booster. Proteins in the crude fraction were separated on SDSPAGE, and the gel was stained with sodium acetate (4 mol/1). A band at around 180 000 was cut and pooled. The fusion protein was eluted by immersing the gel slices in 10 mmol Tris-HCl/1, pH 7-4, including 0-1% SDS at 4 °C for 48 h with constant rotation. The eluate was concentrated on a Centricon-30 filter (Amicon, MA, U.S.A.) and dialysed against 10 mmol NaHC03/l overnight at room temperature. The pro¬ tein concentration in the solution was estimated at 250 pg/ml by silver staining of the mini gel. The solu¬ bilized antigen was kept frozen until used for booster
injections.
strips containing the fusion protein inserted subcutaneously into the nape of the neck of a male New Zealand White rabbit. At the same time, the Ribi adjuvant system (RAS, R-700; Ribi The membrane
was
ImmunoChem Research, MT, U.S.A.) emulsified with physiological saline only was applied to the cut skin. Four weeks after the initial immunization, the mixture of the solubilized protein (25 pg) and R-700 (1 ml) was injected into the animal according to the manufacturer's protocol. The booster shots were repeated every second week from then on.
sampling Blood samples were collected prior to the
Blood
immuniza¬ tion and once a week after the first booster injection. The production of antibodies was monitored by immunohistochemical examination of brain sections from the adult female rat. Biochemical characterization of the antiserum Uterine cytosol (400 pg protein), prepared from immature Sprague-Dawley rats as described previ¬ ously (Aihara, Kobayashi, Kimura et al. 1988), was simultaneously incubated with 10 nmol [3H]OE2/l and various amounts of IgGs obtained from either antiserum or preimmune serum in the presence of a 100fold molar excess of unlabelled OE2 at 4 °C overnight. The separation of bound and free hormone was per¬ formed using a hydroxylapatite method (Watson & Clark, 1980). Specific binding of [3H]OE2 to the ER was calculated by subtracting non-specific binding from total binding. Preimmune and immune serum IgGs were purified
by affinity chromatography
on protein A-Sepharose (Pharmacia Fine Chemicals). The serum, diluted with an equal volume of 3 mol NaCl/1 contain¬ ing 1-5 mol glycine/1 (pH 8-9) was loaded onto the column, which had been equilibrated with the same buffer. After washing with the same buffer, the IgG fraction was eluted with 0-1 mol citric acid/1 (pH 4-0), pooled, dialysed against distilled water and
CL-4B
lyophilized.
To examine whether the antiserum reacts with the ER occupied by OE2, the uterine cytosol (400 pg pro¬ tein) was incubated with 10 nmol [3H]OE2/l in the presence or absence of a 100-fold molar excess of unlabelled OE2 at 4 °C for 3 h. IgG fraction (800 pg protein) was added and incubated further at 4 °C overnight. At the end of the incubation, an equal vol¬ ume (0-25 ml) of TN buffer (10 mmol Tris-HCl/1, pH 7-4, containing 0-45 mol NaCl/1) was added and the sample was applied to Protein A-Sepharose CL-4B in a column (0-5 ml bed volume). After
washing with 20 ml TN buffer, the hormonereceptor-antibody complex was eluted with 5 ml 0-1 mol citric acid/1, pH 30. The radioactivity in an aliquot (2 ml) of the eluate was measured. The rat uterine cytosol was run on SDS-PAGE fol¬ lowed by electroblotting as described above. After quenching non-specific binding sites, the blot was sub¬ sequently incubated with either the antiserum or the
control serum at a final dilution of 1:5000 and 125Ilabelled F(ab')2 (2 pCi/ml ; Amersham International pic, Amersham, Bucks, U.K.) according to the method of Sambrook et al. (1989). The membrane was subjected to autoradiography using Kodak XOMAT film (Eastman Kodak, NY, U.S.A.) with an intensifying screen for 18 h at -80 °C. Protein concentration Unless otherwise stated, protein concentrations in the samples were determined by the method of Lowry, Rosebrough, Farr & Randall (1951) using bovine serum albumin (BSA) as standard.
Immunohistochemistry Adult female Sprague-Dawley rats maintained and bred in our laboratory were used. Animals were per¬ fused transcardically with 50 ml (0-1 mol/1) phos¬ phate-buffered saline solution (PBS) followed by 250 ml 4% paraformaldehyde in phosphate buffer (0-1 mol/1) (PB), pH 7-4, under Nembutal (Abbott Labs) anaesthesia. Tissues were removed, postfixed in the same fixative overnight at 4 °C, dehydrated, and then paraffin-embedded. Sections (5 pm thick) were cut and mounted on gelatin-coated glass slides. The sections were deparaffinized in xylene, hydrated through graded alcohols and rinsed with PBS. Immu¬ nohistochemical examination was done using the streptoavidin biotin staining kit (Histofine; Nichirei, Tokyo, Japan) using the manufacturer's instruction protocol. In brief, the sections were treated with 3% hydrogen peroxide in PBS for 15 min, washed exten¬
and incubated with the blocking re¬ of normal 10% agent goat serum for 30 min. Then the sections were sequentially incubated with the antiserum (1:5000 dilution with PBS including 1% BSA and 0-02% sodium azide) for 24-48 h at 4 °C in a humid moist chamber, the biotinylated goat antirabbit IgG for 10 min and the peroxidase-conjugated streptoavidin for 5 min. Each step was followed by three 5-min washes. The chromogen solution con¬ sisted of 005% DAB in PB (0-1 mol/1), pH 7-4, con¬ taining 00025% nickel chloride and cobalt acetate. The sections were incubated with the solution includ¬ ing 0009% hydrogen peroxide for 10 min. The slides were washed under running tap water, dehydrated
and coverslips applied. Some sections were counterstained with either neutral red or eosin prior to
dehydration.
Immunohistochemical controls
Specificity was established by incubating the sections with the antiserum preabsorbed with the lysate from either the control cells or cells containing the fusion protein. Additional controls were the incubation with the preimmune serum or the omission of primary or secondary antibody in the staining procedure. RESULTS
Expression of the functional ER in the transfected cells was confirmed by specific binding of [3H]OE2. Scatchard analysis of the binding data revealed a single class of high-affinity binding site with a dissoci¬ ation constant (Kd) of 1-49 nmol/l and a receptor density of 118 fmol/mg protein, for the crude fraction
012
008
0-04
sively with PBS,
01 Bound text-figure
1. Scatchard
(nmol/l)
analysis of tritiated oestradiol
([3H]OE2) binding to the fusion protein. Lysate (400 pg
of the cells containing the ß-galactosidase-oestrogen receptor fusion protein was incubated with increasing concentrations of [3H]OE2 in the absence (total binding) or presence (non-specific binding) of a 100-fold excess of unlabelled ligand for 20 min at 37 °C. Non-specific binding was 12-4-26-1% of the total binding. The binding data were analysed according to the method of Scatchard (1949). Dissociation constant was 1-49 nmol/l. Receptor density was 118 fmol/mg protein.
protein)
from cells containing the ER cDNA with orientation (Text-fig. 1). In contrast, that from the cells inserted with the control plasmid showed no
prepared sense
specific binding (data not shown). The Kd value was slightly higher than that for the ER of uterine origin (Kd 0-80 nmol/l, a value obtained in a parallel experi¬ ment employing the nucleus fraction of the rat uterus), which is probably due to conformational change or steric hindrance of the OE2-binding site by the fused ß-galactosidase in the expressed ER. The binding was specific for OE2. The binding of radiolabelled OE2 to the fusion protein was inhibited by increasing concentrations of unlabelled OE2, while
progesterone, corticosterone and
testosterone failed
displace bound [3H]OE2 (Text-fig. 2). Thus, the competition studies above confirmed that the fusion protein containing the functional ER was expressed to
in the E. coli cells.
