Neuroscience Vol. 49, No. 4, pp. 893-902, Printed in Great Britain

0306-4522/92

1992

$5.00 + 0.00

Pergamon Press Ltd 0 1992IBRO

CHANGES OF RAT STRIATAL NEURONAL MEMBRANE MORPHOLOGY AND STEROID CONTENT DURING THE ESTROUS CYCLE M. MORISSETTE,*~L.-M. GARCIA-SEGURAJ A. BBLANGERt and T. Dr PAoLo*t$ *School of Pharmacy, Lava1 University, Ste-Foy, Qdbec, Canada GlK 7P4 TDepartment of Molecular Endocrinology, CHUL Research Centre, Ste-Foy, Quebec, Canada GlV 4G2 SInstituto Cajal, C.S.I.C., Dr. Arce 37, 28002, Madrid, Spain AIrstract-It is well documented that sex steroids affect striatal dopamine systems. However, the mechanism(s) of these hormonal effects in the striatum is still not well understood. We now report that gonadal steroid hormones during the estrous cycle affect the morphology and steroid hormone content of the rat striatum. Rats displaying at least two consecutive estrous cycles were included in this study as well as a group of female rats ovariectomized two weeks before being killed. The striatum was dissected from one half of each brain and used for morphological studies. From the other half of each brain, the striatum was dissected and steroid hormone concentrations in striatum and the remainder of the brain were determined. Tissues and serum concentrations of 17B-estradiol, progesterone and prolactin were measured by specific radioimmunoassays. Serum 17fi-estradiol and prolactin concentrations peaked in proestrus, while progesterone was high in diestrus and proestrus. 17B-Estradiol levels were higher in the striatum than in the rest of the brain; both were also shown to fluctuate during the estrous cycle and with a pattern similar to that observed in serum. Progesterone serum levels showed a similar pattern of changes during the estrous cycle to progesterone concentrations in the striatum and the rest of the brain. The ultrastructure of the striatal dendritic membranes was studied by freeze-fracture. A significant difference in the content of intramembranous particles in dendritic shafts, which are mainly contacted by dopaminergic synapses, was found during the estrous cycle. The numerical density of large (> 10 nm) intramembranous particles was increased in diestrus I and II and in the afternoon of proestrus compared to estrus, the morning of proestrus and ovariectomized rats. In contrast, the numerical density of small (< 10 nm) intramembranous particles was decreased in cycling animals compared to ovariectomized rats and fell in the afternoon of proestrus and then progressively increased in the following days to peak in the morning of proestrus. A negative correlation between steroid concentrations and small intramembranous particle density was observed, while the correlation was positive for large particles. No changes were

observed in the membranes of dendritic spines, the main postsynaptic target for cortical afferents. In summary, this is the first report that concentrations of 17/I-estradiol and progesterone in the striatum fluctuate during the estrous cycle. This is associated with estrous cycle-dependent changes of intramembranous particle density of striatal dendritic membranes. Our data therefore indicate that the striatum is a brain region hormonally modulated under physiological conditions.

Although the striatum is not a major target for sex steroids, several studies on dopamine (DA) transmission throughout the estrous cycle of the rodent have shown the modulation by gonadal steroids on the nigrostriatal DA system. Striatal DA content is decreased on the day of estrus’2*26and diestrus I (DI) in rat and mice, while homovanillic acid is increased12*26and DA turnover is highest on the day of estrus in mice.28 The activity of monoamine oxidase is lowest during the night between estrus and DI.*’ In addition, variations in the uptake of [3H]DA during the estrous cycle are observed with the highest rate

§To whom correspondence should be addressed at: Department of Molecular Endocrinology, CHUL Research Centre, Lava1 University Medical Centre, 2705. Laurier Boulevard, Sainte-Foy,~Quebec, Canada.GlV 4G2. Abbreviations: DA, dopamine; DI, diestrus I; DII, diestrus II; E,, 17)?-estradiol; PAM, proestrus a.m.; PPM, proestrus p.m.

of uptake in DI.13 Finally, Dluzen and Ramirez” observed a perturbation of the rhythmic diurnal fluctuation of spontaneous endogenous in vitro DA release during the estrous cycle of the rat in the afternoon of proestrus (PPM) and the morning of estrus. During the rat estrous cycle, striatal D-2 DA agonist binding sites show an increase of the ratio of high to low D-2 agonist binding sites in diestrus II (DII) while the D-2 antagonist binding sites remain constant.” In contrast, density of striatal D-l antagonist binding sites is higher on the day of DI and DII compared to values for the other stages of the estrous cycle.36 We have also shown that the stage of the estrous cycle at ovariectomy influences striatal D-l DA receptors for several days with a pattern similar to that observed during the estrous cycle.37 In behavioral studies, Robinson et aL4’ show sex differences and estrous cycle-dependent variations in rotational behavior elicited by electrical stimulation 893

M. MOKISSFTTE PI a/

x94 in the Similar rotational

rat

with

results

more

turning

are obtained

behavior

on

the

day

of estrus.

for amphetamine-elicited

in the rat.‘,’

study of the rat arcuate nucleus, an estrogen-sensitive area of the hypothalamus,9.22the quantitative evaluation of freeze-fracture replicas of the perikarya reveal that the density of intramembranous particles varies during the estrous cycle.25 In addition, quantitative analysis of freeze-fracture replicas from the arcuate nucleus has revealed sex differences in the density of intramembranous particles.” Therefore, to investigate the possible mechanism of action of steroid hormones in the striatum, we have evaluated the intramembranous particle density on freeze-fracture replicas of dendritic membranes from striatum during the four-day estrous cycle in relation to striatal gonadal hormone concentrations. In a freeze-fracture

