Haloperidol-sensitive (+)[3H]SKF-10,047 binding sites (σ sites) exhibit a unique distribution in rat brain subcellular fractions

European Journal of Pharmacology - Molecular Pharmacolog), Section, 188 (1990) 211-218

2!1

Elsevier EJPMOL 90069

Haloperidol-sensitive (+) [3HISKF-10,IM7 binding sites (o sites) exhibit a unique distribution in rat brain subcellular fractions David J. M c C a n n a n d T s u n g - P i n g Su Neurochemistry Unit, Neuropharrnacology Laboratory, Addiction Research Center. National Institute on Drag Abuse, P.O. Bo.~ 5t80, Baltimore, hID 21224, ~S.A.

Received 3 November1989,accepted 4 Janua~ 1990

The distribution of haloperidol-sensitive ( +)[3H]N-ailylnormetazocine ((+)[3H]SKF-10,047) binding sites (o sites) "n subeellular fractions of rat brain homogenates was extensively characterized. In synaptosomal fractions, enriched in choline ~cetyltransferase activity, o sites were present in lower concentrations than in whole brain homogenates. On the other hand, microsomal and myelin fractions were found to be enriched in o sites. A similar pattern of enrichment was seen for 5'-nueleotidase activity, a general plasma membrane marker. However, subsequent experiments in which microsomes were subfractionaled o~ ~near sucrose gradients led to the recovery of o sites Gver a significantly lower density range than 5'-nueleotidase activity or ATP-stimulated [3H]ouabain binding, an additional plasma membrane marker. In addition, previously reported distributions of a number of other subceldular markers, including those for endoplasmic reticnium, were found to contrast with the observed distribution of o sites. It is conclvded that rat brain o sites are not concentrated at synaptic regions of plasma membrane. However, the possibility that o sites are localized to specialized areas of nonsynaptie plasma membrane cannot be excluded. Brain (rat); tf Receptors; (Subcellular fractionation)

1. Introduction ( + )N-Allylnormetazocine ((+)SKF-10,047) is a benzomorphan derivative which causes psychoto~fimetie effects in humans (Keats and Telford, 1964). As a result, the demonstration of haloperidol-sensitive binding sites for this drug in brain tissue (Su, 1982) has provoked a great deal of interest; sites at which both psyehotomimetic and antipsychotic drugs bind could be logical candidates for involvement in the pathogenesis of schizophrenia. These haloperidol-sensitive ( + ) S K F 10,047 binding sites, termed o sites (Quirion et al.,

Correspondence to: Dr. Tsung-PingSu, NIDA Addiction Research Center P.O. Box 5180 Baltimore, MD 21224 U.S.A.

1988), have recently been shown to exist in homogenates of peripheral blood leukocytes (Wolfe et al., 1988) and various endocrine tissues (Wolfe et al., 1989). With the additional observation that o sites bind progesterone with high affinity ( S u e t al., 1988), it has been suggested that o sites may function as receptors which link the endocrine, nervous and immune systems. In contrast with the hypothesis that a sites are receptors, it has recently been suggested that they may represent a type of membrane-bound microsomal enzyme (McCann etal., 1989). The suggestion was, in part, based on the demonstration that a sites are more concentrated in mierosomal than in synaptosomal fractions of rat brain homogenates. However, subcellular fractions other than synaptosomes and microsomes were not evaluated

0922-4106/90/$03.50 © 1990 Hsevier Science PublishersB.V. (Biomedical Division)

212 in that study. In addition, due to the presence of plasma membrane vesicles in microsomal fractions of rat brain, the possibility that e sites are localized to nonsynaptic areas of plasma membrane was noted (McCann et al., 1989). The goal of the present study was to examine more thoroughly the subcellular localization of a sites in rat brain with a specific aim to determine if a sites are plasma membrane constituents. Both synaptic and nonsynaptic regions of plasma membrane were considered as potential sites of localization. Correspondingly, direct comparisons were made between distributions of haloperidol-sensifive (+)[3H]SKF-10,047 binding and distributions of established synaptosomal and general plasma membrane markers in subcellular fractions of rat brain.

