Testosterone induces hypothalamic aromatase during early development in quail

Testosterone induces hypothalamic aromatase during early development in quail

Brain Research, 377 (1986)63-72 Elsevier 63 BRE 11811 Testosterone Induces Hypothalamic Aromatase During Early Development in Quail M. SCHUMACHERz ...

840KB Sizes 0 Downloads 35 Views

Brain Research, 377 (1986)63-72 Elsevier

63

BRE 11811

Testosterone Induces Hypothalamic Aromatase During Early Development in Quail M. SCHUMACHERz and J.B. HUTCHISON1 1MRC Unit on the Development and Integration of Behaviour, Madingley, Cambridge CB3 8AA (U. K.) and 2Laboratoire de Biochimie G~n~raleet Compar~e, University of Lidge (Belgium) (Accepted November 12th, 1985) Key words: aromatase - - hypothalamus - - preoptic area - - testosterone - - metabolism of testosterone - - reproductive behavior - brain differentiation - - sexual differentiation

The effects of testosterone on androgen metabolizing enzymes were examined in the developing hypothalamus of male and female quail using an in vitro radiometric assay which measures metabolite formation in individual brain samples. Testosterone (T) administered by subcutaneous silastic implants to gonadectomized 4-day-old chicks increased formation of estradiol-17fl (E2) in both preoptic area + anterior hypothalamus (PA) and posterior hypothalamus + tuberal area (HT) to adult levels. The T-induced increase in E2 formation occurred to the same degree in both sexes. The increase was very small in control non-target areas, neostriatum intermediale + hyperstriatum ventrale (VN), of either sex. Testosterone had no effect on formation of 5a-dihydrotestosterone (5a-DHT), 5fldihydrotestosterone (5fl-DHT) and 5fl-androstane 3a,17fl-diol (5fl,3a-diol) in PA. Kinetic analysis of the rate of E2 production by hypothalamic tissue from castrated chicks (CX-chicks) and castrates treated with T (CX+T-chicks) indicates that the increase in hypothalamic aromatase activity by T corresponds to induction of the enzyme: the Vmax (maximum velocity) was increased by T (CXchicks, 21; CX+T-chicks, 91 fmol/mg FW/h), whereas the Kmwas unaffected (CX-chicks, 5.5; CX+T-chicks, 4.7 × 10-8 M). Testosterone treatment, effective for inducing PA and HT aromatase activity, also activated crowing and caused cloacal gland development; neither of these effects were sexually dimorphic. Our results indicate that: (1) T induces aromatase specifically in the hypothalamus during early post-hatching development, other pathways of T metabolism are not affected; and (2) the inducible aromatase is not sexually dimorphic in the developing brain. Since there are sex differences in adult brain aromatase, we conclude that capacity for induction of the hypothalamic aromatase becomes sexually differentiated after the post-hatching period.

INTRODUCTION Testosterone (T) exerts its activating effects on behavior and gonadotropin release through its metabolites 8'3°'48. A r o m a t i z a t i o n of T to estradiol-17fl (E2) in the preoptic area ( P O A ) is important for the activation of male sexual behavior 33'34'6°'62. In the hypothalamus, T is also converted to 54- and 5fl-dihydrotestosterone (54- and 5fl-DHT) and to the corresponding diols 6'54'61. Although 5 a - D H T facilitates sexual behavior of male quail in synergy with estrogens as an active metabolite 38'53, the 5fl-reduction of T represents an inactivation pathway 32'6~. The intracellular metabolism of T thus determines the sensitivity of the brain to androgens by inactivating T or by producing behaviorally active metabolites 33-35. The activities of the T-metabolizing brain enzymes

in the hypothalamus, and as a consequence the responsiveness of the brain to androgens, are regulated by factors like stress 64, the social e n v i r o n m e n t25'4°, the endocrine status of the animal and its genetic sex. Testosterone induces the activity of aromatase in the P O A of adult doves 62, quail 57 and rats 5°. Thus aromatase serves as a regulator of androgen action. In the event that more T becomes available from the Peripheral circulation, its action may either be further amplified or modified in androgen target-cells by increased production of behaviorally effective E2 (ref. 33). In birds, as in mammals, females are less sensitive to the activating effects of T on behavior than males 3' 12,41,43. In contrast to males, female quail never coi~ulate in response to T and the cloacal glands (an epithelial a n d r o g e n - d e p e n d e n t foam gland behind the

Correspondence: J.B. Hutchison, MRC Unit on the Development and Integration of Behaviour, Madingley, Cambridge, CB3 8AA, U.K. 0006-8993/86/$03.50(~) 1986 Elsevier Science Publishers B.V. (Biomedical Division)