EXPERIMENTAL

PROCEDURES

Adult Sprague-Dawley intact female rats were purchased from Charles River Canada Inc., St-Constant, Quebec. The rats were housed two per cage and maintained at 22-23°C on a 14: 10 light:dark cycle (lights on from 05.00 to 19.00). They received rat chow and water ad libitum. A group of rats were bilaterally ovariectomized and were used for the experiment 14 days later. For intact females, only rats demonstrating at least two consecutive four-day estrous cycles were included in the experiment. Vaginal smears were taken daily to monitor the estrous cycle. Rats were killed by decapitation; the proestrus p.m. (PPM) group in the afternoon (between 16.00 and 17.00) and the other groups in the morning (between 08.30 and 09.30). Trunk blood was collected and serum was separated by centrifugation at 4000 g for 10 min and kept at -20°C until assayed for prolactin, 17a-estradiol (EJ and progesterone. Prolactin was measured in duplicate by double-antibody radioimmunoassays using rat prolactin-I-3 and rabbit antisera (anti-rat prolactin-S-3) kindly provided by the National Hormone and Pituitary Program, Baltimore. In one half of each brain, the striatum was dissected immediately and the remainder of this half-brain was pooled; both these tissues were then frozen in dry ice and kept at - 70°C until assayed for their steroid levels. Striaturn and the remainder of the brain were prepared for the steroid assay as previously described.4’ The frozen tissues were homogenized in 5 ml ethanol:acetone (9: 1) and extracted overnight at 4‘C. After centrifugation at 2OOOg for IOmin at 4”C, the pellet was washed with 5 ml ethanol : acetone (9 : I), recentrifuged and the combined supematants evaporated under nitrogen. The residue was then resuspended in 1 ml of distilled water and extracted twice with 5 ml ether: acetone (9 : 1). The organic phase was dried under a stream of nitrogen. E, and progesterone were determined by specific radioimmunoassays after chromatography on Sephadex LH-20 columns and the levels of steroids were corrected for procedure losses as previously described.6.7 From the other half of each brain, the striatum was dissected immediately, cut into small pieces of l-2 mm, and immersed for 2 h in 1% glutaraldehyde and 1% paraformaldehyde in 0.1 M phosphate buffer, pH 7.4 at room temperature. The pieces-of striatum were removed and then washed in cold 0.1 M ohosnhate buffer. DH 7.4. This procedure of fixation was used to compare possible modifications of intramembranous particle content of striatal

dendritic membranes with variations in hormonal levels m the striatum in the same animals. Therefore. it was no1 possible to fix the tissue in these animals by perfusion. To determine whether the procedure of fixation used may affect intramembrane protein particle counts, a second group 01 rats either ovariectomized or monitored for the determnation of the stage of the estrous cycle, were perfused through the left cardiac ventricle, under ether anesthesia, first for 5 min with 0.9% NaCl and then for 20 min with 1% glutaraldehyde and 1% paraformaldehyde in 0.1 M phosphate buffer, pH 7.4, at room temperature. A ttssue block containing the striatum was removed and further fixed in the same solution for 2 h at room temperature and then washed in cold 0.1 M phosphate buffer, pH 7.4. Serial sections. 300 nm thick. were obtained by use of an Oxford Vibratome. The striatum was punched out, at -3.X mm from bregma, using hollow needles with an inner diameter of 300fim. The quality of fixation was examined by conventional electron microscopy of thin sections and was found to be satisfactory in both perfusionand immersion-fixed tissues. For freeze-fracture, tissue samples were soaked for 2 1~in 0.1 M phosphate buffer, pH 7.4. containing 20% glycerol. They were then frozen in Freon 22, cooled with liquid nitrogen, fractured at - 110°C and replicated by platinum/carbon evaporation” in a Balzers 400D apparatus (Balzers, Liechtenstein). Three to six replicas were prepared from each animal. Neuronal membranes in freeze-fracture replicas were identified by their contiguity to cross-fractured cytoplasmic profiles. Profiles of neuropil predominate in most of the surface of the replicas, while fracture of membranes of neuronal perikarya were too scarce to allow an adequate quantitative analysis. Therefore, we restricted our analysis to membranes of dendritic shafts and dendritic spines. To avoid the possible bias introduced by the observer, no attempt at selection was made when the replicas were photographed. Therefore, for a given replica, all the encountered replicated inner and outer membrane fracture faces of dendritic shafts and dendritic spines were photographed without knowledge of their experimental group. Photographic prints were also evaluated without knowledge of the experimental group from which the picture was taken. When photographic prints were analysed, no neuronal membrane was rejected on criteria other than unsatisfactory shadowing. The number of intramembranous particles on dendritic spines was recorded within a test square grid superimposed on photographic prints (final magnification x 165,000). To avoid counting errors due to the variable radius of curvature of the denritic membrane, the calculation of the correct size of the test square grid was carried out as described previously.*’ The size (diameter) of intramembranous particles, measured with a calibrated eyepiece, was considered to be the length of the base of the triangular shadow projected with the particle. Only flat regions of the membrane were selected for quantitative analysis since the angle of platinum shadowing with respect to the variable curvature of fracture faces may induce changes in the apparent size of the particles. In a similar experiment, a second group of 42 femalc rats either ovariectomized or monitored for the determination of the stage of the estrous cycle were used to determine serum and striatal concentrations of E, in order to confirm the EZ results obtained in the first experiment. Rats were killed by decapitation; the PPM group in the afternoon (between 16.00 and 17.00) and the other groups in the morning (between 08.30 and 09.30). Trunk blood was collected and serum was separated by centrifugation at 4000g for 10 min and kept at -20°C until assayed for El. Serum steroid levels were assayed in individual animals. The striata were dissected immediately, and both left and right striatum of 2- 3 rats were pooled together for the Ez assay. The