2. Materials and methods

2.1. Fractionation procedures 2.1.1. Primary fractionation of rat brain Brains were removed from male Fischer 344 rats and promptly homogenized in 10 volumes of ice-cold 0.32 M sucrose using a motor-driven teflon pestle and tapered glass tube (clearance = 0.15 to 0.23 ram). Homogenates were then subjected to differential centrifugation as described by Gray and Whittaker (1962). Briefly, successive centrifugation steps resulted in the collection of three pellets of decreasing particle size: P1 (termed the nuclear fraction), the pellet formed by centrifugation of the homogenate at 1000 × g for 10 min; P2 (termed the crude synaptosomal fraction), the pellet formed by centrifugation of the resulting supernatant at 20000 × g for 30 rain; and P3 (termed the microsomal fraction), the pellet formed by centrifugation of the second supernatant at 100000 × g for 60 rain. As in all subsequent fractionation procedures, centrifugation steps and collection of fractions were carded out at 4°C. Samples of original homogenates and aliquets of each fraction, suspended in 0.32 M sucrose, were used fresh for (+)[3H]SKF-10,047 binding assays or frozen at - 2 0 ° C for subsequent assays of marker enzymes.

2.1.2. Subfractionation of P2 Freshly prepared P2 fractions of rat brain homogenates, suspended in 0.32 M sucrose, were subfractionated as described by Gray and Whittaker (1962). After centrifugation of the sample on a discontinuous gradient of 0.8 and 1.2 M sucrose at 113000 x g for 70 min, three subfractions were collected: P2A (termed the myelin fraction), the material collected at the 0.32/0.8 M sucrose interface; P2B (termed the synaptosomal fraction), the material collected at the 0.8/1.2 M sucrose interface; and P2c (termed the mitochondrial fraction), the pellet formed under the gradient. Fractions P2A and P2B were diluted to 25 ml with ice-cold 0.32 and 0.45 M sucrose, respectively, and collected by centrifugation at 100000 x g for 60 min. Pellets were then resuspended in ice-cold 0.32 M sucrose.

2.1.3. Subfractionation of P3 A modification of the primary fractionation procedure described above was utilized to obtain P3 fractions. The initial centrifugation step (1000 x g for 10 rain) was omitted and, as a result, P1 and P2 fractions were collected together, as one pellet, following the 10 000 × g centrifugation step. The 10000 x g supernatant was then centrifuged as described above and the resulting pellet (P3) was suspended in ice-cold 0.25 M sucrose containing 3 m M imidazole HC1 (pH 7.4 at 25 ° C). Four ml of the suspended P3 fraction, containing one-half of the material resulting from fractionation of a whole rat brain, was loaded onto a 34 ml linear sucrose gradient (20% to at least 60% w / v ) and centrifuged at l l 3 0 0 0 x g for 17 h. Sucrose solutions used to form the gradient were buffered with imidazole HC1 (3 mM, pH 7.4 at 25 ° C). From each gradient, ten subfractions of approximately equal volume were collected, from top to bottom, using Pasteur pipets. Concentrations of sucrose were then measured using a refractometer. Each subfraction was subsequently diluted to a volume of 24 ml, with a final sucrose concentration of 0.25 M, by the addition of appropriate volumes of ice-cold H20 and 0.25 M sucrose. Following centrifugation at 100 000 x g for 60 rain, pellets were suspended in ice-cold 0.32 M sucrose.

2t3

2.2. Assay procedures 2.2.1. Haloperidol-sensitive ( + )[sH]SKF-10,047

binding In a final volume of 0.25 ml, membranes were incubated with 50 nM (+)[3H]SKF-10,047 in the presence of 50 mM Tris HCI buffer (pH 8.0 at 25°C). All incubations were carried out for 60 min at 25°C, conditions appropriate for attainment of equilibrium binding. Haloperidol-sensitive binding of ( + )[3H]SKF-10,047 was defined as the difference between total binding and that measured in the presence of 10/tM haloperidol. In all experiments, concentrations of tissue were such that less than 5% of radioligand was bound. Incubations were stopped by adding 5 ml of ice-cold 100 mM Tris HCI (pH 7.7 at 25°C; rinse buffer) followed by vacuum filtration through Whatman G F / C glass fiber filters which had been pretreated by soaking in 0.5% polyethylenimine. Filters were then rinsed three times with 5 ml of rinse buffer and radioactivity retained on filters was measured by liquid scintillation spectrometry. In all experiments, haloperidol-sensitive binding to filters alone was simultaneously measured and subtracter, ¢'rom the total haloperidol-sensitive binding in the presence of tissue to obtain the amount of haloperidol-sensitive binding to each tissue sample.

