- Email: [email protected]
Estrogen-induced suppression of female rat forebrain neurons with axons to ventral midbrain
Neuroscience Letters, 119 (1990) 171-174
171
Elsevier Scientific Publishers Ireland Ltd. NSL 07278
Estrogen-induced suppression of female rat forebrain neurons with axons to ventral midbrain Takeshi Hasegawa and Yasuo Sakuma Department of Physiology L Hirosaki University School of Medicine, Hirosaki (Japan) (Received 4 June 1990; Revised version received 18 July 1990; Accepted 18 July 1990)
Key words: Antidromic activation; Axonal excitability; Preoptic area; Sex steroid hormone; Ventral tegmental area In estrogen-treated and non-treated ovariectomized female rats under urethan anesthesia, 278 neurons were antidromically identified in the medial preoptic and other basal forebrain areas by electrical stimulation of their terminals in the ventral tegmental area. Responses from the estrogen-treated rats had a significantly higher threshold, longer latency and absolute refractory period than those from the non-treated rats. Estrogen may decrease excitability of the preoptic axons along its length.
Estrogen (E) implants in the medial preoptic area (POA) of female rats facilitate reproductive behavior and wheel running activity, presumably via E-sensitive neurons in this region [14]. Anatomical studies showed dense projections of POA cells to the ventral midbrain, specifically to the ventral tegmental area (VTA) [5]. Thus, the expression of certain E-dependent functions may be realized through this system. However, systematic evaluation of E effects on POA efferents has yet to be pursued. Here we report E-induced changes in parameters of VTA-induced antidromic activation of POA neurons in the ovariectomized (OVX) female rats. Thirty-eight OVX Wistar female rats were used. Twenty were used without further treatment; 18 had a 5 mm Silastic capsule (inner diameter, 0.059 inch; outer diameter, 0.077 inch) containing 1.5-2.0 mg 17fl-estradiol (Sigma), implanted under the subaxillary skin 5-8 days before the experiment. The treatment stabilizes the serum E at the intact proestrous level [11]. Before they were anesthetized by urethan (1.0 g/kg b.wt., i.p. in a 0.25 g/ml solution), all rats with the capsule showed lordosis reflex in response to male mounts, while none of the untreated rats did. They were placed in a stereotaxic frame, with the incisor bar put 5 mm above the center of the ear bars. The parietal cortex was exposed and covered with warm agar in saline (2 %). Rectal temperature was maintained between 37 and 38°C. A coaxial bipolar electrode, constructed from 30 gauge hypodermic tubing and 120-/tm-thick stainless-steel inner wire with Correspondence: Y. Sakuma, Department of Physiology I, Hirosaki University School of Medicine, 5 Zaifucho, Hirosaki 036, Japan. 0304-3940/90/$ 03.50 © 1990 Elsevier Scientific Publishers Ireland Ltd.
a DC resistance of approximately 60 kf2, was inserted into the VTA. The coordinates were: 4.2 mm posterior to the bregrna, 0.7 mm lateral to the midline, and 7.5 mm below the dural surface. The basal forebrain was systematically explored while 500 as negative rectangular pulses were applied to the VTA at 1.0 Hz. Current intensity was monitored as a voltage across a calibrated 1 kI2 resistor. The intensity did not exceed 1700 gA, which was well below the threshold for the induction of body movement. Recordings of extracellular potentials were made with glass micropipettes constructed from Pyrex tubes of 1.5 mm outer diameter and were filled with 0.5 M sodium acetate solution, which contained 2% Pontamine sky blue 6B. The DC resistance of the pipette was no more than 20 MI2. Extracellular potentials were amplified and displayed by conventional methods. Potentials which remained smaller than 1.0 mV in the peak-to-peak amplitude were discarded. Microelectrode tracks were in the transverse planes approximately 2.0 mm anterior to the bregma. The latency, threshold and absolute refractory period were determined. The threshold was the minimal current at which 50 % following of the response was achieved [2]. The responses with a threshold exceeding 2.0 mA were discarded. The absolute refractory period was the maximal interval between the stimulus pair, at which the second fraction at 3 times the threshold consistently failed to elicit a response. Six to eight electrode penetrations, with a separation of 100 or 200 am, were usually made. Following completion of each penetration, cathodal current of 3/zA was passed for 15 min via the micropipette at two points in each track, at the deepest and
172