895

Estrous cycle and striatal membrane morphology Table 1. Estradiol concentration in serum and brain of rats during the estrous cycle and after ovariectomy 17/3-Estradiol concentration (nM) Group ~a~ectorni~ Ektrus Diestrus I Diestrus II Proestrus a.m. Proestrus p.m.

Striatum

Serum 0.032 f 0.030 * 0.010 * 0.020 f 0.027 f 0.196 It

0.021 0.015 0.008 0.008 0.014 O.OSO**

1.1% * 1.716 f 1.414 f 1.291 f 1.147 + 4.595 +

0.202 0.474 0.139 0.170 0.134 2.199*

Brain 0.101 f 0.010 0.144 if 0.030 0.166 f0.017 0.168 Ifi 0.024 0.218 IO.038 0.719 f 0.145**

In one half of each brain, the striatum was dissected and e&radio1 concentrations in striatum and the remainder of the brain were determined; these were defined as striatum and brain in the table. Results shown are the mean f S.E.M. from 5-13 rats analysed individually. *P < 0.05, **P < 0.01 vs estrus day values. steroids were measured in duplicate by specific radioimm~o~ays as previously described in the texth7 Statistical evaluation of hormone and in~~embr~ous particle data were performed by analysis of variance followed by pahwise comparisons with the Duncan-Rramer Multiple range test.% Correlation analysis between steroid levels and intramembranous particle density were performed. To protect against inflation of the type I error rate associated with multiple testing for significant correlations, probability levels for individuals test of significance. were adjusted for the total number of tests by the Bonferroni method.43 RESULTS

brain steroid coneentr~ti#ns during the

prolactin levels were as previously observed.‘* To confirm the striatal results, this experiment was repeated in another group of rats where pools of left and right striata from 2-3 rats were assayed. Serum and striatum E, concentrations were measured during the estrous cycle and in ovariectomized rats as shown in Table 4 and were essentially the same as those for the first experiment reported in Table 1. Results in Tables 1 and 2 as well as Table 4 show a positive correlation between steroid levels (E, or progesterone) in the serum, compared to striatum or brain concentrations (data not shown).

estrous cycle

Stratus cycle

Et concentrations in serum, striatum and the rest of the rat brain during the estrous cycie and after ovariectomy are shown in Table 1. E, levels peak in the afternoon of proestrus in serum, striatum and the rest of the brain. In contrast, progesterone concentrations were more phasic with one peak on the day of DI and DII and one on PPM in serum, striatum and the rest of the brain (Table 2). The surge of serum prolactin was seen only during the afternoon of proestrus (Table 3). Serum E,, progesterone and

The morphology of strlatum in freeze-fracture replicas was similar to that observed on thin sections. Areas where myelinated fibers predominated were excluded from the analysis. Dendritic shafts and dendritic spines were frequently observed in freezefracture replicas from the neuropil. These images provide large amounts of membrane fracture faces to be analysed (Fig. 1). Examples of freeze-fracture replicas from the striaturn are shown in Fig 1. Quantitative assessment of

Sewn

and

~~rphoi#gy and steroids during the estrow

Table 2. Progesterone ~n~nt~tion in serum and brain of rats during the estrous cycle and after ovariectomy Progesterone concentration (nM) Group Ovariectomized Estrus Diestrns I Dies&us II Proestrus a.m. Proestrus p.m.

Serum 1.08 f 22.83 f 50.23 a 55.54 k 4.80 + 50.04 f

0.38* 5.47 2.83** 10.30” 1.30. 5.47*

Striatum 0.50 + 5.90 f 33.19 + 52.26 + 5.02 * 61.18 +

Brain

0.14 0.03 * 0.01 3.06 4.51 It: 1.11 10.68* 16.17 f 4.22** 13.51** 25.72 jI 3.62** 1.60 7.25 If: 2.17 15.17+* 28.30 f 3.01**

In one half of each brain, the striatum was dissected and progesterone concentrations in striatum and the remainder of the brain were determined, these were defined as strati and brain in the table. Results shown are the means f S.E.M. from 5-13 rats analysed individually. lP < 0.05, l*P > 0.01 vs estrus day values.

M. MORISS~:T-I-I~1rd.

896

Table 3. Serum prolactin concentration of rats during the estrous cycle and after ovariectomy

Prolactin concentration (rig/ml) Serum

Group Ovariectomized Es&us Diestrus 1 Diestrus 11 Proestrus a.m. Proesttus p.m.