2.2.2. Marker enzymes Choline acetyltransferase activity was assayed by monitoring the formation of [14C]acetylcholine using [14C]acetyl coenzyme A as an acetyl donor, as described by Fonnum (1975). The activity of 5'-nucleotidase was assayed by monitoring the conversion of [14C]adenosine 5'-monophosphate to [14C]adenosine, as described by Gentry and Olsson (1975). ATP-stimulated [3H]ouabain binding, which directly reflects the amount of ouabain-sensitive Na +, K+-ATPase in a sample, was measured by a modification of the method described by Caspers et al. (1987). Briefly, membranes were incubated for 60 min at 25 ° C with 40 nM [3H]ouabain in the presence of 50 mM imidazole HC1 (pH 7.4 at 25°C), 10 mM NaCI, and 10 mM MgCI 2. ATP-stimulated binding was defined as the difference between that measured in

the presence and absence of 5 mM ATP. Incubations were stopped and filters (Whatman GF/C) were rinsed three times with 5 ml of ice-cold 5 mM Tris HC1 (pH 7.4 at 25°C).

2.3. Data analysis and expression of results Results of studies in which rat brain homogenates were subjected to primary fractionation, as well as results of studies in which P2 fractions were subfractionated, are expressed both in terms of the relative specific activity and the percentage of total recovered activity corresponding to each fraction or subfraction. Relative specific activities were calculated for choline acetyltransferase and 5'-nucleotidase by dividing the specific activity measured in each fraction or subfraction by that measured in a sample of the original homogenate. For haloperidol-sensitive (+)[3H]SKF-10,047 binding, values of the amount of binding/rag protein were utilized in a similar fashion. Thus, for each assay the relative specific activity of the original homogenate is defined as 1.0 and values greater than 1.0 indicate enrichment of activity. Protein content was measured by the method of Lowry et al. (19S~) using bovine serum albumin as a standard. Density distribution histograms, constructed as described by Beaufay et al. (1964), were used to analyze data from P3 subfractionation studies. Histograms were normalized by the method of Bowers and de Duve (1967) to allow averaging of data from individual experiments. The median equilibrium density of ligand binding or enzyme activity was calculated from each normalized histogram.

2.4. Drugs and chemicals Ultrapure sucrose (Boehringer Mannheim Bitchemicals, Indianapolis, IN) was used in all experiments. (+)[3H]SKF-10,047 (31.6 to 59.0 C i / mmol), [3H]Ouabain (14.6 Ci/mmol), [14C]adenosine 5'-raonophosphate (561 mCi/mmol), and [laC]acetyl coenzyme A (4.0 mCi/mmol) were obtained from NEN Research Products (Boston, MA). Haloperidol was supplied by McNeil Labs Inc. (Fort Washington, PA). Unlabelled acetyl

214 an average of 71 dpn~(N = 6; 50 n M radioligand). This level of binding was measured during each experiment and subtracted from the total haloperidol-sensitive binding measured in the presence of tissue to obtain the a m o u n t of haloperidol-sensitive binding to tissue alone. The distribution profiles of haloperidol-sensitive (+)[3H]SKF-10,047 binding, choline acetyltransferase activity (synaptosomal marker), and 5'-nuclcotidase activity (general plasma membrane marker) were compared in three subcelhilar fractions obtained by differential centrifugation of rat brain homogenates (fig. 1). The apparent distributions of a sites and choline acetyltransferase activity are dissimilar; notably, while the relative specific activity of a sites is greatest in the microsomal fraction, the relative specific activity of choline acetyltransferase is greatest in the crude synaptosomal fraction. O n the other hand, no substantial difference is apparent between the distribution of a sites and that of the general plasma membrane marker, 5'-nuclcotidase. Crude synaptosomal fractions (i.e., P2; fig- 1) were further fractionated into components of low density (P2A; myelin fraction), medium density (P2B; synaptosomal fraction), and high density

coer~yme A was obtained from ICN Biochemicals (Cleveland, OH). All other chemicals and reagents were obtained from Sigma Chemical C o m p a n y (St. Louis, MO).

3.