the shallowest points with antidromic responses. At the termination of experiment, the rat was deeply anesthetized and perfused through the heart with 10 % formalin. A frozen serial section of the brain (100/zm) was made and stained with Cresyl violet. A total of 278 neurons were antidromically activated from the VTA. Each had a discrete threshold and stable latency. The waveform of the antidromic action potentials was biphasic, with the initial deflection positive to the animal ground. In 256 (92.1%) potentials, a notch was present in their leading phase, which presumably corresponds to the slowed propagation of excitation from the initial segment to the somatodendritic (SD) membrane. The failure of antidromic propagation into the SD complex was frequently observed when a single stimulus at 1.0 Hz repetition was used (Table I). When the interval between a paired set of stimulation pulses was shortened into the absolute refractory period, the SD response to the second pulse almost always delayed or failed. Spontaneously generated action potentials canceled the antidromic potentials. The absence of spontaneous activity in 232 cells (83.5 %) precluded the possibility of observing the collision. The antidromic nature of these cells depended on stable latency and solid ability to follow paired stimuli at short intervals (3-5 ms). In response to gradual changes in the stimulus intensity, the latency jumps occurred in 53/103 (51.5%) in the nontreated, and in 66/128 (51.6%) in the E-treated animals. Discrete reductions in the latency jumps amounted to as large as 17.7 ms (mean 3.2 ms). The 278 recordings were made up of 154 from the Etreated rats and 124 from those without the treatment. They are compared in Table I. In some cells, the refractory period could not be determined due to high threshold or blockade of activation to the stimulus pair. Mann-Whitney U-test detected a significantly higher threshold, longer latency and absolute refractory period in the E-treated rats than those without treatment. The
proportion of spontaneously active cells, none of which had discharge rates exceeding 0.2 Hz regardless of the Etreatment, and the percentage of the blockade of the SD potentials, were not different between the animal groups. The number of antidromic responses encountered in each electrode penetration was 0.95+0.10 (mean + S.E.M., standard error of the mean, for n = 127 penetrations) in the non-treated and 1.22 -t- 0.13 (n = 139) in the E-treated rats; statistics do not warrant a difference. The antidromically identified cells were located in the POA, which scattered into the nucleus accumbens, the diagonal band of Broca and the bed nucleus of the anterior commissure. The distribution pattern is basically similar to that revealed by the retrograde tracer studies, and covers a wider area than that of the E-concentrating cells in the forebrain [5, 14]. Since no difference existed between the neuronal distribution in the OVX nontreated and OVX E-treated rats, data on the locations of 278 antidromically identified cells in each group are combined and shown in Fig. 1. Stimulation sites in the VTA are also given. The latency jumps are usually associated with the activation of different branches of the terminal arborization. Because fluorescent-dye tracer is taken up in the VTA and retrogradely transported to the POA, the stimulation sites in this study were most likely in a terminal field of the POA axons, which presumably project through the medial forebrain bundle [5, 14]. Perkins and Whitehead [12] activated 13.5% of POA cells antidromically from the medial forebrain bundle at the mammillary level, which continued caudally. POA stimulation, in turn, causes transsynaptic activation of VTA at short latencies [7]. Individual variations in the placement of the stimulating electrode, if any, were unlikely to precipitate systematic difference in the threshold between the animal groups. The observed lengthening effect of E on the latency contrasted to that in the paraventricular oxytoci-
TABLE I C O M P A R I S O N S B E T W E E N R E C O R D I N G S F R O M T H E C O N T R O L O V A R I E C T O M I Z E D (OVX) A N D T H E E S T R O G E N - T R E A T E D (OVX + E) F E M A L E R A T S Values are means + S.E.M. Values in the parentheses are the range and the number of observations (n). SD, somatodendritic complex (see text). Antidromic activation latency, (ms)
Antidromic activation threshold, C A )
Absolute refractory period, (ms)
Spontaneously active ceils
Failure of SD invasion
OVX
12.3 +0.7 (1.6-38.0, n = 124)
624-t- 39 (110-1700, n = 124)
1.33 + 0.05 (0.6-2.0, n = 6 3 )
21/124 (16.9%)
29/124 (23.4%)
OVX+E
14.5_+0.5"* (2.2-36.0, n = 154)
743-t-36" ( 100-I 950, n = 154)
1.48+0.4" (0.9-2.4, n = 67)
25/154(16.2%)
33/154(21.4%)
**P < 0.002, *P < 0.02, M a n n Whitney U-test.
173
A7.4
A A 7.8 A
8-2
,;'
"--_
~
\i\,
, MS,
,'.._ ,,t~ ~
L,--,.J "--").~,._,,
,
-
-------,~"
,~
'.