1.16 ) 0.53 5.20+2.12 3.27f 2.07 0.92 + 0.60 1.25+ 0.52 140.23 + I f5.97**

Results shown are the mean _) S.E.M. from S-13 rats analysed individualIy. **P < 0.01 vs es&us day values.

intramembranous particle numerical density in dendritic shafts and dendritic spines did not reveal any significant differences (P > 0.2) among the tissues fixed by immersion and the tissues fixed by perfusion. Differences in intramembranous particle numerical density were found during the estrous cycle (Table 5). In the inner membrane face of dendritic shafts, the numerical density of small (< 10 nm) intramembranous particles fell in PPM and then progressively increased in the following days to peak in the morning of proestrus (PAM). The variation in the numerical density of large intramembranous particles (> 10 nm) showed a different pattern: an increase from PAM to PPM was followed by a decrease from PPM to estrus, increasing again in Dl and DII. Similar changes were observed in the outer membrane face of dendritic shafts. In contrast, none of the above-mentioned changes were observed in the inner or outer membrane fracture faces of dendritic spine membranes. In ovariectomized rats, the number of small intramembranous particles was significantly increased compared to the values of cycling animals, whereas the number of large intramembranous particles was fow, showing values similar to those found in PAM and estrus rats. No effect of ovariectomy was detected in the number of intramembranous particles in the membranes of dendritic spines. Correlation between striatal intramembranous particle density and steroid concentrations

Since the same pattern of changes (and high positive correlation) was observed for the variations of intramembranous particle density when measured in the inner or outer membrane faces for small as well as large particles, we used the total ~ntramembranous

Table 4. Serum and striatum 17~~-es~r~i~li(~l conc‘t’nJ~atmrr< ii, a group of 42 rats either ovariectnmized or drtrmp the estrous crclc 170.estradiol concentration (inn) Group (No. of rats) Ovariectomizcd (8) Estrus (8) Diestrus I (7) Diestrus II (7) Proestrus a.m. (4) Proestrus p.m. (8)

Serum

Strialum

0.025 IO.004 0.03 I & 0.006 0.018 i 0.002 0.056 * 0.013* 0.039 & 0.009 0.086 + O.OiI **

I.lli i_O.lZO I.220 + 0.207 I.363rl‘ 0.270 0.958 i 0. t 93 0.X86 .I: 0.008 2.I?:! I_0.234*

Both left and right striatum of 2 3 rats were pooled together for the striatum assay while serum steroid levels were assayed in individual animals. Serum and striatum estradial levels were significantly correlated iR -::0.X6: P = 0.0016; d.f. = 9). *P < 0.05, **P < 0.01 vs estrus day values.

particle density (inner plus outer membrane faces) for a multiple correlation analysis with steroid levels. Table 6 shows a general negative correlation between steroid concentrations and small intramembranous particle density, while a positive correlation was observed for large particles. The correlation between progesterone levels (in the striatum, brain or serum) with ~ntramembranous particle density (small or large) was highly significant. Brain E, concentrations correlated significantly with striatal large intramembranous particle density. DlSCUSSlON

Serum and brain steroid concentrations

This is the first report that striatal Ievels of EZ and progesterone fluctuate during the rat estrous cycle; a similar variation of these steroids was also observed for the rest of the brain. Interestingly, the rise in brain steroid levels occurred at the same time as the steroid peak serum levels during the estrous cycle. In addition, E, and progesterone concentrations measured in the striatum were higher than those observed for the rest of the brain, thus suggesting a certain specificity in the distribution of these steroids in the brain. In a recent study, we reported the comparison between plasma and whole brain steroid concentrations after an acute injection of E, or progesterone to ovariectomized female rats.4’ These experiments showed that plasma and brain E, or progesterone

Fig. I. Examples of freeze-fracture replicas from the striatum of female rats. (a) Panoramic view from the striatum of an ovariectomized rat showing the typical appearance of a dendritic shaft (D) giving rise to a dendritic spine (S). The fracture face exposed for the dendrite and the denritic spine is the inner membrane face. In the lower part of the figure, the plane of fracture has broken into the cytoplasm (C) of the dendrite. Some axonal profiles (A) are also observed in this field. Magnification is x 35,000. (b) High magnification of the striatum from a rat in the afternoon of proestrus. This replica shows the inner membrane faces of a dendrite (D) and a dendritic spine (S). Large (110 nm, long arrow) and small (< IO nm, short arrow) intramembranous particles are observed in the membrane. Magnification is x 60.000.

Fig.

1

M. MORISSETTE ef al.

898

Table 5. Number of intramembranous particles per pm* in the neuronal membranes of the dendritic shafts and dendritic spines from the rat striatum measured by freeze-fracture during the estrous cycle and in ovariectomized rats E

ovx Dendritic shafts Inner membrane IMP i 10 nm IMP > 10 nm Outer membrane IMP < 10 nm IMP > 10 nm Dendritic spines Inner membrane IMP < 10 nm IMP> IOnm Outer membrane IMP < 10 nm IMP > 10 nm

Intramembranous DI

particles (IMP)/pm2 DII PAM

PPM

face 915 f 114** 227 & 29 220 + 24 139k28 137 & 17 474 + 56++

336 rf: 41 523 _+63**

407 + 51** 106 + 19

107 _+17 513 rf: 73**

52 + 7 73 rf:8**

119 * 12** 31+4

17i6 79 k 28**

face 171 * 22** 24 4 3

58 & 7 29 + 5

41*4 75 + 10**

face 321 + 34 90 & 9

267+20 81 +6

317+28 85 & 8

273 + 23 87 k 7

272 k 18 77 + 5

363 k 45 109 + 14

41+5 8kl

29 + 5 10*2

27 k 7 9+2

33 f 14 13,5

face 38 + 6 10*2

32 k 7 II&2

Results shown are the mean + S.E.M. from 5-13 rats analysed individually and for which serum and brain hormone levels were measured as shown in Tables 1-3. OVX, ovariectomized rat; E, estrus. **P < 0.01 vs respective E day values

concentrations in ovariectomized rats peak at approximately the same time following injection of these steroids. Moreover, brain and plasma E, or progesterone concentrations were correlated. These results and the present data suggest that measuring changes of serum or plasma levels of Ez or progesterone is a good reflection of the variations occurring in the rat brain, and more specifically, in the striatum. Striatal intramembranous particle density and hormones