Results

Preliminary (+)[3H]SKF-10,047 binding assays, in which the method of Largent et at. (1986) was utilized without modification, revealed a substantial level of haloperidol-sensitive binding to W h a t m a n G F / B filters in the absence of tissue; using 50 n M radioligand, this binding averaged 476 d p m (N = 4). Despite intensive efforts, a pretreatment for filters superior to soaking in 0.5% polyethylenimine could not be found. A change to W h a t m a n G F / C filters was prompted by the observation of an approximately 50% reduction in haloperidol-sensitive binding. In addition, increasing the concentration of the rinse buffer (Tris-HC1) from 5 m M to 100 m M was found to provide a noticeable degree of protection against radioligand binding to filters. With these changes, halopeddol-sensitive (+)[3H]SKF-10,047 binding to filters in the absence of tissue was reduced to

i

:1 III I 31

~<

0

++i o:

--I--

i

~

.

3

.

0

2

.

.

-

---r-

I NI-I

:® P1

P2

P3

HALOPERIDOL~E (+)I~I]SKF-10,047 BINDING

0

P1

P2

-

P3

CHOLINE ACETYLTRANSFERASE ACTIVITY

P1

F2

P3

S'-NUCLEO1]DASE ACTIVITY

Fig. 1. Distribution patterns of haloperidnl-sensitive(+)[ 3H]SKF-10,047 binding, choline acetyltransferaseactivity and 5'-nucleotidase activityin primary particulate fractions of rat brain homogenates.P1, the nuclear pellet formed by centrifugationof homogenate at 1000× g for 10 rain; P2, the crude synaptosomalpellet formed by centrifngation of the resulting supernatant at 20000x g for 30 rain.; I)3, the microsomalpellet formed by centrifugation of the second supernatant at 100000x g for 60 min. Open bars: relative specific activity,with the relative specificactivity of the original homogenate defined as 1.00. Shaded bars: percentage of recovered activity, with reference to the total amount of activity recovered in P1, P2 and P3 as 100%. Means and associated S.E. are shown for three different experiments.

215 n

-r 2 |

I

I

o r'--I

r-'~

I

0

I. r - - 7

2O P2A

P2B

P~'C

HALOPERIDOL-SENSITIVE (÷)[3HJSKF-10,047BINDING

PaA

P2II

P2C

CHOLINEACETYLTRANSFERASE ACTIVITY

P2A

PgB

P2C

5'-NUCLEOI"~DASE ACTIVITY

Fig. 2. Distribution patterns of haloperidol-sensitive (+)[3H]SKF-10,047 binding, choline acetyltransferase activity and 5'-nucleotiaase activity in P2 subfractions. Following primary fractionation of rat brain homogenates, as described in the legend to fig. 1, crude synaptosomal fractions were subfractionated by isopycnic cemrifugation as described by Gray and Whittaker (1962). Pza, the myelin fraction collected at the 0.32/0.8 M sucrose interface; P2B, the synaptosomal fraction collected at the 0,8/1.2 M sucrose interface; P2c, the mitochondrial fraction collected as a pellet under the gradient. Open bars: relative specific activity, with the relative specific activity of the original homogenate defined as 1.00. Shaded bars: percentage of recovered activity, with reference to the total amount of activity recovered in P2A, P2B and P2c as 100%, Means and associated S.E. are shown for three different expedraents.

(P2c; mitochondrial fraction). Distribution profiles of o sites and the two marker enzymes were then determined (fig. 2). The apparent distributions of a sites and choline acetyltransferase are clearly distinguishable. In contrast with haloperidol-sensirive (+)[3H]SKF-10,047 binding, which is most enriched in the myelin fraction, choline acetyltransferase is most enriched in the synaptosomal fraction. In terms of recovery (shaded bars), it is noteworthy that 857o of choline acetyltransferase activity was recovered in the synaptosomal fraction. In contrast, only 39% of haloperidol-sensitive (+)[3H]SKF-10,047 binding was synaptosomal. Again, however, a definite similarity is apparent with regard to the distribution of o sites and that of the general plasma membrane marker, 5 '-nucleotidase. Figure 3 shows results of experiments in which rat brain microsomes were subfractionated on linear sucrose gradients. Average median buoyant densities (indicated by vertical arrows) and their 95% confidence intel"vals (indicated by associated horizontal bars) are 1.113 (i.096-1.130) g / m l for haloperidol-sensitive (+)[3H]SKF-10,047 binding, 1.139 (1.135-1.143) g / m l for 5'-nucleotidase activity, and 1.147 (1.138-1.156) g / m l for ATP-stimu-

lated [3H]ouabain binding. An analysis of variance (Studentized range test) revealed that the median buoyant density of haloperidol-sensitive (+)[3H]SKF-10,047 binding is significantly lower than the median buoyant density of each of the two plasma membrane markers (k = 0.010 for P = 0.05).