ß-galactosidase. The minor species may represent its truncated form. From these observations, we consid¬ ered that the highest band of about 180 000 in lane 1 represented the fusion protein. The band on the membrane was cut, pooled and used for immunization. Three months after the start of immunization, the serum sample showed a positive reaction in the screen¬ ing. The specificity of the serum to the ER was con¬ firmed by three methods, i.e. binding assay, immunoblot analysis and histochemical examination. The effect of the antiserum on [3H]OE2 binding to the ER from the rat uterus is shown in Text-fig. 3. The IgG fraction obtained from the antiserum inhib¬ ited OE2 binding to the receptor in a dose-dependent manner, while that of the preimmune control serum had no effect. 100
100
)
80 60 60
o
oa
40
o
m
40
20 20. 10
0
text-figure
W- 1
100 1000 10 Final concentration (nmol/l)
2. Effects of steroid hormones
on
tritiated
([3H]OE2) binding to the fusion protein. Lysate (400 pg protein) of the cells containing the fusion protein was incubated with 10 nmol [3H]OE2/l in the pres¬ oestradiol
ence
of various concentrations of unlabelled steroid hor¬ OE2 (O), progesterone ( ), corticosterone (D)
mones:
and testosterone
100
1000
IgG ^g/tube)
(A).
The fusion protein in the crude fraction was electrophoretically separated and blotted on the nitrocellu¬ lose membrane. PI. 1, fig. 1, lane 1 shows the membrane which was stained with amido black. By immunoblot analysis employing the anti-human ER antibody, two bands were specifically detected (PI. 1, fig. 1, lane 2). The MT of the major band was around
180 000 which agrees with the estimated Mr of the fusion protein, i.e. 65 000 for the ER and 116 000 for
3. Effects of the antiserum on tritiated oes¬ tradiol ([3H]OE2) binding to the oestrogen receptor in the rat uterus. Cytosol fraction (400 pg protein) of the rat uterus was incubated with 10 nmol [3H]OE2/l or [3H]OE2 plus a 100-fold excess of unlabelled OE2 in the presence of various amounts of IgG purified from either the immune serum ( ) or the preimmune serum ( ), and specific binding was determined as described in the Materials and Methods. Specific binding, represented as a percentage of control binding (without IgG), is shown. Each value rep¬ resents the mean of duplicate determinations. text-figure
The antiserum was also able to recognize the OE2 ER complex. When the [3H]OE2-ER complex was
reacted with the IgG fraction from the antiserum and then applied to the Protein A-Sepharose CL-4B col¬ umn, 170-7 out of 206-2 fmol ER/mg protein applied was recovered in the eluate (recovery 82-8%). On the contrary, the IgG fraction of the preimmune serum failed to bind the [3H]OE2-ER complex. Eventually,
preoptic nucleus, the bed nucleus of the stria termin¬ alis, the septohypothalamic nucleus (PI. 2, figs 4 and 5), the arcuate nucleus, the ventrolateral part of the ventromedial hypothalamic nucleus (PI. 2, fig. 6), the
lateral habenula nucleus, the basolateral and medial nuclei of the amygdala and the medial central grey in the midbrain (data not shown). Positive signals in the brain were also restricted to the cell nuclei (PI. 2, fig. 5, thick arrows), although the staining density varied among cells even within the same region. Preincubation of the antiserum with the crude frac¬ tion of cells expressing the ß-galactosidase-ER fusion protein resulted in a marked reduction of labelling, while incubation with that of the control cells had little effect on the staining when used in immunohisto¬
chemistry (data
not
shown).
DISCUSSION 4. Binding of the antiserum to oestradiol (OE2)-oestrogen receptor (ER) complexes. Cytosol frac¬ tion (400 pg protein) of the rat uterus was incubated with
text-figure
10 nmol [3H]OE2/l for 3 h at 4 °C. The mixture was incuba¬ ted with IgG (800 pg) obtained from the antiserum or the preimmune serum and further incubated at 4 °C overnight. Immune complexes were isolated using a Protein ASepharose CL-4B column and radioactivity was measured. Each value represents the mean of duplicate determinations.
little radioactivity was eluted from the column under the acidic condition, i.e. only 1 5 fmol ER/mg protein (recovery 0-7%) (Text-fig. 4). It was therefore clear that the antiserum bound the ER molecules regardless of whether they were occupied or unoccupied by the ·
ligand.
The uterine
cytosol
was
subjected
to
immunoblot
the antiserum and ,25I-labelled antirabbit F(ab')2 fraction. The autoradiogram shows that the antiserum detected a single band of Mr 67 000 (PI. 1, fig. 2, lane 2), which was in good agreement with the Mr of the rat ER (Lubahn, McCarty &
analysis using
Koike et al. 1987). Histochemical examination was performed on 5 pm thick pituitary and brain sections from the adult female rat. When the pituitary section was reacted with the antiserum and then counterstained with eosin, positive signals were found only in the anterior lobe and not in the intermediate or posterior lobes (PI. 1, fig. 3). Immunoreactivity was restricted to the cell nuclei. Most of the cells in the anterior lobe were positive, but some cells remained negative. In the brain, positively stained cells were found in the cells adjacent to the vascular organ of the lamina termin¬ alis, the anteromedial preoptic nucleus, the medial
McCarty, 1985;
The ER
belongs
to the
steroid-thyroid
hormone
receptor superfamily and shares the common amino acid sequence with the other receptors in this family
(Evans, 1988). As we used a full length of the ER protein for the immunization, it was possible that the polyclonal antibody obtained in the present study might cross-react with the other receptors. However, several lines of evidence suggest that the antibody was directed exclusively to the ER. First, the binding of OE2
to the
receptor in the
uterus was inhibited in a
dose-dependent by the antibody, and the OE2 ER complex recognized by the antibody. Secondly, in immunoblot analysis of the uterus cyto¬ sol, the antibody detected a single band of Mr 67 000 which is in good agreement with the Mr of the rat ER (Lubahn et al. 1985; Koike et al. 1987). Finally, the distribution of the immunoreactivity in the rat tissue detected by the antibody was comparable with that demonstrated by autoradiography (Pfaff & Keiner, manner was
1973; Sheridan et al. 191 A) and immunohistochemis¬ try (Sar & Parikh, 1986). Moreover the distribution
of the ER mRNA detected by in-situ hybridization (Simerly, Chang, Muramatsu & Swanson, 1990; Hayashi & Okamura, 1992) was comparable with the
present study.
Recently, Furlow et al. (1990) produced antisera to synthetic peptide which corresponds to an amino acid sequence found in the hinge region of the rat ER. a
Those antisera also bind to both denatured and native receptors, and interact with unoccupied ER as well as that bound to OE2. It is therefore possible that the antibody obtained in this study may have components which recognize the same sequence as above, and/or other components which recognize the N-terminal part of the receptor sequence, because this region is
hypervariable in size and amino acid composition among the steroid and thyroid receptor superfamily (Evans, 1988) and is known to be highly antigenic (Wränge & Gustafsson, 1978). Oestrogen receptor immunoreactivity in the rat has been demonstrated by several groups, such as Sar & Parikh (1986), using a monoclonal antibody directed to the calf uterine cytosolic ER, and Gee et al. ( 1990) using a dinitrophenyl hapten-labelled rat anti-human ER monoclonal antibody. The former group showed ER immunoreactivity in the nuclei of target cells after OE2 stimulation. However, in the absence of hor¬ mone, they observed neither cytoplasmic nor nuclear staining in the brain. In contrast, we have seen posi¬ tive signals in the brain not only from ovary-intact rats but also from ovariectomized animals (data not shown). This discrepancy is probably due to the fact that our antiserum could recognize both forms of receptor, both occupied and unoccupied by the hor¬ mone. Although the rat brain contains only about 10% ER on a weight basis compared with the pituitary and uterus (Korach & Muldoon, 1974), the titre of our antibody is sufficient to detect the ER in paraffinembedded brain sections with a standard staining pro¬ cedure. On the other hand, when the monoclonal anti¬ body to the human receptor (Green et al. 1984) is applied to the paraffin sections, some modifications are required, such as DNase treatment (Shintaku & Said, i987) and cobalt chloride intensification (Hiort, Kwan & DeLellis, 1988). In the present study, we have also used nickel and cobalt bases, along with DAB. The main purpose, however, was to obtain pur¬ ple reaction products which allow counterstaining of the cytoplasm with red or pink dyes. The specific staining was exclusively confined to the nuclei of OE2 target cells. This result supports previ¬ ous findings that the receptor protein resides in the nuclear compartment regardless of the hormonal environment (King & Green, 1984; Welshons, Lieberman & Gorski, 1984). However, Ramm, Lauretano, Vrabel et al. (1988) have suggested that the preferential localization of ERs in the nuclei obtained with immunohistochemical techniques is due to a loss of loosely bound receptor in the cytosol dur¬ ing the fixation step. In this regard, it is of interest that we observed a variable density of specific staining among cells and tissues. In order to determine whether the staining density reflects the physiological and hor¬ monal conditions of animals, further studies are ongoing.
a polyclonal antibody to the rat ER produced by utilizing the cDNA which encodes a full length of the receptor protein. Several lines of evidence demonstrate that the antibody exclusively recognizes the ER. The titre of the antibody is
In summary,
was
sufficient to detect small amounts of ER in paraffinembedded sections, regardless of the hormonal environment. This antibody would therefore serve as a powerful tool to detect and analyse the ER in the rat.