,
[ ,b
, ~t,, ~ i . ,
.
"J_K~.~ ~ DBB
D { ~ ~
Imm
B
A 1.8 A
2.2
...-.,
C'~/
ML
I
A 2. Mr....'?
/ ~
...
SN
.-,
Imm
Fig. 1. A: Dots represent approximate locations of neurons in the medial preoptic area (POA) and adjacent basal forebrain which displayed antidromic responses to ventral tegmental area stimulation. Tip locations of the stimulation electrodes are shown in B. AC, anterior commissure; ACB, nucleus accumbens; CP, cerebral peduncle; DBB, diagonal band of Broca; IP, interpeduncular nucleus; ML, medial lemniscus; MS, medial septum; OC, optic chiasma; R, red nucleus; SC, suprachiasmatic nucleus; SN, substantia nigra; SO, supraoptic nucleus. Numbers accompanying each panel denote distances from bregma in mm.
nergic cells [1]. Since the exact trajectory of POA axons is not known, minor bias in the stimulation sites could not be excluded to cause such a difference in the latency. Nevertheless, this may indicate that the change in the excitability occurs along certain lengths of the axon and is not limited to the axon terminals. E raised the antidromic activation threshold and prolonged the absolute refractory period. Several possible mechanisms have been proposed for E-induced changes in the neuronal excitability. Some POA neurons with axons to the ventral midbrain concentrate E [5]. The lordosis reflex in the OVX rat, which was used in the present study as a behavioral clue to the E effects, requires 48 hr to emerge following the administration [14]. The long latency may be needed for E-dependent synthesis of functional molecule through genomic activation [14]. Such possibility has been demonstrated [8]. A subset of cells in the rostral part of the ventromedial nucleus of the hypothalamus are specifically excited by E [2, 17]. Their topography and axonal projection to the midbrain differentiate them from the E-concentrating local neurons [14]. Proposed nongenomic effects of E in-
clude changes in the membrane fluidity to alter neuronal or myometrial excitability [4]. In the OVX female rat, E activates Na+,K+-ATPase in the POA, but diminishes it in the medial basal hypothalamus [15]. The electrogenic pump activity, which would affect the neuronal excitability through shifts in the resting potential, is associated with the membrane fluidity [20]. In the medial amygdala neurons in tissue slices, an increased K* conductance was accounted for E-induced hyperpolarization [9]. Analogous mechanism may be responsible for the lack of neuronal accommodation in the amygdala neurons in the slices from the E-treated rats [19]; altered turnover of amino acid receptors may also modulate neuronal excitability in response to ambient molecules such as glutamate [16]. Traditionally, E-induced inhibition of neuronal activity in the POA has been associated with a negative feedback effect of E on gonadotropin release [3]. The present group of POA cells, however, is clearly different from those involved in gonadotropin secretion, which are distributed in the anteroventral periventricular region [21]. The VTA, which is innervated by the POA and send off
174
efferents to the spinal cord and to the basal ganglia, occupy a strategic position in exercising behavioral control [14]. E initiates maternal behavior through POA, however, Numan et al. [10] suggest that the lateral hypothalamus may intervene in the descending path. The present direct POA projection to the VTA does not fit their scheme. Although a definite role of POA neurons in the regulation of E-dependent sexual receptivity in the female rat has been much debated [6], we have successfully suppressed the lordosis reflex by electrical stimulation of the POA [13] as well as the VTA [18]. The E-induced suppression of the POA would, therefore, culminate in the disinhibition of the reflex. However, there is still other behavior, such as aggression and running wheel activity [14] in which POA has been implicated as a site of E action. Therefore, until the neural circuit for each of these and other functions is better known, the role of the current subset of POA cells remains an open question. 