4. Discussion

Results of the present study are in agreement with a previous report describing low concentrations of o sites in rat brain synaptosomes when compared to rrficrosomes (McCann et al., 1989). Based on relative specific activity values shown in figs, 1 and 2, o sites were found to be three to four times more concentrated in microsomes than in synaptosomes. Through a more extensive evaluation of rat brain subcellnlar fractions, the present study revealed that tr sites are enriched not only in the microsomal fraction (P3; fig. 1), but also in the myelin fraction (P~; fig. 2). This distribution was found to sharply contrast with that of the synaptosomal marker enzyme, choline acetyltransferase, providing strong evidence that o

216

HALOPERIDOL-SENSiTIVE (,)[3H]SKF.10,047 BINDING

~,o

i F~-q i., -I-

20

T

--t--T-

[

__

__

uCLEOTIDASE ACTIVITY

ATP-STIMULATED [SHJOUABAIN BINDING

~ 10 5

1.0B

1 3 1 8 DENSITY(g/m|)

1.23

I ~ i t i i 30 35 40 45 50 SS SUCROSE CONOENTRAllON (% WN)

15

Fig. 3. Distribution patterns of haloperidol-sensitive (+)[3H]SKF-10,047 binding, 5'-nucleotidase activity, and ATPstimulated [3H]ouabain binding as established by isopycnic eentrifugation of rat brain microsomzsin linear sucrosegradients. Abscissa:densityis divided into six equal regionswithin the range of 1.08 to 1.23g/ml using an increment of 0.025; shaded regions represent all material recovered at densities below 1.08 and above 1.23 g/ml and are each constructedover an arbitrary width of 0.025 g/ml. Ordinate: frequency= the amountof activity recoveredin a givenregion/the total amount of activity recovered/the density spanned by the region. As such, the total area of each histogramequals 1 and the area of any regionis equal to the fractionof activity recovered in that region. Each histogramrepresentsmeans and associated S.E. for three different experiments.Vertical arrowsindicate average median densities and associated horizontal bars represent the corresponing95% confidenceintervals.

sites are not concentr,:ted at synapses within rat brain. Based on the results shown in figs. 1 and 2, the subcellular distribution of o sites can not be differentiated from that of the general plasma membrane marker, 5'-nucleotidase. However, upon further fractionation, microsomal o sites were shown to exhibit a significantly lighter median buoyant density than either 5'-nucleotidase cr the additional plasma membrane marker, ATP-stimulated [3Hlouabain binding (fig. 3). In a similar study, microsomal musearinic receptors from rat brain were shown to exhibit a density distribution similar to that of 5'-nueleotidase (Laduron and Janssen, 1979). Taken together, these findings suggest that o sites are not located on vesicles formed from plasma membrane. This conclusion is tempered by the possibility that some plasma membrane constituents might, due to localization to specialized regions of plasma membrane or to specific types of cells, demonstrate density distributions which differ from those of the general plasma membrane markers examined. Nevertheless, subcellular organclles other than plasma membrane must be considered as potential sites of o-site localization. The possibility that o sites are a component of myelin is suggested by their enrichment in the P2A fraction; cleetron microscopic studies have demonstrated that the P2A fraction is enriched in myelin (e.g., Gray and Whittaker, 1962). In addition, subfractionation of mierosomes on linear sucrose gradients has revealed that CNPase, a myelin marker enzyme, is also distributed over a lower-density range than are plasma membrane markers (Bansal et al., 1985). However, results of autoradiographic studies using (+)[3H]-3-(3-hydroxyphenyl)-N-(1-propyl)piperidine to label o sites (Gundlach et al., 1986) bring into serious question the possibility that o sites are associated with myelin. In both guinea pig and rat spinal cord, autoradiographic grains were confined mainly to areas of gray matter. Comparisons of the observed distribution of o sites in rat brain subecllular fractions with previously reported distributions of several other marker enzymes failed to reveal any striki~:~g similarities. For example, although enriehmen~ of o