ACKNOWLEDGEMENTS
This study was supported in part by Grants-in-Aid for Scientific Research from the Ministry of Education, Science and Culture to H.O.(No. 03760212) and to S.H.(No.03640636). Excellent technical assistance from H. Ueda is greatly appreciated.
REFERENCES Aihara, M., Kobayashi, H., Kimura, T., Hayashi, S. & Kato, J. (1988). Changes in uterine estrogen receptor concentrations
persistent estrous and persistent diestrous rats. Endocrinologìa Japónica 35, 57-70. Balthazart, J., Gahr, M. & Surlemont, C. (1989). Distribution of estrogen receptors in the brain of the Japanese quail : an immunocytochemical study. Brain Research 501, 205-214. Evans, M. R. (1988). The steroid and thyroid hormone receptor superfamily. Science 240, 889-895. Furlow, J. D., Ahrens, H., Mueller, G. C. & Gorski, J. (1990). Antisera to a synthetic peptide recognize native and denatured rat estrogen receptors. Endocrinology 127, 1028-1032. Gee, J. M. W., Nicholson, R. I., Jasani, B., Newman, G R. & Amselgruber, W. M. (1990). An immunocytochemical method for localization of estrogen receptors in rat tissues using a dinitrophenyl (DNP)-labeled rat monoclonal primary antibody. Journal of Histochemistry and Cytochemistry 38, 69-78. Gottesman, M. E., Adhya, S. & Das, A. (1980). Transcription by bacteriophage lambda gene product. Journal of Molecular Biology 140, 57-75. Green, G. L., Sobel, N. B., King, W. J. & Jensen, E. V. (1984). Immunochemical studies of estrogen receptors. Journal of Steroid Biochemistry 20, 51-56. Hayashi, S. & Okamura, H. (1992). Factors regulating sexual in
differentiation of the brain : Neonatal steroid treatment and development of the reproductive brain in the rat. In Brain Control of the Reproductive System, pp. 49-68. Ed. A. Yokoyama. Tokyo : Japan Scientific Societies Press and Boca Raton : CRC Press. Hiort, O., Kwan, P. W. L. & DeLellis, R. A. (1988). Immunohis¬ tochemistry of estrogen receptor protein in paraffin sections. American Journal of Clinical Pathology 90, 559-563. King, W. J. & Green, G L. (1984). Monoclonal antibodies local¬ ize oestrogen receptor in the nuclei of target cells. Nature 307, 745-747. Koch, M. (1990). Effects of treatment with estradiol and parental experience on the number and distribution of estrogenbinding neurons in the ovariectomized mouse brain. Neuroendocrinology 51, 505-514. Koike, S., Sakai, M. & Muramatsu, M. (1987). Molecular cloning and characterization of rat estrogen receptor cDNA. Nucleic Acids Research 15, 2489-2513.
Korach, K. S., Horigome, Y., Tomooka, Y., Yamashita, S., Newbold, R. R. & McLachlan, J. A. (1988). Immunodetection
of estrogen receptors in epithelial and mesenchymal tissues of the neonatal mouse uterus. Proceedings of the National Academy of Sciences of the U.S.A. 85, 3334-3337. Korach, K. S. & Muldoon, T. G (1974). Characterization of the interaction between 17ß-estradiol and its cytoplasmic receptor in the rat anterior pituitary gland. Biochemistry 13, 1932-1938. Laemmli, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680-685. Lowry, O. H., Rosebrough, N. J„ Farr, A. L. & Randall, R. J. (1951). Protein measurement with the Folin phenol reagent. Journal of Biological Chemistry 193, 265-275. Lubahn, D. B., McCarty, K. S„ Jr & McCarty, K. S., Sr. (1985). Electrophoretic characterization of purified bovine, porcine, murine, rat, and human uterine estrogen receptors. Journal of Biological Chemistry 260, 2515-2526. Moncharmont, B., Su, J. L. & Parikh, I. (1982). Monoclonal antibodies against estrogen receptor: interaction with different molecular forms and functions of the receptor. Biochemistry 21, 6916-6921. Pfaff, D. W. & Keiner, M. (1973). Atlas of E2-concentrating cells in the CNS of the female rat. Journal of Comparative Neurology 151, 121-158. Ramm, S., Lauretano, A. M., Vrabel, D. M., Pappas, C. A. & Tamura, H. (1988). Nuclear localization of hormone-free estrogen receptors by monoclonal antibodies could be a tissuefixation dependent artifact. Steroids 51, 425-439. Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning: A Laboratory Manual, edn 2. Cold Spring Harbor: Cold Spring Harbor Laboratory Press. Sar, M. & Parikh, I. (1986). Immunohistochemical localization of estrogen receptor in rat brain, pituitary and uterus with monoclonal antibodies. Journal of Steroid Biochemistry 24, 497-503. Scatchard, G (1949). The attraction of proteins for small molecules and ions. Annals of the New York Academy of Sciences 51, 660-672. Sheridan, P. J., Sar, M. & Stumpf, . E. (1974). Autoradio¬ graphic localization of 3H-estradiol or its metabolites in the central nervous system of the developing rat. Endocrinology 94, 1386-1390.
P. & Said, J. W. (1987). Detection of estrogen receptors with monoclonal antibodies in routinely processed formalin-fixed paraffin sections of breast carcinoma. American Journal of Clin¬ ical Pathology SI, 161-167. Simerly, R. B., Chang, C, Muramatsu, M. & Swanson, L. W. (1990). Distribution of androgen and estrogen receptor mRNA-containing cells in the rat brain : An in situ hybridiza¬ tion study. Journal of Comparative Neurology 194, 76-95. Stanley, K. K. & Luzio, J. P. (1984). Construction of a new family of high efficiency bacterial expression vectors: identification of cDNA clones coding for human liver proteins. EMBO Journals, 1429-1434. Stumpf, W. E., Narbitz, R. & Sar, M. (1980). Estrogen receptors in the fetal mouse. Journal of Steroid Biochemistry 12, 55-64. Towbin, H., Staehelin, T. & Gordon, J. (1979). Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets : procedure and some applications. Proceedings of the National Academy of Sciences of the U.S.A. 76, 4350-4354. Traish, ., Kim, N. & Wotiz, . H. (1989). Characterization of polyclonal antibodies to preselected domains of the human estrogen receptor. Endocrinology 125, 172-179. Walker, K. J., Bouzubar, N„ Robertson, J., Ellis, I. O., Elston, C. W., Blarney, R. W., Wilson, D. W., Griffiths, K. & Nicholson, R. I. (1988). Immunohistochemical localization of estrogen receptor in human breast tissue. Cancer Research 48, 6517-6522. Watson, C. S. & Clark, J. H. (1980). Heterogeneity of estrogen binding sites in mouse mammary cancer. Journal of Receptor Research 1, 91-111. Welshons, W. V., Lieberman, M. E. & Gorski, J. (1984). Nuclear localization of unoccupied oestrogen receptors. Nature 307, 747-749. Wränge, O. & Gustafsson, J. A. (1978). Separation of the hor¬ mone and DNA-binding sites of the hepatic glucocorticoid receptor by means of proteolysis. Journal of Biological Chemistry 253, 856-865. Yamamoto, K. R. (1985). Steroid receptor regulated transcription of specific genes and gene networks. Annual Review of Genetics 19, 209-252. Zaino, R. J., Clarke, C. L., Feil, P. D. & Satyaswaroop, P. G. (1989). Differential distribution of estrogen and progesterone receptors in rabbit uterus detected by dual immunofluorescence. Endocrinology 125, 2728-2734.