Our results, showing that intramembranous particle density in the dendritic membrane of striatal neurons varied during the estrous cycle, suggest that the ultrastructure of striatal neuronal membranes is affected by ovarian secretions. This conclusion is further supported by the effect of ovariectomy, resulting in an increased density of small intramembranous particles in dendritic shafts. Striatal changes in the

number of large intramembranous particles is positively correlated with the levels of progesterone. Both progesterone levels in striatum and the density of large particles in dendritic shafts were high in DI, DII and PPM and low in estrus, PAM and ovariectomized rats. In contrast, the density of small intramembranous particles is negatively correlated with the levels of progesterone. The correlations between E2 levels and intramembranous particle density revealed a similar pattern to the progesterone correlations, that is a negative correlation for small particles and a positive correlation for large particles. This may suggest that E2 and/or progesterone favor large, to the detriment of small, intramembranous particles. The absence of intramembranous particle density changes in the head of dendritic spines during the estrous cycle and after ovariectomy is interesting, because these structures are mainly the postsynaptic targets for cortical afferents, while dendritic shafts are

Table 6. Correlation analysis between steroid (estradiol or progesterone) concentrations and striatal small (< 10 nm) and large (> 10 nm) intramembranous particles (sum of inner and outer membrane faces of fracture) on dendritic shafts Intramembranous particles density Correlation coefficient, probability Tissue (No. of observations) Progesterone-striatum (35) Progesterone-brain (36) Progesterone-serum (36) E,-striatum (31) E,-brain (29) E,-serum (34)

lOnm 0.57, 0.67, 0.60, 0.27, 0.53, 0.35,

0.0003* O.OOOl* 0.0001* 0.1323 0.0029* 0.0410

*Statistically significant if P < 0.0083 (P of 0.10/12) when probability levels are adjusted for the total number of comparisons by means of the Bonferroni method.43 To counterbalance this basically conservative approach, we chose a liberal overall probability level of P -C0.10 for accepting results as significant.

Estrous cycle and striatal membrane morphology mainly contacted by dopaminergic projections from the nigra and ventral tegmentum.Wy4g,51This may suggest that the hormonal regulation of membrane ultrastruct~e could be restricted to areas of the neuron contacted by dopaminergic terminals. The membrane changes observed may be either the result of a transsynaptic effect, consecutive to the modulation by ovarian secretions of the release of transmitters, or the result of a direct effect of gonadal steroids on neuronal membranes. Arcuate nucleus compared to striutal intrarnembranous particles

Numerous studies in the arcuate nucleus of the rat, an estrogen-sensitive area of the hypothalamus,g,22 have shown the importance of E2 on neurogenesis, synaptogenesis and synaptic rem~elling.4z For instance, quantitative analysis of freeze-fracture replicas from the arcuate nucleus has revealed that a population of small-sized intramembranous particles is enriched in female neuron cell membranes when compared to males, whereas a population of large intramembranous particles is enriched in male neuronal membranes.*’ These sex differences can be abolished by pharmacological administration of E, to adult female rats.M Also, the effects of E, on neuronal membrane organization can be very rapid. For example, the density of exo-endocytotic pits in the arcuate neurons of the rat h~othal~us is increased within 1 min following perfusion with medium containing a physiological concentration of E, and the effect of E2 observed on arcuate neurons was postulated to be mediated by a direct effect of this steroid on neuronal membranes.24 The effects of E, on sexual differentiation of the rat brain, occurring during the perinatal period, can be extended to a new concept of lifelong effects of sex steroid on the central nervous system.42 This concept is supported by the observation in the arcuate nucleus of the rat, that the density of intramembranous particles and synapses on perikarya of neurons varies during the ovarian cycle while mature male rats maintain the same morphological characteristics.” This synaptic remodelling of these neurons in cycling female rats shows the complex dynamics of synaptology and synaptic plasticity present in this brain structure. It is interesting to observe in this study that the effect of gonadal steroids on neuronal membranes is not specific to areas responsible for the control of gonadotropin release or to areas containing sexsteroid sensitive neurons. The striatum is known to be implicated in non-sexual behaviors of rodents. The present study demonstrates that the effects of gonadal steroids on neuronal cell membranes can also at%.3 neurons in areas of the brain where no, or very few, steroid receptors are observed.45 However, the patterns of changes of large and small intramembranous particles in the striatum during the estrous cycle are different from those observed in the arcuate nucleus,*’ suggesting regionaf specificity.