217 sites in the P3 fraction (fig. 1) would be expected if o sites were a constituent of endoplasmic reficulum, enrichment in the P2A fraction (fig. 2) would not be expected. Three different endoplasmic reticulum markers were reported to exhibit a relative specific activity of less than 1 in the P2A fraction (Possmayer et al., 1979). In contrast, enrichment by a factor of more than 2 was observed for o sites in the P2A fraction (fig. 2). Thus, while an association of o sites with endoplasmic reticulum cannot be ruled out based on the present data, it seems that endoplasmic reticulum cannot be the only locus of enrichment for o sites. Other comparisons strongly suggest that o sites are not prim,~rily localized to nuclei or mitochondria. It is reasonable to expect that localization to nuclei would result in the majority of o sites being recovered in the Pt fraction, as demonstrated for D N A (e.g., Kai and Hawthorne, 1966), and that localization to mitochondria would result in a marked enrichment of o sites in the P2c fraction, as demonstrated for succinic dehydrogenase (e.g., Whittaker, 1962; Kai and Hawthorne, 1966). These patterns of distribution are clearly not apparent from figs. 1 and 2. Because no established subcellular marker appears to precisely parallel the distribution of o sites, the subcellular localization cf o sites remains subject to speculation. In fight of the recent finding that progesterone is capable of binding to o sites with high affinity ( S u e t al., 1988), details of a report by Bramiey and Menzies (1988) describing the association of endogenous progesterone with a unique particulate fraction of h u m a n corpus luteum m a y be noteworthy. Following fractionation of corpus luteum homogenates on linear sucrose gradients, membrane-associated progesterone was recovered over a lighter density range than were plasma membrane markers, a result similar to that obtained for o sites in the present study (fig. 3)~ Despite the measurement of numerous other subcellular markers, a distribution pattern similar to that of progesterone was not observed. The possibility exists, therefore, that membrane-associated progesterone m a y reflect binding of the hormone to corpus luteal o sites. This possibility is strengthened by the additional finding that o sites are extremely dense in maturing ovarian follicles

(Wolfe et al., 1989). While the suggested relationship between membrane-associated progesterone and o sites is highly speculative, it would seem to merit investigation. In summary, o sites were differentiated from both synaptosomal and general plasma membrane markers based on their distribution in subcelhilar fractions of rat brain. Additional comparisons with previously reported distributions of several other subcellular markers also failed to reveal compelling similarities. As a result, the exact subeellular localization of o sites remains subject to further investigation. One noteworthy possibility is that o sites m a y be plasma membrane constituents which, due to locaUzation to specific regions of plasma membrane or to specific types of cells, demonstrate subtle differences from the general plasma m e m b : a n e markers examined in the present study. O n the other hand, locali7ation of o sites to synaptic regions of plasma membrane appears highly unlikely based on the dramatic differences seen in comparisons with the synaptosomal marker, choline acetyltransferase. It is concluded that electron microscopic studies, perhaps utilizing immunocytochemical techniques, will be essential for defining the subcellular location of o sites.

References Bansal, V.S., H. Hatt*~fi, D. Orihel and J.N. Kanfer, 1985, Distribution of selectedphospholipidmodifyingenzymesin rat brain rnicrosomal subfractions prepared by density gradient zonal rotor centrifugation, Neurochem. Res. 10. 439. Beaufay.H., P. Jacques, P. Baudhuin, O.Z. SelLinger,J. Bcrthet and C. de Duve, 1964, Tissue fractionation studies 18: resolution of mitochondrial fractions from rat liver into three distinct populations of cytoplasmicparticles by means of density equilibration in various gradients, Biochem. J. 92, 184. Bowers. W.E. and C. de Duve, 1967, Lysosomesin lymphoid tissue If: ilitracellular distribution of acid hydrolases. J. Cell Biol. 32, 339. Bramley. T.A. and G.S. Menzies, 1988, Association of progesterone with a unique particulate fraction of the human corpus lutenm, J. Endocrinol. 116, 307. Caspers, M.L., ILD. Schwartz, R, Labarca and S.M. Paul 1987, Autoradiographic visualization and characterization of [3H]ouabain binding to the Na ÷, K+-ATPas¢ of rat brain and pineal, Brain Res. 409, 335.