Shintaku,
DESCRIPTION OF PLATES Plate 1 figure 1. Immunoblot analysis of the fusion protein. The cell lysate (80 pg/lane) was run on 7-5% SDS-PAGE under reducing conditions, and electroblotted on a nitro¬ cellulose membrane. The blot was stained with amido black (lane 1 ) or immunostained employing the antihuman oestrogen receptor (ER) monoclonal antibody (lane 2). Two bands of the precipitation were detected in lane 2 : a major band at around 180 000 (arrow) and a minor one just below it. Relative molecular weights (MT) of the standard proteins are shown on the left. See text for details. figure 2. Immunoblot analysis of the ER in the rat uterus. Cytosol fractions ( 16 pg protein/lane) were subjected to 10% SDS-PAGE under reducing conditions, and blotted onto the membrane. The blot was incubated with the or the immune serum (lane 2) dilution of 1:5000 for 2 h at 4 °C, and was further incubated with l25I-labelled F(ab')2 fragment of anti-rabbit IgG (2 pCi/ml) for 2 h at room temperature. After extensive washing, the membrane was dried and autoradiographed at —80 °C for 18 h. A single band of precipitation at 67 000 was detected in lane 2 as indicated by an arrow.
preimmune serum (lane 1 ) at
a
3. ER-positive cells in the pituitary. In the anterior (a), the majority of the cells were positive (thick arrows), but some cells remained negative (thin arrows), while in the intermediate (i) and posterior (p) lobes no positive cells were found. Counterstained with eosin. figure
lobe
Bar
=
50 pm.
Plate 2 4. ER-positive neurones in the rostral hypothala¬ They were concentrated in the medial preoptic nucleus (pom). The positive cells were also found in the figure
mus.
bed nucleus of the stria terminalis (bst) and in the septohypothalamic nucleus (shy), ac, the anterior commis¬ sure; ox, the optic chiasm. Frontal section. Bar= 1 mm.
ER-positive (thick arrows) and -negative (thin in the medial preoptic nucleus in a parasagittal section counterstained with neutral red. figure
5.
arrows)
neurones
Rostral is the left side. Bar
=
50 pm.
figure 6. ER-positive neurones in the mediobasal hypo¬ thalamus. They were found in the arcuate nucleus (ar) and in the ventrolateral portion of the ventromedial hypo¬ thalamic nucleus (vm). v, the third ventricle. Frontal sec¬ tion. Bar= 1 mm.
PLATE
1
Fig.
Fig.
(Facing
p.
342)
Fig.
1
3
2
PLATE
Fig.
4
Fig.
Fig.
5
6
2
polyclonal antibody to the rat oestrogen receptor expressed in Escherichia coli: characterization and application to immunohistochemistry H.
Okamura, K. Yamamoto, S. Hayashi, A. Kuroiwa and M. Muramatsu
Department of Anatomy and Embryology, Tokyo Metropolitan Institute for Neuroscience, 2-6 Musashidai, Fuchu City, Tokyo 183, Japan *Department of Biology, School of Education, Waseda University, Shinjuku-ku, Tokyo 169-50, Japan tDepartment of Cell Biology, Research Institute for Tuberculosis and Cancer, Tohoku University, Sendai, Miyagi 980, Japan |Department of Biochemistry, Faculty of Medicine, University of Tokyo, Bunkyo-ku, Tokyo 113, Japan (Requests for offprints should be addressed to S. Hayashi) received 28 February 1992 ABSTRACT
A rat oestrogen receptor-\g=b\-galactosidase fusion protein was expressed using a pEX2/rat oestrogen receptor cDNA construct. Scatchard analysis of [3H]oestradiol-17\g=b\binding to the cell lysate revealed that the fusion protein had functional binding sites specific for oestradiol with a dissociation constant of 1\m=.\49nmol/l. The relative molecular weight (Mr) of the fusion protein was determined as 180 000 by immunoblot analysis of the cell lysate employing a monoclonal antibody to the human oestrogen receptor. The protein was isolated by means of SDS-PAGE and subsequent electroblotting. By immunization with the purified materials on nitrocellulose membrane, a polyclonal antibody to the rat oestrogen receptor was raised in a rabbit. Binding of [3H]oestradiol to the oestrogen receptor from the rat uterus was inhibited by the antibody in a dose\x=req-\ dependent manner. The antibody was also able to recognize the oestrogen receptor occupied by
[3H]oestradiol. Thus, the antibody could react with both forms of the receptor molecule, either occupied or unoccupied by the hormone. In immunoblot analysis of the cytosol fraction of the rat uterus, a single band of Mr 67 000, the size of the oestrogen receptor, was detected by the antibody. Moreover, when the antibody was applied to immunohistochemical examination of paraffin-embedded pituitary and brain sections of the rat, immunostaining was observed in cells of the anterior pituitary and in neurones in specific regions of the brain. The immunoreactivity was restricted exclusively to cell nuclei in both tissues. These results demonstrate that the polyclonal antibody obtained in the present study was specific to the oestrogen receptor, and that it would be a powerful tool to detect and analyse the receptors in various target tissues for oestrogen. Journal of Endocrinology (1992) 135, 333\p=n-\341
INTRODUCTION
Classically the oestrogen receptor (ER) has been detected in target tissues by autoradiography using tritium-labelled oestradiol-17ß (OE2) (Pfaff & Keiner, 1973; Sheridan, Sar & Stumpf, 1974). Although this technique satisfies all of the criteria of receptor binding (Stumpf, Narbitz & Sar, 1980), it is a disad¬ vantage that the method requires a very long process¬ ing time. On the other hand, antibodies to the ER have been shown to offer direct and accurate localiza¬ tion of the receptors by an immunohistochemical method. Several groups have produced antibodies
Oestrogen is responsible for a variety of reproductive actions which are mediated by a specific intracellular receptor in target tissues. Following binding, the hor¬ mone-receptor complex functions as a nuclear transcription factor that alters the expression of spe¬ cific genes (Yamamoto, 1985). Therefore, in order to understand the mechanism of action of oestrogen, it is of primary importance to determine the function and localization of the receptor in detail.
directed to the receptor protein from human mam¬ mary tumours (Green, Sobel, King & Jensen, 1984) or calf uterus (Moncharmont, Su & Parikh, 1982), or antibodies directed to synthesized peptides which correspond to preselected regions in the amino acid sequence of the molecule (Traish, Kim & Wotiz, 1989 ; Furlow, Ahrens, Mueller & Gorski, 1990). Of these antibodies, the monoclonal antibody to the human ER, which is commercially available, has been widely used to detect the ERs in a variety of animals includ¬ ing man (King & Green, 1984; Walker, Bouzubar, Robertson et al. 1988), mice (Korach, Horigome, Tomooka et al. 1988; Koch, 1990), rabbits (Zaino, Clarke, Feil & Satyaswaroop, 1989) and birds (Balthazart, Gahr & Surlemont, 1989). This rat monoclonal antibody, however, may not be suitable to determine the localization of the ER in rat tissues, since the secondary antibody (anti-rat IgG) binds throughout the tissue to endogenous immunoglobulins and produces widespread background staining (Gee, Nicholson, Jasani et al. 1990). Thus, to study the precise localization of the ER in the rat, antibodies to the ER raised in any species other than the rat are desirable. Recent success in the cloning of a cDNA for the rat ER (Koike, Sakai & Muramatsu, 1987) enabled us to obtain the receptor protein in large amounts by its expression in a culture of Escherichia coli cells. The expressed protein contained a functional ER which bound specifically to OE2. The protein, purified by means of sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) and subsequent electroblotting, was injected into a rabbit and an antiserum was obtained. Using biochemical and immunohisto¬ chemical examinations, we proved that the antiserum contained antibodies directed specifically to the ER.