1 Akaishi, T. and Sakuma, Y., Estrogen excites oxytocinergic, but not vasopressinergic cells in the paraventricular nucleus of female rat hypothalamus. Brain Res., 335 (1985) 302-305. 2 Akaishi, T. and Sakuma, Y., Projections of estrogen-sensitive neurones from the ventromedial hypothalamic nucleus of the female rat, J. Physiol., 372 (1986) 207-220. 3 Dyer, R.G., An electrophysiological dissection of the hypothalamic regions which regulate the pre-ovutatory secretion of luteinizing hormone in the rat, J. Physiol., 234 (1973) 421-442. 4 Erulker, S.D. and Wetzel, D.M., Steroid effects on excitable membranes, Curr. Topics Membr. Transport, 31 (1987) 141-190. 5 Fahrbach, S.E., Morrell, J.I. and Pfaff, D.W., Identification of medial preoptic neurons that concentrate estradiol and project to the midbrain in the rat, J. Comp. Neurol., 247 (1986) 364-382. 6 Kondo, Y., Shinoda, A., Yamanouchi, K. and Arai, Y., Recovery of lordotic activity by dorsal deafferentiation of the preoptic area in male and androgenized female rats, Physiol. Behav., 37 (1986) 495-498. 7 Maeda, H. and Mogenson, G.J., An electrophysiological study of inputs to neurons of the ventral tegmental area from the nucleus accumbens and medial preoptic-anterior hypothalamic areas, Brain Res., 197 (1980) 365-377. 8 Mobbs, C.V., Fink, G., Pfaff, D.W., Hip-70: A protein induced by
estrogen in the brain and LH-RH in the pituitary, Science, 247 (1990) 1477-1479. 9 Nabekura, J., Oomura, Y., Minami, T., Mizuno, Y. and Fukuda, A., Mechanism of the rapid effect of 17fl-estradiol on medial amygdala neurons, Science, 233 (1986) 226-228. 10 Numan, M., Morrell, J.I. and Pfaff, D.W., Anatomical identification of neurons in selected brain regions associated with maternal behavior deficits induced by knife cuts of the lateral hypothalamus in rats, J. Comp. Neurol., 237 (1985) 552-564. 11 Parsons, B., MacLusky, N.J., Krey, L., Pfaff, D.W. and McEwen, B.S., The temporal relationship between estrogen-inducible progestin receptors in the female rat brain and the time course of estrogen activation of mating behavior, Endocrinology 107 (1980) 774-779. 12 Perkins, M.N. and Whitehead, S.A., Responses and pharmacological properties of preoptic/anterior hypothalamic neurones following medial forebrain bundle stimulation, J. Physiol., 279 (1978) 347 360. 13 Pfaff, D.W. and Sakuma, Y., Facilitation of the lordosis reflex from the ventromedial nucleus of the hypothalamus, J. Physiol., 288 (1979) 189-202. 14 Pfaff, D.W., Cellular mechanisms of female reproductive behaviors. In E. Knobil and J. Nell (Eds.), The Physiology of Reproduction, Raven, New York, 1988, pp. 1487-1568. 15 Rodriguez del Castillo, A., Battaner, E., Guerra, M., Alonso, T. and Mas, M., Regional changes of brain Na÷,K+-transporting adenosine triphosphatase related to ovarian function, Brain Res., 416(1987) 113-118. 16 Sah, P. Hestrin, S., Nicoll, R.A., Tonic activation of NMDA receptors by ambient glutamate enhances excitability of neurons, Science, 246 (1989) 815-818. 17 Sakuma, Y. and Akaishi, T., Cell size, projection path, and localization of estrogen-sensitive neurons in the rat ventromedial hypothalamus, J. Neurophysiol., 57 (1987) 1148-1159. 18 Sakuma, Y., Akitsu, H., Hasegawa, N. and Hasegawa, T., Interruption of lordosis behavior from electrical stimulation of the female rat ventral tegmental area, Proc. Int. Union Physiol. Sci., 17 (1989) 310. 19 Schiess, M.C., Joels, M., Shinnick-GaUagher, P., Estrogen priming affects active membrane properties of medial amygdala neurons, Brain Res., 440 (1988) 380-385. 20 Szego, C.M. and Pietras, R.J., Membrane recognition and effector sites in steroid hormone action, Biochem. Act. Horm., 8 (1981) 307~,63. 21 Terasawa, E., Wiegand, S.J. and Bridson, W.E., A role for medial preoptic nucleus on afternoon proestrus in female rats, Am. J. Physiol., 238 (1980) E533-E539.