Striatal intramembranous proteins

899 particles

and membrane

The general pattern of changes of the D-l receptors is more related to the changes of large intramembranous particles while the fluctuations of the D-2 subtype shows more similarities with the modulation of small intramembranous particles during the estrous cycle. During most of the estrous cycle the ratio of high to low D-2 agonist binding sites is low compared to ovariectomized rats, with the exception of the day of DII for which values such as for ovariectomized rats are more than twice as high.” By comparison, we also observed that the density of small intramembranous particles on the inner and outer membrane faces was also lower in the striatum of cycling animals compared to values of ovariectomized rats. Furthermore, as for agonist D-2 sites, small intramembranous particle density were higher only at one stage during the estrous cycle; whereas for D-2 receptors the peak is on the day of DII, the in~amembranous particle density increase occurred a day after, namely in PAM. In general, the density (B-Max) of striatal D-l receptors is higher throughout the estrous cycle compared to the values of ovariectomized rats.36 By comparison, we also observed that the density of large intramembranous particles was generally higher in the striatum of cycling animals compared to that of ovariectomized rats. More specifically, the density of D-l receptors and large intramembranous particles on the inner and outer membrane faces shows a similar pattern of fluctuations during the estrous cycle and in ovariectomized rats, except in PPM where the density of D-l receptors and large intramembranous particles display changes in opposite directions. Since intramembranous particles represent, at least in part, membrane proteins,23*38it is possible that these proteins may be intramembranous components responsible for the coupling of dopa~nergic receptors to a second messenger transduction system via an intramembranous effector or an intramembranous protein implicated in the modification of membrane fluidity, thus regulating the activity of these receptors. It is difficult to associate changes in intramembranous particle density in the striatum with another neurotransmitter system in this structure. A literature search revealed very little data on the modulation of membrane receptors in striatum during the estrous cycle for neurotransmitters other than DA, except for serotonin. Hence, Biegon et al? have found during the estrous cycle in the basal forebrain of the rat (including the h~othalamus, septum and preoptic area), a similar pattern of serotonin receptor changes as we have observed for striatal DA receptors with higher binding in the days of diestrus than in proestrus, whereas no effect of the cycle on serotonin binding was observed in the striatum.

900 Membrane

M. MORISSETTE (at(I/. t$ects

qf’steroids

Several studies have shown that estrogen has rapid effects on DA transmission in the striatum. For instance, after an acute intravenous injection of Ez, electrophysical studies have shown an alteration in the basal firing rate and autoreceptor sensitivity of DA neurons of substantia nigra” as well as reversal by E, of the inhibitory effects of DA applied iontophoretically onto striatal neurons within 2-6 h of E,.’ Using in vitro superfusion studies and in viuo by microdialysis, estradiol was shown to rapidly increase striatal DA release.4,’ We have shown that an injection of E, at a physiological dose rapidly increases striatal dihydroxyphenylacetic acid and homovanillic acid concentrations,‘5.4’ decreases the density of the D-2 DA receptor high-affinity agonist sitq3’ increases the density of the D-2 DA receptor low-affinity agonist site,>” and increases DA uptake siteq4” whereas DA concentrations and total density of D-2 DA receptors as well as affinity of the D-2 agonist and antagonist sites and DA uptake sites remain unchanged. These results show a rapid and short-lasting effect of E, and could be correlated with elevated plasma EZ concentrations; it is likely that this is a membrane-mediated. non-genomic effect of E,. Progesterone could also alter DA transmission in the striatum by a membrane-mediated effect as observed for E,. We have previously shown that a physiological dose of progesterone acutely increases DA and its metabolites in rat striatum.‘” Dluzen and Ramirez” observed a bimodal effect of progesterone (stimulation followed by inhibition) on in vitro DA release from the corpus striatum for ovariectomized rats primed with E,. These authors observed the stimulatory phase between 2 and 12 h after the progesterone injection and postulated it to involve a non-genomic mechanism through a direct action on nerve terminals in the striatum.18

These results and the present data suggest that there may be a membrane receptor, independent of the classical genomic steroid receptor. for estrogen and progesterone responsible for the effects observed. Indeed, membrane receptors for estrogen and progesterone have been reported in the brain.‘” In contrast, Roy et al.48 have recently shown that E, can label striatal membrane proteins analysed by gel electrophoresis, but no saturable membrane binding sites were observed for Ez. suggesting the absence or. alternatively, the presence of a different receptor for E, in rat striatal membranes. The presence of membrane receptors for sex steroid hormones is. however, not necessarily required for rapid membrane effects of these hormones. For example, sexual steroids may interact with membrane phospholipids producing alterations in membrane fluidity.‘4,” Thus, the effects of gonadal steroids on neuronal membranes could generate alteration in membrane transport.3’~“~” ion conductance enhancement of endocytosis.- ‘1,)1.3:.13 changes producing rapid changes in neuronal excitability,2Y and production of rapid effects on neuronal membrane ultrastructure.‘4.4”

CONCLUSIONS

In summary, our results show for the first time that striatal E, and progesterone concentrations fluctuate during the estrous cycle and that this is associated with changes of intramembranous particle density of striatal dendritic membranes. Changes in the levels of E, and progesterone in the striatum during the estrous cycle suggest that the differences in intramembranous particle density found were probably a membrane-linked effect of steroid hormones. Acknowledgements-This

work was supported by a MRC grant (T.D.P.) and DGICYT (Spain) grant No. PM89-0004. We thank J. Sancho and E. Valero for their technical assistance. M. M. is holder of a studentship from the Fonds de la Recherche en Sante du Quebec.

REFERENCES

1. Arnauld E., Dufy B., Pestre M. and Vincent J. D. (1981) Effects of estrogens on the response of caudate neurons to microiontophoretically applied dopamine. Neurosci. Lett. 21, 325-33 1.