MATERIALS AND METHODS
Expression The pEX2
of the rat ER in E. coli
(Stanley & Luzio, 1984) was pur¬ Boehringer Mannheim GmBH (Mannheim, Germany) and host strain N4830-1, a PI transductant of N4830 (Gottesman, Adhya & Das,
chased
vector
from
1980), was obtained from Pharmacia Fine Chemicals (Uppsala, Sweden). cDNA for the rat ER (Koike et al. 1987) was modified to introduce an EcoRI site at
the 5' terminus on the sequence for the construction produce the ß-galactosidase-rat ER fusion protein. The fragment encompassing from codon 68, through translational termination codon to some additional 3' non-coding sequence was excised by SacII and EcoRI double digestion. The EcoRI site at the 3' end of the fragment was artificially introduced during cDNA to
et al. 1987). To introduce another EcoRI site at the 5' end of the fragment, oligonucleotides corresponding to the region from 391 to 461 in the cDNA (5-AATTCAACGCCGCCGCCGCCGCCGCG and 5-GCGGCGGCGGCGGCGTTG) were synthesized with a single base substitution G at position 393 of the cDNA sequence to A. This substitution creates the EcoRI site at the 5' end of the oligonucleotide without changing the amino acid sequence of the rat ER portion. The oligonucleotides were annealed to form a double strand, ligated with the SacII-EcoRI fragment and the resulting chimeric fragment was cloned into the EcoRI site of pUC18. The sequence of the exchanged region was then deter¬ mined. The EcoRI fragment was then excised, trans¬ ferred to the EcoRI site of the pEX2 vector and a clone containing the cDNA with the sense orientation for ß-galactosidase was selected. The final construct was introduced into N4830-1 host cells for protein production. The rat ER portion in the fusion protein corresponds to amino acid 61 to carboxyl terminus of native rat ER and contains DNA binding and ligand binding domains. The protein was expressed in the host cells according to the method described by Sambrook, Fritsch & Maniatis (1989). Briefly, the cells were incubated in two x YT broth (bacto-trypton, 16 g/l ;bacto-y east extract, 10 g/l ; NaCl, 5 g/l; pH 7-0) containing 50 pg ampicillin/ml at 30 °C with constant agitation. Absorbance at 650 nm was monitored. When it reached 0-8, the culture was transferred to 42 °C and incubated for a further 2 h with constant agitation. The cell pellet was stored at —80 °C until further processing.
cloning (Koike
Biochemical characterization of the fusion protein The E. coli cells were sonicated and centrifuged at 10 000 g for 15 min. The pellet was washed three times with 10 mmol Tris-HCl/1, pH 7-4, containing 1 mmol EDTA/1,1 mmol dithiothreitol/1 and 10 mmol sodium molybdate/1 (TEDM buffer) and resuspended in the buffer. The suspension (crude fraction) containing 400 pg protein was incubated with increasing concen¬ trations of [3H]OE2 (New England Nuclear, MA, U.S.A.) in a total volume of 0-25 ml with or without a 100-fold excess of unlabelled OE2 for 20 min at 37 °C. The mixture was then cooled on ice and centri¬ fuged at 1000 g for 30 min. After washing the pellets three times with 3 ml TEDM buffer, radiolabelled materials were extracted with 1 ml absolute ethanol. Scintillation fluid (3 ml) was added to each extract and the radioactivity was measured. Specific binding was calculated as the difference between the radio¬ activity bound in the absence and that bound in the presence of unlabelled ligand. Binding data were cal¬ culated by the method of Scatchard (1949) and ana¬ lysed using a suitable computer program.
In order to examine the
cross-reactivity
of the
expressed ER with other steroid hormones, the crude fraction (400 pg protein) was incubated with 10 nmol
[3H]OE2/l in the presence of various concentrations of unlabelled OE2, progesterone, corticosterone or tes¬ tosterone. Binding of [3H]OE2 to the ER in each incu¬ bation was measured as described above. The relative molecular weight (Mr) and the quan¬ tity of the fusion protein in the crude fraction were determined by immunoblot analysis. The E. coli cells were sonicated, dialysed and then lyophilized. Sam¬ ples were resuspended in the SDS-PAGE sample buffer containing ß-mercaptoethanol and boiled to denature. Proteins were separated on discontinuous 7-5% SDS-PAGE (Laemmli, 1970), followed by electroblotting onto nitrocellulose membranes (Towbm, Staehelin & Gordon, 1979). The blot was immuno¬ stained employing the anti-human ER monoclonal antibody and reagents kit (Abbott Labs, IL, U.S.A.) according to the manufacturer's instructions. A band specific for the ER was visualized by incubating the membranes for 1 min with diaminobenzidine (DAB) solution. Alternatively, the membrane was stained with amido black and the staining density of the band corresponding to the fusion protein was compared with that of a standard protein in order to estimate the amount of the fusion protein in the crude fraction. Immunization The lyophilized E. coli cell crude fraction was sub¬ jected to SDS-PAGE and electroblotting as described above. The blot was briefly stained with amido black, destained with water, and the band corresponding to the Mr of the fusion protein, 180 000, was cut. The strips were pooled and kept at 4 °C until used for first immunization. The total amount of protein on the strips was roughly estimated at 180 pg from the silver staining density of the minigel (Phast system, Pharmacia Fine Chemicals, Uppsala, Sweden) (data not
shown).
Solubilized antigen was prepared for the booster. Proteins in the crude fraction were separated on SDSPAGE, and the gel was stained with sodium acetate (4 mol/1). A band at around 180 000 was cut and pooled. The fusion protein was eluted by immersing the gel slices in 10 mmol Tris-HCl/1, pH 7-4, including 0-1% SDS at 4 °C for 48 h with constant rotation. The eluate was concentrated on a Centricon-30 filter (Amicon, MA, U.S.A.) and dialysed against 10 mmol NaHC03/l overnight at room temperature. The pro¬ tein concentration in the solution was estimated at 250 pg/ml by silver staining of the mini gel. The solu¬ bilized antigen was kept frozen until used for booster
injections.
strips containing the fusion protein inserted subcutaneously into the nape of the neck of a male New Zealand White rabbit. At the same time, the Ribi adjuvant system (RAS, R-700; Ribi The membrane
was
ImmunoChem Research, MT, U.S.A.) emulsified with physiological saline only was applied to the cut skin. Four weeks after the initial immunization, the mixture of the solubilized protein (25 pg) and R-700 (1 ml) was injected into the animal according to the manufacturer's protocol. The booster shots were repeated every second week from then on.
sampling Blood samples were collected prior to the
Blood
immuniza¬ tion and once a week after the first booster injection. The production of antibodies was monitored by immunohistochemical examination of brain sections from the adult female rat. Biochemical characterization of the antiserum Uterine cytosol (400 pg protein), prepared from immature Sprague-Dawley rats as described previ¬ ously (Aihara, Kobayashi, Kimura et al. 1988), was simultaneously incubated with 10 nmol [3H]OE2/l and various amounts of IgGs obtained from either antiserum or preimmune serum in the presence of a 100fold molar excess of unlabelled OE2 at 4 °C overnight. The separation of bound and free hormone was per¬ formed using a hydroxylapatite method (Watson & Clark, 1980). Specific binding of [3H]OE2 to the ER was calculated by subtracting non-specific binding from total binding. Preimmune and immune serum IgGs were purified
by affinity chromatography
on protein A-Sepharose (Pharmacia Fine Chemicals). The serum, diluted with an equal volume of 3 mol NaCl/1 contain¬ ing 1-5 mol glycine/1 (pH 8-9) was loaded onto the column, which had been equilibrated with the same buffer. After washing with the same buffer, the IgG fraction was eluted with 0-1 mol citric acid/1 (pH 4-0), pooled, dialysed against distilled water and
CL-4B
lyophilized.