171
Elsevier Scientific Publishers Ireland Ltd. NSL 07278
Estrogen-induced suppression of female rat forebrain neurons with axons to ventral midbrain Takeshi Hasegawa and Yasuo Sakuma Department of Physiology L Hirosaki University School of Medicine, Hirosaki (Japan) (Received 4 June 1990; Revised version received 18 July 1990; Accepted 18 July 1990)
Key words: Antidromic activation; Axonal excitability; Preoptic area; Sex steroid hormone; Ventral tegmental area In estrogen-treated and non-treated ovariectomized female rats under urethan anesthesia, 278 neurons were antidromically identified in the medial preoptic and other basal forebrain areas by electrical stimulation of their terminals in the ventral tegmental area. Responses from the estrogen-treated rats had a significantly higher threshold, longer latency and absolute refractory period than those from the non-treated rats. Estrogen may decrease excitability of the preoptic axons along its length.
Estrogen (E) implants in the medial preoptic area (POA) of female rats facilitate reproductive behavior and wheel running activity, presumably via E-sensitive neurons in this region [14]. Anatomical studies showed dense projections of POA cells to the ventral midbrain, specifically to the ventral tegmental area (VTA) [5]. Thus, the expression of certain E-dependent functions may be realized through this system. However, systematic evaluation of E effects on POA efferents has yet to be pursued. Here we report E-induced changes in parameters of VTA-induced antidromic activation of POA neurons in the ovariectomized (OVX) female rats. Thirty-eight OVX Wistar female rats were used. Twenty were used without further treatment; 18 had a 5 mm Silastic capsule (inner diameter, 0.059 inch; outer diameter, 0.077 inch) containing 1.5-2.0 mg 17fl-estradiol (Sigma), implanted under the subaxillary skin 5-8 days before the experiment. The treatment stabilizes the serum E at the intact proestrous level [11]. Before they were anesthetized by urethan (1.0 g/kg b.wt., i.p. in a 0.25 g/ml solution), all rats with the capsule showed lordosis reflex in response to male mounts, while none of the untreated rats did. They were placed in a stereotaxic frame, with the incisor bar put 5 mm above the center of the ear bars. The parietal cortex was exposed and covered with warm agar in saline (2 %). Rectal temperature was maintained between 37 and 38°C. A coaxial bipolar electrode, constructed from 30 gauge hypodermic tubing and 120-/tm-thick stainless-steel inner wire with Correspondence: Y. Sakuma, Department of Physiology I, Hirosaki University School of Medicine, 5 Zaifucho, Hirosaki 036, Japan. 0304-3940/90/$ 03.50 © 1990 Elsevier Scientific Publishers Ireland Ltd.
a DC resistance of approximately 60 kf2, was inserted into the VTA. The coordinates were: 4.2 mm posterior to the bregrna, 0.7 mm lateral to the midline, and 7.5 mm below the dural surface. The basal forebrain was systematically explored while 500 as negative rectangular pulses were applied to the VTA at 1.0 Hz. Current intensity was monitored as a voltage across a calibrated 1 kI2 resistor. The intensity did not exceed 1700 gA, which was well below the threshold for the induction of body movement. Recordings of extracellular potentials were made with glass micropipettes constructed from Pyrex tubes of 1.5 mm outer diameter and were filled with 0.5 M sodium acetate solution, which contained 2% Pontamine sky blue 6B. The DC resistance of the pipette was no more than 20 MI2. Extracellular potentials were amplified and displayed by conventional methods. Potentials which remained smaller than 1.0 mV in the peak-to-peak amplitude were discarded. Microelectrode tracks were in the transverse planes approximately 2.0 mm anterior to the bregma. The latency, threshold and absolute refractory period were determined. The threshold was the minimal current at which 50 % following of the response was achieved [2]. The responses with a threshold exceeding 2.0 mA were discarded. The absolute refractory period was the maximal interval between the stimulus pair, at which the second fraction at 3 times the threshold consistently failed to elicit a response. Six to eight electrode penetrations, with a separation of 100 or 200 am, were usually made. Following completion of each penetration, cathodal current of 3/zA was passed for 15 min via the micropipette at two points in each track, at the deepest and
172