218 Founum, F., 1975, A rapid biochemical method for the determinati,~vt of choline acetyhransferase, J. Neurochem. 24, 407. Gentry, M.K. and R.A. Olsson, 1975, A simple, specific, radioisotopic assay for 5'-nucleotidase, Anal. Biochem. 64, 624. Gray, E.G. and V.P. Whittaker, 1962, The isolation of nerve endings from brain: an electron-microscopic study of cell fragments derived by homogenization and centrlf:;gafion, J. Anat. 95, 79. Gandlach, A.L., B.L. Largent, and S.H. Snyder, 1986, Autoradiographic localization of sigma receptor binding sites in guinea pig and rat central nervoaz system with (+)[3H!-3(3-hydroxyphenyl)-N-(1-p:opyl)piperidive, J. NeurcscL 6, 1757. Kai, M. and J.N. Hawthorne, 1966, Incorporation of injected [3~P]phospha~e into fl'~e phosphoinositides of subcelhiar fractions from young rat brain, Biochem. J. 98, 62. Keats, A. and J. Telford, 1964, Narcotic antagonists as analgesics, clinical aspects, in: Molecular Modification in Drug Design, Advances in Chemistry, Series 45, ed. R.F. Gould, (American Chemical Society, Washington, D.C.) p. 170. Ladaron, P.M. and P.F.M. Janssen, 1979, Characterization and subcellular localization of brain muscarinic receptors labelled in vivo by [3H]dexetimide, J. Nenrochem. 33, 1223. Largent, B.L., A.L. Gundlaeh, and S.H. Snyder, 1986, Pharmacologi~zal and autoradiographic discrimination of sigma and phencyclidine receptor binding sites in brain with (+)-[3H]SKF-10,047, (+)-[3Hl-3-[3-hydroxyphenyl]N-(1-propyl)piperidine and [3H]-l-[1-(2-thienyl)cyclohcx-yl]piperidine, J. Pharmacol. Exp. Ther. 238, 739. Lowry, O.H., N.J. Rosebrough, A.L. Farr and R.J. Randall, 1951, Protein measurement with the Folin phenol reagent, J. Biol. Chem. 193, 265.

McCann, D.J., R.A. Rabin, S. Rens-Domiano and J.C. Winter, 1989, Phencyclidine/SKF-10,047 b~nding sites: evaluation of function, Pharmacol. Biochem. Behav. 32, 87. Possmayer, F., L. Kleine, G Duwe, P.J. Stewart-DeHaan, T. Wong, C.F.C. MacPherson and P.G.R. Harding, 1979, Differences in the subcellular and subsynaptoscmal distribution of the putative endoplasmic reticularn markers, NADPH-cytochrome c reductase, estrone sulfate sulfohydrolase and CDP-choline-diacylglycerol chollnephosphotransferee in rat brain, J. Neurochem. 32, 889. Qu~6on, R., R. Chicheportiehe, P.C. Contreras, K.M. Johnson, D. Lodge, S.W. Tam, LH. Woods and S.R. Zakin, 1988, Classification of nomenclature of phencyclidine and sigma receptor sites, in: Sigrga and Phencyciidine-llke Compounds as Molecular Pr,.~bes in Biology, eds. E.F. Domino and J.oM. Kamenka (NPP Bookz, Ann Arbor) p. 601. Su, T.-P., 1982, Evidence for sigma opioid receptor: Binding of [3H]SKF-10,047 to etorphint~inaccessible sites in guinea-pig brain, J. Phan-nacol. Exp. Th,:r. 223, 284. Su, T.-P., E.D. London and J.H. Jaffe, !988, Steroid binding at o receptors suggests a link between endocrine, nervous, and immune systems, Science 240, 219. Whittaker, V.P., 1962, Pharmacological studies with isolated cell components, Biochem. Pharmacol. 9, 61. Wofle, S.A, S.G. Culp and E.B. De Souza, 1989, o-Receptors in endocrine organs: identification, characterization, and autoradiographic localization in rat pituitary, adrenal, testis, and ovary, Endocrinology 124, 1160. Wolfe, S A., C. Kulsakdinan, G. Battaglla, J.H. Jaffe and E.B. De Souza, 1988, Initial identification and characterization of sigma receptors on human peripheral blood leukocytes, J. Pharmacol. Exp. Ther. 2r47, 1114.