To examine whether the antiserum reacts with the ER occupied by OE2, the uterine cytosol (400 pg pro¬ tein) was incubated with 10 nmol [3H]OE2/l in the presence or absence of a 100-fold molar excess of unlabelled OE2 at 4 °C for 3 h. IgG fraction (800 pg protein) was added and incubated further at 4 °C overnight. At the end of the incubation, an equal vol¬ ume (0-25 ml) of TN buffer (10 mmol Tris-HCl/1, pH 7-4, containing 0-45 mol NaCl/1) was added and the sample was applied to Protein A-Sepharose CL-4B in a column (0-5 ml bed volume). After
washing with 20 ml TN buffer, the hormonereceptor-antibody complex was eluted with 5 ml 0-1 mol citric acid/1, pH 30. The radioactivity in an aliquot (2 ml) of the eluate was measured. The rat uterine cytosol was run on SDS-PAGE fol¬ lowed by electroblotting as described above. After quenching non-specific binding sites, the blot was sub¬ sequently incubated with either the antiserum or the
control serum at a final dilution of 1:5000 and 125Ilabelled F(ab')2 (2 pCi/ml ; Amersham International pic, Amersham, Bucks, U.K.) according to the method of Sambrook et al. (1989). The membrane was subjected to autoradiography using Kodak XOMAT film (Eastman Kodak, NY, U.S.A.) with an intensifying screen for 18 h at -80 °C. Protein concentration Unless otherwise stated, protein concentrations in the samples were determined by the method of Lowry, Rosebrough, Farr & Randall (1951) using bovine serum albumin (BSA) as standard.
Immunohistochemistry Adult female Sprague-Dawley rats maintained and bred in our laboratory were used. Animals were per¬ fused transcardically with 50 ml (0-1 mol/1) phos¬ phate-buffered saline solution (PBS) followed by 250 ml 4% paraformaldehyde in phosphate buffer (0-1 mol/1) (PB), pH 7-4, under Nembutal (Abbott Labs) anaesthesia. Tissues were removed, postfixed in the same fixative overnight at 4 °C, dehydrated, and then paraffin-embedded. Sections (5 pm thick) were cut and mounted on gelatin-coated glass slides. The sections were deparaffinized in xylene, hydrated through graded alcohols and rinsed with PBS. Immu¬ nohistochemical examination was done using the streptoavidin biotin staining kit (Histofine; Nichirei, Tokyo, Japan) using the manufacturer's instruction protocol. In brief, the sections were treated with 3% hydrogen peroxide in PBS for 15 min, washed exten¬
and incubated with the blocking re¬ of normal 10% agent goat serum for 30 min. Then the sections were sequentially incubated with the antiserum (1:5000 dilution with PBS including 1% BSA and 0-02% sodium azide) for 24-48 h at 4 °C in a humid moist chamber, the biotinylated goat antirabbit IgG for 10 min and the peroxidase-conjugated streptoavidin for 5 min. Each step was followed by three 5-min washes. The chromogen solution con¬ sisted of 005% DAB in PB (0-1 mol/1), pH 7-4, con¬ taining 00025% nickel chloride and cobalt acetate. The sections were incubated with the solution includ¬ ing 0009% hydrogen peroxide for 10 min. The slides were washed under running tap water, dehydrated
and coverslips applied. Some sections were counterstained with either neutral red or eosin prior to
dehydration.
Immunohistochemical controls
Specificity was established by incubating the sections with the antiserum preabsorbed with the lysate from either the control cells or cells containing the fusion protein. Additional controls were the incubation with the preimmune serum or the omission of primary or secondary antibody in the staining procedure. RESULTS
Expression of the functional ER in the transfected cells was confirmed by specific binding of [3H]OE2. Scatchard analysis of the binding data revealed a single class of high-affinity binding site with a dissoci¬ ation constant (Kd) of 1-49 nmol/l and a receptor density of 118 fmol/mg protein, for the crude fraction
012
008
0-04
sively with PBS,
01 Bound text-figure
1. Scatchard
(nmol/l)
analysis of tritiated oestradiol
([3H]OE2) binding to the fusion protein. Lysate (400 pg
of the cells containing the ß-galactosidase-oestrogen receptor fusion protein was incubated with increasing concentrations of [3H]OE2 in the absence (total binding) or presence (non-specific binding) of a 100-fold excess of unlabelled ligand for 20 min at 37 °C. Non-specific binding was 12-4-26-1% of the total binding. The binding data were analysed according to the method of Scatchard (1949). Dissociation constant was 1-49 nmol/l. Receptor density was 118 fmol/mg protein.
protein)
from cells containing the ER cDNA with orientation (Text-fig. 1). In contrast, that from the cells inserted with the control plasmid showed no
prepared sense
specific binding (data not shown). The Kd value was slightly higher than that for the ER of uterine origin (Kd 0-80 nmol/l, a value obtained in a parallel experi¬ ment employing the nucleus fraction of the rat uterus), which is probably due to conformational change or steric hindrance of the OE2-binding site by the fused ß-galactosidase in the expressed ER. The binding was specific for OE2. The binding of radiolabelled OE2 to the fusion protein was inhibited by increasing concentrations of unlabelled OE2, while
progesterone, corticosterone and
testosterone failed
displace bound [3H]OE2 (Text-fig. 2). Thus, the competition studies above confirmed that the fusion protein containing the functional ER was expressed to
in the E. coli cells.
ß-galactosidase. The minor species may represent its truncated form. From these observations, we consid¬ ered that the highest band of about 180 000 in lane 1 represented the fusion protein. The band on the membrane was cut, pooled and used for immunization. Three months after the start of immunization, the serum sample showed a positive reaction in the screen¬ ing. The specificity of the serum to the ER was con¬ firmed by three methods, i.e. binding assay, immunoblot analysis and histochemical examination. The effect of the antiserum on [3H]OE2 binding to the ER from the rat uterus is shown in Text-fig. 3. The IgG fraction obtained from the antiserum inhib¬ ited OE2 binding to the receptor in a dose-dependent manner, while that of the preimmune control serum had no effect. 100
100
)
80 60 60
o
oa
40
o
m
40
20 20. 10
0
text-figure
W- 1
100 1000 10 Final concentration (nmol/l)
2. Effects of steroid hormones
on
tritiated
([3H]OE2) binding to the fusion protein. Lysate (400 pg protein) of the cells containing the fusion protein was incubated with 10 nmol [3H]OE2/l in the pres¬ oestradiol
ence
of various concentrations of unlabelled steroid hor¬ OE2 (O), progesterone ( ), corticosterone (D)
mones:
and testosterone
100
1000
IgG ^g/tube)
(A).
The fusion protein in the crude fraction was electrophoretically separated and blotted on the nitrocellu¬ lose membrane. PI. 1, fig. 1, lane 1 shows the membrane which was stained with amido black. By immunoblot analysis employing the anti-human ER antibody, two bands were specifically detected (PI. 1, fig. 1, lane 2). The MT of the major band was around
180 000 which agrees with the estimated Mr of the fusion protein, i.e. 65 000 for the ER and 116 000 for
3. Effects of the antiserum on tritiated oes¬ tradiol ([3H]OE2) binding to the oestrogen receptor in the rat uterus. Cytosol fraction (400 pg protein) of the rat uterus was incubated with 10 nmol [3H]OE2/l or [3H]OE2 plus a 100-fold excess of unlabelled OE2 in the presence of various amounts of IgG purified from either the immune serum ( ) or the preimmune serum ( ), and specific binding was determined as described in the Materials and Methods. Specific binding, represented as a percentage of control binding (without IgG), is shown. Each value rep¬ resents the mean of duplicate determinations. text-figure
The antiserum was also able to recognize the OE2 ER complex. When the [3H]OE2-ER complex was
reacted with the IgG fraction from the antiserum and then applied to the Protein A-Sepharose CL-4B col¬ umn, 170-7 out of 206-2 fmol ER/mg protein applied was recovered in the eluate (recovery 82-8%). On the contrary, the IgG fraction of the preimmune serum failed to bind the [3H]OE2-ER complex. Eventually,
preoptic nucleus, the bed nucleus of the stria termin¬ alis, the septohypothalamic nucleus (PI. 2, figs 4 and 5), the arcuate nucleus, the ventrolateral part of the ventromedial hypothalamic nucleus (PI. 2, fig. 6), the
lateral habenula nucleus, the basolateral and medial nuclei of the amygdala and the medial central grey in the midbrain (data not shown). Positive signals in the brain were also restricted to the cell nuclei (PI. 2, fig. 5, thick arrows), although the staining density varied among cells even within the same region. Preincubation of the antiserum with the crude frac¬ tion of cells expressing the ß-galactosidase-ER fusion protein resulted in a marked reduction of labelling, while incubation with that of the control cells had little effect on the staining when used in immunohisto¬
chemistry (data
not
shown).