the shallowest points with antidromic responses. At the termination of experiment, the rat was deeply anesthetized and perfused through the heart with 10 % formalin. A frozen serial section of the brain (100/zm) was made and stained with Cresyl violet. A total of 278 neurons were antidromically activated from the VTA. Each had a discrete threshold and stable latency. The waveform of the antidromic action potentials was biphasic, with the initial deflection positive to the animal ground. In 256 (92.1%) potentials, a notch was present in their leading phase, which presumably corresponds to the slowed propagation of excitation from the initial segment to the somatodendritic (SD) membrane. The failure of antidromic propagation into the SD complex was frequently observed when a single stimulus at 1.0 Hz repetition was used (Table I). When the interval between a paired set of stimulation pulses was shortened into the absolute refractory period, the SD response to the second pulse almost always delayed or failed. Spontaneously generated action potentials canceled the antidromic potentials. The absence of spontaneous activity in 232 cells (83.5 %) precluded the possibility of observing the collision. The antidromic nature of these cells depended on stable latency and solid ability to follow paired stimuli at short intervals (3-5 ms). In response to gradual changes in the stimulus intensity, the latency jumps occurred in 53/103 (51.5%) in the nontreated, and in 66/128 (51.6%) in the E-treated animals. Discrete reductions in the latency jumps amounted to as large as 17.7 ms (mean 3.2 ms). The 278 recordings were made up of 154 from the Etreated rats and 124 from those without the treatment. They are compared in Table I. In some cells, the refractory period could not be determined due to high threshold or blockade of activation to the stimulus pair. Mann-Whitney U-test detected a significantly higher threshold, longer latency and absolute refractory period in the E-treated rats than those without treatment. The
proportion of spontaneously active cells, none of which had discharge rates exceeding 0.2 Hz regardless of the Etreatment, and the percentage of the blockade of the SD potentials, were not different between the animal groups. The number of antidromic responses encountered in each electrode penetration was 0.95+0.10 (mean + S.E.M., standard error of the mean, for n = 127 penetrations) in the non-treated and 1.22 -t- 0.13 (n = 139) in the E-treated rats; statistics do not warrant a difference. The antidromically identified cells were located in the POA, which scattered into the nucleus accumbens, the diagonal band of Broca and the bed nucleus of the anterior commissure. The distribution pattern is basically similar to that revealed by the retrograde tracer studies, and covers a wider area than that of the E-concentrating cells in the forebrain [5, 14]. Since no difference existed between the neuronal distribution in the OVX nontreated and OVX E-treated rats, data on the locations of 278 antidromically identified cells in each group are combined and shown in Fig. 1. Stimulation sites in the VTA are also given. The latency jumps are usually associated with the activation of different branches of the terminal arborization. Because fluorescent-dye tracer is taken up in the VTA and retrogradely transported to the POA, the stimulation sites in this study were most likely in a terminal field of the POA axons, which presumably project through the medial forebrain bundle [5, 14]. Perkins and Whitehead [12] activated 13.5% of POA cells antidromically from the medial forebrain bundle at the mammillary level, which continued caudally. POA stimulation, in turn, causes transsynaptic activation of VTA at short latencies [7]. Individual variations in the placement of the stimulating electrode, if any, were unlikely to precipitate systematic difference in the threshold between the animal groups. The observed lengthening effect of E on the latency contrasted to that in the paraventricular oxytoci-
TABLE I C O M P A R I S O N S B E T W E E N R E C O R D I N G S F R O M T H E C O N T R O L O V A R I E C T O M I Z E D (OVX) A N D T H E E S T R O G E N - T R E A T E D (OVX + E) F E M A L E R A T S Values are means + S.E.M. Values in the parentheses are the range and the number of observations (n). SD, somatodendritic complex (see text). Antidromic activation latency, (ms)
Antidromic activation threshold, C A )
Absolute refractory period, (ms)
Spontaneously active ceils
Failure of SD invasion
OVX
12.3 +0.7 (1.6-38.0, n = 124)
624-t- 39 (110-1700, n = 124)
1.33 + 0.05 (0.6-2.0, n = 6 3 )
21/124 (16.9%)
29/124 (23.4%)
OVX+E
14.5_+0.5"* (2.2-36.0, n = 154)
743-t-36" ( 100-I 950, n = 154)
1.48+0.4" (0.9-2.4, n = 67)
25/154(16.2%)
33/154(21.4%)
**P < 0.002, *P < 0.02, M a n n Whitney U-test.
173
A7.4
A A 7.8 A
8-2
,;'
"--_
~
\i\,
, MS,
,'.._ ,,t~ ~
L,--,.J "--").~,._,,
,
-
-------,~"
,~
'.