2. Becker J. B. and Ramirez V. D. (1981) Sex differences in the amphetamine-stimulated release of catecholamines from rat striatal tissues in vitro., Bruin Res. 2Q4, 361-372. 3. Becker J. B., Robinson T. E. and Lorenz K. A. (1982) Sex differences and estrous cycle variations in amphetamineelicited rotational behavior. Eur. J. Pharmacol. 80, 65-72. 4. Becker J. B. (1990) Estrogen rapidly potentiates amphetamine-induced striatal dopamine release and rotational behavior during microdialysis. Neurosci. Lett. 118, 169-171. 5. Becker J. B. (19%) Direct effect of Illbeta-estradiol on striatum: sex differences in dopamine release. Synapse $157-164. 6. B&anger A., Cusan L., Caron S., Barden N. and DuPont A. (1981) Ovarian progestins, androgens and estrogen throughout the 4-day estrous cycle in the rat. Biol. Reprod. 24, 591-596. I. Belanger A., Caron S. and Picard V. (1980) Simultaneous radioimmunoassay of progestins, androgens and estrogens in rat testis. J. Steroid Biochem. 13, 185190. 8. Biegon A., Bercovitz H. and Samuel D. (1980) Serotonin receptor concentration during the estrous cycle of the rat. Brain Res. 187, 221-225.

9. Brawer J. R. (1971) The role of arcuate nucleus in the brain-pituitary-gonad axis. J. corny. Neural. 143, 411-446. 10. Butcher R. L., Collins W. E. and Fugo N. W. (1974) Plasma concentration of LH, FSH, prolactin, progesterone and estradiol-178 throughout the 4-day estrous cycle of the rat. Endocrinology 94, 1704-l 708. I 1. Chiodo L. A. and Caggiula A. R. (1983) Substantia nigra dopamine neurons: alterations in basal discharge rates and autoreceptor sensitivity induced by estrogen. Neuropharmacology 22, 593-599

Estrous cycle and striatal membrane morphology

901

12. Crowley W. R., G’Donohue T. L. and Jacobowitz D. M. (1978) Changes in catecholamine content in discrete brain nuclei during the estrous cycle of the rat. Bruin Res. 147, 315-336. 13. Davis C. F., Davis B. F. and Halaris A. E. (1977) Variations in the uptake of [%Jdopamine during the estrous cycle. Life Sci. 20, 13191332. 14. Deliconstantinos G. and Fotiou S. (1986) Sex steroid and prostaglandin interactions upon the purified rat myometrial plasma membranes. Molec. Cell. Endocrinol. 45, 149-156. 15. Di Paolo T., Rouillard C. and B&lard P. J. (1985) 17&estradiol at a physiological dose acutely increases dopamine turnover in rat brain. Eur. J. Pharmacol. 117, 197-203. 16. Di Paolo T., L&sque D. and Daigle M. (1986) A physiolo~~al dose of progesterone affects rat striatum biog~ic amine metabolism. Eur. J. Ph~~~oZ. 125, 11-16. 17. Di Pa010 T., Falardeau P. and Morissette M. (1988) Striatal D-2 dopamine agonist binding sites fluctuate during the rat estrous cycle. Life Sci. 43, 665-672. 18. Dluzen D. and Ramirez V. D. (1984) Bimodal effect of progesterone on in vitro dopamine function of the rat corpus striatum. Neuroendocrinology 39, 149-155. 19. Dluzen D. and Ramirez V. D. (1985) In uitro donamine release from the rat striatum: diurnal rhvthm and its modification by the estrous cycle.‘Neu&endocrinology 41, 97-100. 20 Garcia-Segura L. M. and Perrelet A. (1984) Postsynaptic membrane domains in the molecular layer of the cerebellum: a correlation between presynaptic inputs and postsynaptic plasma membrane organization. Brain Res. 321, 255-266. 21. Garcia-Segura L. M., Baetens D. and Naftolin F. (1985) Sex differences and maturational changes in arcuate nucleus neuronal plasma membrane organization. Deal Brain Rex 19, 146149. 22. Garcia-Segura L. M., Baetens D. and Naftolin F. (1986) Synaptic remodelling in arcuate nucleus after injection of estradiol vaierate in adult female rats. Brain Res. 366, 131-136. 23. Garcia-Segura L. M., De Cozar M., Moreno M. C. and Monreal J. (1986) Freeze-fracture characterization of proteolipid protein and basic protein of central nervous system myelin incorporated to liposomes. Bruin 1pes. 3&I, 261-266. 24. Garcia-Segura L. M., Olmos G., Tranque P. and Naftolin F. (1987) Rapid effects of gonadal steroids upon hypothalamic neuronal membrane ultrastructure. J. Steroid Biochem. 27, 615623. 25. Garcia-Segura L. M., Hernandez P., Olmos G., Tranque P. A. and Naftolin F. (1988) Neuronal membrane remodelling during the oestrous cycle: a freeze fracture study in the arcuate nucleus of the rat hypothalamus. J. Neurocytol. 17, 377-383. 26. Greengrass P. M. and Tonge S. R. (1971) Changes in brain monoamine concentrations during the oestrous cycle in the mouse: possible pharmacological implications. J. Pharm. Pharmac. 23, 897-898. 27. Holzbauer M. and Youdim M. B. H., (1973) The oestrous cycle and monoamine oxidase activity. Br. J. Pharmac. 48, 600-608.