DISCUSSION 4. Binding of the antiserum to oestradiol (OE2)-oestrogen receptor (ER) complexes. Cytosol frac¬ tion (400 pg protein) of the rat uterus was incubated with
text-figure
10 nmol [3H]OE2/l for 3 h at 4 °C. The mixture was incuba¬ ted with IgG (800 pg) obtained from the antiserum or the preimmune serum and further incubated at 4 °C overnight. Immune complexes were isolated using a Protein ASepharose CL-4B column and radioactivity was measured. Each value represents the mean of duplicate determinations.
little radioactivity was eluted from the column under the acidic condition, i.e. only 1 5 fmol ER/mg protein (recovery 0-7%) (Text-fig. 4). It was therefore clear that the antiserum bound the ER molecules regardless of whether they were occupied or unoccupied by the ·
ligand.
The uterine
cytosol
was
subjected
to
immunoblot
the antiserum and ,25I-labelled antirabbit F(ab')2 fraction. The autoradiogram shows that the antiserum detected a single band of Mr 67 000 (PI. 1, fig. 2, lane 2), which was in good agreement with the Mr of the rat ER (Lubahn, McCarty &
analysis using
Koike et al. 1987). Histochemical examination was performed on 5 pm thick pituitary and brain sections from the adult female rat. When the pituitary section was reacted with the antiserum and then counterstained with eosin, positive signals were found only in the anterior lobe and not in the intermediate or posterior lobes (PI. 1, fig. 3). Immunoreactivity was restricted to the cell nuclei. Most of the cells in the anterior lobe were positive, but some cells remained negative. In the brain, positively stained cells were found in the cells adjacent to the vascular organ of the lamina termin¬ alis, the anteromedial preoptic nucleus, the medial
McCarty, 1985;
The ER
belongs
to the
steroid-thyroid
hormone
receptor superfamily and shares the common amino acid sequence with the other receptors in this family
(Evans, 1988). As we used a full length of the ER protein for the immunization, it was possible that the polyclonal antibody obtained in the present study might cross-react with the other receptors. However, several lines of evidence suggest that the antibody was directed exclusively to the ER. First, the binding of OE2
to the
receptor in the
uterus was inhibited in a
dose-dependent by the antibody, and the OE2 ER complex recognized by the antibody. Secondly, in immunoblot analysis of the uterus cyto¬ sol, the antibody detected a single band of Mr 67 000 which is in good agreement with the Mr of the rat ER (Lubahn et al. 1985; Koike et al. 1987). Finally, the distribution of the immunoreactivity in the rat tissue detected by the antibody was comparable with that demonstrated by autoradiography (Pfaff & Keiner, manner was
1973; Sheridan et al. 191 A) and immunohistochemis¬ try (Sar & Parikh, 1986). Moreover the distribution
of the ER mRNA detected by in-situ hybridization (Simerly, Chang, Muramatsu & Swanson, 1990; Hayashi & Okamura, 1992) was comparable with the
present study.
Recently, Furlow et al. (1990) produced antisera to synthetic peptide which corresponds to an amino acid sequence found in the hinge region of the rat ER. a
Those antisera also bind to both denatured and native receptors, and interact with unoccupied ER as well as that bound to OE2. It is therefore possible that the antibody obtained in this study may have components which recognize the same sequence as above, and/or other components which recognize the N-terminal part of the receptor sequence, because this region is
hypervariable in size and amino acid composition among the steroid and thyroid receptor superfamily (Evans, 1988) and is known to be highly antigenic (Wränge & Gustafsson, 1978). Oestrogen receptor immunoreactivity in the rat has been demonstrated by several groups, such as Sar & Parikh (1986), using a monoclonal antibody directed to the calf uterine cytosolic ER, and Gee et al. ( 1990) using a dinitrophenyl hapten-labelled rat anti-human ER monoclonal antibody. The former group showed ER immunoreactivity in the nuclei of target cells after OE2 stimulation. However, in the absence of hor¬ mone, they observed neither cytoplasmic nor nuclear staining in the brain. In contrast, we have seen posi¬ tive signals in the brain not only from ovary-intact rats but also from ovariectomized animals (data not shown). This discrepancy is probably due to the fact that our antiserum could recognize both forms of receptor, both occupied and unoccupied by the hor¬ mone. Although the rat brain contains only about 10% ER on a weight basis compared with the pituitary and uterus (Korach & Muldoon, 1974), the titre of our antibody is sufficient to detect the ER in paraffinembedded brain sections with a standard staining pro¬ cedure. On the other hand, when the monoclonal anti¬ body to the human receptor (Green et al. 1984) is applied to the paraffin sections, some modifications are required, such as DNase treatment (Shintaku & Said, i987) and cobalt chloride intensification (Hiort, Kwan & DeLellis, 1988). In the present study, we have also used nickel and cobalt bases, along with DAB. The main purpose, however, was to obtain pur¬ ple reaction products which allow counterstaining of the cytoplasm with red or pink dyes. The specific staining was exclusively confined to the nuclei of OE2 target cells. This result supports previ¬ ous findings that the receptor protein resides in the nuclear compartment regardless of the hormonal environment (King & Green, 1984; Welshons, Lieberman & Gorski, 1984). However, Ramm, Lauretano, Vrabel et al. (1988) have suggested that the preferential localization of ERs in the nuclei obtained with immunohistochemical techniques is due to a loss of loosely bound receptor in the cytosol dur¬ ing the fixation step. In this regard, it is of interest that we observed a variable density of specific staining among cells and tissues. In order to determine whether the staining density reflects the physiological and hor¬ monal conditions of animals, further studies are ongoing.
a polyclonal antibody to the rat ER produced by utilizing the cDNA which encodes a full length of the receptor protein. Several lines of evidence demonstrate that the antibody exclusively recognizes the ER. The titre of the antibody is
In summary,
was
sufficient to detect small amounts of ER in paraffinembedded sections, regardless of the hormonal environment. This antibody would therefore serve as a powerful tool to detect and analyse the ER in the rat.
ACKNOWLEDGEMENTS
This study was supported in part by Grants-in-Aid for Scientific Research from the Ministry of Education, Science and Culture to H.O.(No. 03760212) and to S.H.(No.03640636). Excellent technical assistance from H. Ueda is greatly appreciated.
REFERENCES Aihara, M., Kobayashi, H., Kimura, T., Hayashi, S. & Kato, J. (1988). Changes in uterine estrogen receptor concentrations
persistent estrous and persistent diestrous rats. Endocrinologìa Japónica 35, 57-70. Balthazart, J., Gahr, M. & Surlemont, C. (1989). Distribution of estrogen receptors in the brain of the Japanese quail : an immunocytochemical study. Brain Research 501, 205-214. Evans, M. R. (1988). The steroid and thyroid hormone receptor superfamily. Science 240, 889-895. Furlow, J. D., Ahrens, H., Mueller, G. C. & Gorski, J. (1990). Antisera to a synthetic peptide recognize native and denatured rat estrogen receptors. Endocrinology 127, 1028-1032. Gee, J. M. W., Nicholson, R. I., Jasani, B., Newman, G R. & Amselgruber, W. M. (1990). An immunocytochemical method for localization of estrogen receptors in rat tissues using a dinitrophenyl (DNP)-labeled rat monoclonal primary antibody. Journal of Histochemistry and Cytochemistry 38, 69-78. Gottesman, M. E., Adhya, S. & Das, A. (1980). Transcription by bacteriophage lambda gene product. Journal of Molecular Biology 140, 57-75. Green, G. L., Sobel, N. B., King, W. J. & Jensen, E. V. (1984). Immunochemical studies of estrogen receptors. Journal of Steroid Biochemistry 20, 51-56. Hayashi, S. & Okamura, H. (1992). Factors regulating sexual in
differentiation of the brain : Neonatal steroid treatment and development of the reproductive brain in the rat. In Brain Control of the Reproductive System, pp. 49-68. Ed. A. Yokoyama. Tokyo : Japan Scientific Societies Press and Boca Raton : CRC Press. Hiort, O., Kwan, P. W. L. & DeLellis, R. A. (1988). Immunohis¬ tochemistry of estrogen receptor protein in paraffin sections. American Journal of Clinical Pathology 90, 559-563. King, W. J. & Green, G L. (1984). Monoclonal antibodies local¬ ize oestrogen receptor in the nuclei of target cells. Nature 307, 745-747. Koch, M. (1990). Effects of treatment with estradiol and parental experience on the number and distribution of estrogenbinding neurons in the ovariectomized mouse brain. Neuroendocrinology 51, 505-514. Koike, S., Sakai, M. & Muramatsu, M. (1987). Molecular cloning and characterization of rat estrogen receptor cDNA. Nucleic Acids Research 15, 2489-2513.