,
[ ,b
, ~t,, ~ i . ,
.
"J_K~.~ ~ DBB
D { ~ ~
Imm
B
A 1.8 A
2.2
...-.,
C'~/
ML
I
A 2. Mr....'?
/ ~
...
SN
.-,
Imm
Fig. 1. A: Dots represent approximate locations of neurons in the medial preoptic area (POA) and adjacent basal forebrain which displayed antidromic responses to ventral tegmental area stimulation. Tip locations of the stimulation electrodes are shown in B. AC, anterior commissure; ACB, nucleus accumbens; CP, cerebral peduncle; DBB, diagonal band of Broca; IP, interpeduncular nucleus; ML, medial lemniscus; MS, medial septum; OC, optic chiasma; R, red nucleus; SC, suprachiasmatic nucleus; SN, substantia nigra; SO, supraoptic nucleus. Numbers accompanying each panel denote distances from bregma in mm.
nergic cells [1]. Since the exact trajectory of POA axons is not known, minor bias in the stimulation sites could not be excluded to cause such a difference in the latency. Nevertheless, this may indicate that the change in the excitability occurs along certain lengths of the axon and is not limited to the axon terminals. E raised the antidromic activation threshold and prolonged the absolute refractory period. Several possible mechanisms have been proposed for E-induced changes in the neuronal excitability. Some POA neurons with axons to the ventral midbrain concentrate E [5]. The lordosis reflex in the OVX rat, which was used in the present study as a behavioral clue to the E effects, requires 48 hr to emerge following the administration [14]. The long latency may be needed for E-dependent synthesis of functional molecule through genomic activation [14]. Such possibility has been demonstrated [8]. A subset of cells in the rostral part of the ventromedial nucleus of the hypothalamus are specifically excited by E [2, 17]. Their topography and axonal projection to the midbrain differentiate them from the E-concentrating local neurons [14]. Proposed nongenomic effects of E in-
clude changes in the membrane fluidity to alter neuronal or myometrial excitability [4]. In the OVX female rat, E activates Na+,K+-ATPase in the POA, but diminishes it in the medial basal hypothalamus [15]. The electrogenic pump activity, which would affect the neuronal excitability through shifts in the resting potential, is associated with the membrane fluidity [20]. In the medial amygdala neurons in tissue slices, an increased K* conductance was accounted for E-induced hyperpolarization [9]. Analogous mechanism may be responsible for the lack of neuronal accommodation in the amygdala neurons in the slices from the E-treated rats [19]; altered turnover of amino acid receptors may also modulate neuronal excitability in response to ambient molecules such as glutamate [16]. Traditionally, E-induced inhibition of neuronal activity in the POA has been associated with a negative feedback effect of E on gonadotropin release [3]. The present group of POA cells, however, is clearly different from those involved in gonadotropin secretion, which are distributed in the anteroventral periventricular region [21]. The VTA, which is innervated by the POA and send off
174
efferents to the spinal cord and to the basal ganglia, occupy a strategic position in exercising behavioral control [14]. E initiates maternal behavior through POA, however, Numan et al. [10] suggest that the lateral hypothalamus may intervene in the descending path. The present direct POA projection to the VTA does not fit their scheme. Although a definite role of POA neurons in the regulation of E-dependent sexual receptivity in the female rat has been much debated [6], we have successfully suppressed the lordosis reflex by electrical stimulation of the POA [13] as well as the VTA [18]. The E-induced suppression of the POA would, therefore, culminate in the disinhibition of the reflex. However, there is still other behavior, such as aggression and running wheel activity [14] in which POA has been implicated as a site of E action. Therefore, until the neural circuit for each of these and other functions is better known, the role of the current subset of POA cells remains an open question. 