28. Jori A., Colturani F., Dolfini E. and Rutczynski M. (1976) Modification of the striatal dopamine metabolism during the estrous cycle in mice. Neuroendocrinology 21, 262-266. 29. Kelly M. J., Moss R. L., Dudley C. A. and Fawcett C. P. (1977) The specificity of the response of preoptic-septal area neurons to estrogen: 17-or-estradiol vs 17-#-estradiol and the response of extrah~othalamic neurons. ExpI Brain Res. 30,43-52.

30. Kita H. and Kitai S. T. (1988) Glutamate decarboxylase immunoreactive neurons in rat neost~atum: their mo~holo~cal types and ~pulations. Brain Res. 447, 346-352. 31. Koenig H., Goldstone A. and Lu C. Y. (1982) Testosterone induces a rapid stimulation of endocytosis, amino acid and hexose transport in mouse kidney cortex. Biochem. biophys. Res. Comm. 106, 346-353. 32. Koenig H., Goldstone A. and Lu C. Y. (1983) Androgenic stimulation of endocytosis, amino acid and hexose transport in mouse kidney cortex involves increased calcium fluxes. Biochim. biophys. Acta 762, 366-371. 33. Koenig H., Goldstone A. and Lu C. Y. (1983) Polyamines regulate calcium fluxes in a rapid plasma membr%ne response. Nature 305, 530-534.

34. Kramer C. Y. (1956) Extension of multiple range tests for group means with unequal number of replications. Biometrics 12, 307-310.

35. Levesque D. and Di Paolo T. (1988) Rapid conversion of high into low striatal D-2 dopamine receptor agonist binding states after an acute physiological dose of 17fl-estradiol. Neurosci. Lett. 88, 113-l 18. 36. L&esque D., Gagnon S. and Di Paolo T. (1989) Striatal Dl dopamine receptor density fluctuates during the rat estrous cycle. Neurosei. Left. Q8, 345-350. 37. L&esque D. and Di Paolo T. (1990) Effect of the rat estrous cycle at ovariectomy on striatal D-l dopamine receptors. Brain Res. Bufl. 24, 281-284.

38. Marchesi V. T., Tillack T. W., Jackson R. L., Segrest J. P. and Scott R. E, (1972) Chemical chara~te~zation and surface orientation of the major glycoprotein of the human erythrocyte membrane. Proc. natn. Acad. 5%. U.S.A. 69, 1445-1449. 39. Moor H. and Miihlethaler K. (1963) Fine structure of frozen-etched yeast cells. J. Cell Biol. 17, 609-628. 40. Morissette M., Biron D. and Di Paolo T. (1990) Effect of estradiol and progesterone on rat striatal dopamine uptake sites. Brain Res. Bull. 25, 419-422. 41. Morissette M., L-evesque D., Belanger A. and Di Paolo T. (1990) A physiological dose of estradiol with progesterone affects striatum biogenic amines. Can. J. Physiol. Pharmacol. 68, 152&1526. 42. Naftolin F., Garcia-Segura L. M., Keefe D., Leranth C., Maclusky N. J. and Brawer J. R. (1990) Estrogen effects on the synaptology and neural membranes of the rat hypothalamic arcuate nucleus. Biol. Reprod. 42, 21-28. 43. Neter J. and Wasserman W. (1974) In Applied Linear Statistical Models (ed. Irwin R. D.), pp. 146-149. R. D. Irwin Inc., Illinois. 44. Ohnos G., Aguilera P., Tranque P., Naftolin F. and Garcia-Segura L. M. (1987) Estrogen-induced synaptic remodelling in adult rat brain is accompanied by the reorganization of neuronal membranes. 3rain Res. 425, 5764. 45. Pelletier G., Liao N., Follea N. and Govindan M. V. (1988) Mapping of estrogen rotor-prying cells in the rat brain by in situ hyb~di~tion. Neurosci. ht. !M, 23-28. 46. Rambo C. 0, and Szego C. M. (1983) Estrogen action at endometrial membranes: alterations in luminal surface detectable within seconds. J. Cell Bioi. 97, 679-685.

902

M. MOKISSETTE 6’1ul.

47. Robinson T. E., Camp D. M., Jacknow D. S. and Becker J. 8. (1982) Sex differences and estrous cycle dependent variation in rotational behavior elicited by electrical stimulation of the mesostriatal dopamine system. Bihar. Bwin Rev. 6, 273-287. 48. Roy E. J., Buyer D. R. and Licari V. A. (1990) Estradiol in the striatum: effects on behavior and dopamine receptut s but no evidence for membrane steroid receptors. Brain Res. Bull. 25, 221-227. 49. Somogyi P., Bolam J. F. and Smith A. D. (1981) Monosynaptic cortical input and local axon collaterals of identi~ed striatonigral neurons. A light and electron miocroscopic study using the Golgi-peroxidase transport-degeneratjon procedure. J. camp. Neural. 195, 567--584. 50. Towle A. and Sze P. Y. (1983) Steroid binding to synaptic plasma membrane: differential binding of glucocorticoids and gonadal steroids. Steroid B’iochem. 18, 135.-143. 51. Wilson C. and Croues P. (1980) Fine structure and synaptic connections of the common spiny neurons of the rat neostriatum: a study employing intracellular injections of horseradish peroxidase. J: camp. Nwrol. 194, 599-615. (Accepted

11 February 1992)