Korach, K. S., Horigome, Y., Tomooka, Y., Yamashita, S., Newbold, R. R. & McLachlan, J. A. (1988). Immunodetection
of estrogen receptors in epithelial and mesenchymal tissues of the neonatal mouse uterus. Proceedings of the National Academy of Sciences of the U.S.A. 85, 3334-3337. Korach, K. S. & Muldoon, T. G (1974). Characterization of the interaction between 17ß-estradiol and its cytoplasmic receptor in the rat anterior pituitary gland. Biochemistry 13, 1932-1938. Laemmli, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680-685. Lowry, O. H., Rosebrough, N. J„ Farr, A. L. & Randall, R. J. (1951). Protein measurement with the Folin phenol reagent. Journal of Biological Chemistry 193, 265-275. Lubahn, D. B., McCarty, K. S„ Jr & McCarty, K. S., Sr. (1985). Electrophoretic characterization of purified bovine, porcine, murine, rat, and human uterine estrogen receptors. Journal of Biological Chemistry 260, 2515-2526. Moncharmont, B., Su, J. L. & Parikh, I. (1982). Monoclonal antibodies against estrogen receptor: interaction with different molecular forms and functions of the receptor. Biochemistry 21, 6916-6921. Pfaff, D. W. & Keiner, M. (1973). Atlas of E2-concentrating cells in the CNS of the female rat. Journal of Comparative Neurology 151, 121-158. Ramm, S., Lauretano, A. M., Vrabel, D. M., Pappas, C. A. & Tamura, H. (1988). Nuclear localization of hormone-free estrogen receptors by monoclonal antibodies could be a tissuefixation dependent artifact. Steroids 51, 425-439. Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning: A Laboratory Manual, edn 2. Cold Spring Harbor: Cold Spring Harbor Laboratory Press. Sar, M. & Parikh, I. (1986). Immunohistochemical localization of estrogen receptor in rat brain, pituitary and uterus with monoclonal antibodies. Journal of Steroid Biochemistry 24, 497-503. Scatchard, G (1949). The attraction of proteins for small molecules and ions. Annals of the New York Academy of Sciences 51, 660-672. Sheridan, P. J., Sar, M. & Stumpf, . E. (1974). Autoradio¬ graphic localization of 3H-estradiol or its metabolites in the central nervous system of the developing rat. Endocrinology 94, 1386-1390.
P. & Said, J. W. (1987). Detection of estrogen receptors with monoclonal antibodies in routinely processed formalin-fixed paraffin sections of breast carcinoma. American Journal of Clin¬ ical Pathology SI, 161-167. Simerly, R. B., Chang, C, Muramatsu, M. & Swanson, L. W. (1990). Distribution of androgen and estrogen receptor mRNA-containing cells in the rat brain : An in situ hybridiza¬ tion study. Journal of Comparative Neurology 194, 76-95. Stanley, K. K. & Luzio, J. P. (1984). Construction of a new family of high efficiency bacterial expression vectors: identification of cDNA clones coding for human liver proteins. EMBO Journals, 1429-1434. Stumpf, W. E., Narbitz, R. & Sar, M. (1980). Estrogen receptors in the fetal mouse. Journal of Steroid Biochemistry 12, 55-64. Towbin, H., Staehelin, T. & Gordon, J. (1979). Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets : procedure and some applications. Proceedings of the National Academy of Sciences of the U.S.A. 76, 4350-4354. Traish, ., Kim, N. & Wotiz, . H. (1989). Characterization of polyclonal antibodies to preselected domains of the human estrogen receptor. Endocrinology 125, 172-179. Walker, K. J., Bouzubar, N„ Robertson, J., Ellis, I. O., Elston, C. W., Blarney, R. W., Wilson, D. W., Griffiths, K. & Nicholson, R. I. (1988). Immunohistochemical localization of estrogen receptor in human breast tissue. Cancer Research 48, 6517-6522. Watson, C. S. & Clark, J. H. (1980). Heterogeneity of estrogen binding sites in mouse mammary cancer. Journal of Receptor Research 1, 91-111. Welshons, W. V., Lieberman, M. E. & Gorski, J. (1984). Nuclear localization of unoccupied oestrogen receptors. Nature 307, 747-749. Wränge, O. & Gustafsson, J. A. (1978). Separation of the hor¬ mone and DNA-binding sites of the hepatic glucocorticoid receptor by means of proteolysis. Journal of Biological Chemistry 253, 856-865. Yamamoto, K. R. (1985). Steroid receptor regulated transcription of specific genes and gene networks. Annual Review of Genetics 19, 209-252. Zaino, R. J., Clarke, C. L., Feil, P. D. & Satyaswaroop, P. G. (1989). Differential distribution of estrogen and progesterone receptors in rabbit uterus detected by dual immunofluorescence. Endocrinology 125, 2728-2734.
Shintaku,
DESCRIPTION OF PLATES Plate 1 figure 1. Immunoblot analysis of the fusion protein. The cell lysate (80 pg/lane) was run on 7-5% SDS-PAGE under reducing conditions, and electroblotted on a nitro¬ cellulose membrane. The blot was stained with amido black (lane 1 ) or immunostained employing the antihuman oestrogen receptor (ER) monoclonal antibody (lane 2). Two bands of the precipitation were detected in lane 2 : a major band at around 180 000 (arrow) and a minor one just below it. Relative molecular weights (MT) of the standard proteins are shown on the left. See text for details. figure 2. Immunoblot analysis of the ER in the rat uterus. Cytosol fractions ( 16 pg protein/lane) were subjected to 10% SDS-PAGE under reducing conditions, and blotted onto the membrane. The blot was incubated with the or the immune serum (lane 2) dilution of 1:5000 for 2 h at 4 °C, and was further incubated with l25I-labelled F(ab')2 fragment of anti-rabbit IgG (2 pCi/ml) for 2 h at room temperature. After extensive washing, the membrane was dried and autoradiographed at —80 °C for 18 h. A single band of precipitation at 67 000 was detected in lane 2 as indicated by an arrow.
preimmune serum (lane 1 ) at
a
3. ER-positive cells in the pituitary. In the anterior (a), the majority of the cells were positive (thick arrows), but some cells remained negative (thin arrows), while in the intermediate (i) and posterior (p) lobes no positive cells were found. Counterstained with eosin. figure
lobe
Bar
=
50 pm.
Plate 2 4. ER-positive neurones in the rostral hypothala¬ They were concentrated in the medial preoptic nucleus (pom). The positive cells were also found in the figure
mus.
bed nucleus of the stria terminalis (bst) and in the septohypothalamic nucleus (shy), ac, the anterior commis¬ sure; ox, the optic chiasm. Frontal section. Bar= 1 mm.
ER-positive (thick arrows) and -negative (thin in the medial preoptic nucleus in a parasagittal section counterstained with neutral red. figure
5.
arrows)
neurones
Rostral is the left side. Bar
=
50 pm.
figure 6. ER-positive neurones in the mediobasal hypo¬ thalamus. They were found in the arcuate nucleus (ar) and in the ventrolateral portion of the ventromedial hypo¬ thalamic nucleus (vm). v, the third ventricle. Frontal sec¬ tion. Bar= 1 mm.
PLATE
1
Fig.
Fig.
(Facing
p.
342)
Fig.
1
3
2
PLATE
Fig.
4
Fig.
Fig.
5
6
2