1 Akaishi, T. and Sakuma, Y., Estrogen excites oxytocinergic, but not vasopressinergic cells in the paraventricular nucleus of female rat hypothalamus. Brain Res., 335 (1985) 302-305. 2 Akaishi, T. and Sakuma, Y., Projections of estrogen-sensitive neurones from the ventromedial hypothalamic nucleus of the female rat, J. Physiol., 372 (1986) 207-220. 3 Dyer, R.G., An electrophysiological dissection of the hypothalamic regions which regulate the pre-ovutatory secretion of luteinizing hormone in the rat, J. Physiol., 234 (1973) 421-442. 4 Erulker, S.D. and Wetzel, D.M., Steroid effects on excitable membranes, Curr. Topics Membr. Transport, 31 (1987) 141-190. 5 Fahrbach, S.E., Morrell, J.I. and Pfaff, D.W., Identification of medial preoptic neurons that concentrate estradiol and project to the midbrain in the rat, J. Comp. Neurol., 247 (1986) 364-382. 6 Kondo, Y., Shinoda, A., Yamanouchi, K. and Arai, Y., Recovery of lordotic activity by dorsal deafferentiation of the preoptic area in male and androgenized female rats, Physiol. Behav., 37 (1986) 495-498. 7 Maeda, H. and Mogenson, G.J., An electrophysiological study of inputs to neurons of the ventral tegmental area from the nucleus accumbens and medial preoptic-anterior hypothalamic areas, Brain Res., 197 (1980) 365-377. 8 Mobbs, C.V., Fink, G., Pfaff, D.W., Hip-70: A protein induced by
estrogen in the brain and LH-RH in the pituitary, Science, 247 (1990) 1477-1479. 9 Nabekura, J., Oomura, Y., Minami, T., Mizuno, Y. and Fukuda, A., Mechanism of the rapid effect of 17fl-estradiol on medial amygdala neurons, Science, 233 (1986) 226-228. 10 Numan, M., Morrell, J.I. and Pfaff, D.W., Anatomical identification of neurons in selected brain regions associated with maternal behavior deficits induced by knife cuts of the lateral hypothalamus in rats, J. Comp. Neurol., 237 (1985) 552-564. 11 Parsons, B., MacLusky, N.J., Krey, L., Pfaff, D.W. and McEwen, B.S., The temporal relationship between estrogen-inducible progestin receptors in the female rat brain and the time course of estrogen activation of mating behavior, Endocrinology 107 (1980) 774-779. 12 Perkins, M.N. and Whitehead, S.A., Responses and pharmacological properties of preoptic/anterior hypothalamic neurones following medial forebrain bundle stimulation, J. Physiol., 279 (1978) 347 360. 13 Pfaff, D.W. and Sakuma, Y., Facilitation of the lordosis reflex from the ventromedial nucleus of the hypothalamus, J. Physiol., 288 (1979) 189-202. 14 Pfaff, D.W., Cellular mechanisms of female reproductive behaviors. In E. Knobil and J. Nell (Eds.), The Physiology of Reproduction, Raven, New York, 1988, pp. 1487-1568. 15 Rodriguez del Castillo, A., Battaner, E., Guerra, M., Alonso, T. and Mas, M., Regional changes of brain Na÷,K+-transporting adenosine triphosphatase related to ovarian function, Brain Res., 416(1987) 113-118. 16 Sah, P. Hestrin, S., Nicoll, R.A., Tonic activation of NMDA receptors by ambient glutamate enhances excitability of neurons, Science, 246 (1989) 815-818. 17 Sakuma, Y. and Akaishi, T., Cell size, projection path, and localization of estrogen-sensitive neurons in the rat ventromedial hypothalamus, J. Neurophysiol., 57 (1987) 1148-1159. 18 Sakuma, Y., Akitsu, H., Hasegawa, N. and Hasegawa, T., Interruption of lordosis behavior from electrical stimulation of the female rat ventral tegmental area, Proc. Int. Union Physiol. Sci., 17 (1989) 310. 19 Schiess, M.C., Joels, M., Shinnick-GaUagher, P., Estrogen priming affects active membrane properties of medial amygdala neurons, Brain Res., 440 (1988) 380-385. 20 Szego, C.M. and Pietras, R.J., Membrane recognition and effector sites in steroid hormone action, Biochem. Act. Horm., 8 (1981) 307~,63. 21 Terasawa, E., Wiegand, S.J. and Bridson, W.E., A role for medial preoptic nucleus on afternoon proestrus in female rats, Am. J. Physiol., 238 (1980) E533-E539.