64 cloaca 51) are less developed in females than in males treated with the same amount of hormone 3'12. The sexual dimorphism in behavior is not due to a higher peripheral catabolism of androgens in females, but is caused by sex differences in brain sensitivity to androgens 12'44. Recently, it has been shown that T increases hypothalamic aromatase activity to a lesser extent in adult female quail than in males 57. As the activation of copulation by androgen requires aromatization of T to E24, this quantitative enzymatic difference between adult males and females could contribute to the behavioral insensitivity of females to androgen. Behavioral insensitivity of the female quail to T is thought to be due to irreversible demasculinizing actions of ovarian estrogens during prehatching development 1'2. However, it is now well established that the process of female behavioral demasculinization is not restricted to a critical embryonic period as suggested by Adkins 2, but extends into the first weeks of life 1°'37'55. In order to understand the process of sexual differentiation and maturation of the brain, it is important to know whether brain metabolism of androgen in the female differs from that of the male, particularly with respect to aromatase activity after hatching when sexual differentiation of behavior is being completed. The present study examined: (1) the relationship between aromatase and other metabolic enzymes in the developing brain; (2) the inductive effects of T on these metabolic pathways; and (3) compared the enzymatic changes induced by T with effects on behavior and the cloacal gland. MATERIALS AND METHODS

Experimental animals and treatments Quail chicks were hatched in our laboratory (day of hatching = day 0). During the experiments, chicks were raised in heterosexual groups in a floor pen at 40 °C under a long photoperiod (16 h light:8 h dark). Water and food were available continuously. Chicks were castrated (C-birds) or sham-operated (1-birds) on day 3 using cryogenic anesthesia 31'37. A single incision was made on the left side posterior to the last rib through which either both testes or the left ovary were removed using a suction probe. The right ovary is not developed, and does not regenerate even after removal of the left gonad 26. On day 4, birds were im-

planted subcutaneously with 10 mm (Expt. 1) or 20 mm (Expt. 2) Silastic capsules (Dow Corning Silastic Tubing 602-265; 1.57 mm i.d., 2.41 mm o.d.) under cryogenic anesthesia. The implants remained empty or were filled with crystalline T (Sigma, T-1500). All the implants were preincubated for at least 12 h in a 0.9% saline solution at 40 °C to initiate steroid diffusion through the tube wall. On the day of sacrifice, growing chicks were weighed to the nearest g and their gonads and cloacal glands were measured with a calliper (greatest length x greatest width in mm 2, see ref. 51).

Topographical dissection of the brain Birds were killed by decapitation, and their brains were quickly removed and frozen on dry-ice with the ventral surface uppermost. Brain samples corresponding to the preoptic area + anterior hypothalamus (PA), the posterior hypothalamus + tuberal area (HT) and neostriatum intermediale + hyperstriatum ventrale (VN) as control area were dissected out under a Zeiss operating microscope equipped with a graticule (width x height, 2.1 x 1.25 mm; weight, 4.3 + 0.2 mg, X + S.E.M.). The dissection procedure has been described elsewhere 31.

Measurement of enzymatic activities In vitro microassay of small samples from individual animals has been described 31. Brain samples were weighed to the nearest 0.01 mg and homogenized in 90/A of ice-cold STMM buffer (0.25 M sucrose, 50 mM Trizma pH 7.4 at 37 °C (Sigma T-4378), 5 mM MgC12, 12.8 mM 2-mercaptoethanol) using glass microhomogenizers (100 #1, Jencons, Hemel Hempstead) which were rinsed with further 50 ~1 of buffer. After homogenization, homogenates were snapfrozen in an acetone dry-ice bath and thawed in an ice-water bath before incubation. Immediately after thawing, 100ktl fractions of the homogenates were incubated at 40 °C for 15 min in the presence of 18.8 nM of [la,2a-3H]T (spec. act. 60 Ci/mmol, the Radiochemical Centre, Amersham) and 100 /A of STMM-buffer containing 2.4 mM of NADPHz type III (Sigma, N-6505). In these conditions, the enzymatic reactions studied here are linear for time and substrate concentrations 31. Steroids were extracted 3 times with diethyl ether after addition of 5 ktg of carrier steroids and about

65 2000 counts of [4-14C]E2 and of [4-14C]T for recovery calculations of estrogens and androgens, respectively (all isotopes were from the Radiochemical Centre, Amersham and were purified by thin layer chromatography (TLC) before use). After addition of one volume of 1 M - N a O H to the dry steroids, androgens were extracted 3 times with a mixture of toluene/cyclohexane (1:1, v/v). After neutralization of the aqueous phase, estrogens were extracted 3 times with diethyl ether. Both neutral and phenolic steroids were separated by TLC on silica gel (Machery-Nagel no. 804023) in a mixture of dichloromethane-ether (85:15, v/v). The different metabolites were eluted in 100/~1 of absolute methanol and in 3 ml of Insta-fluor II (Packard) and counted in a Tri-Carb 3255 Liquid Scintillation Spectrometer using the preset 3HQ/14CQwindow. Counts were corrected for spillover, counting efficiency, tracer recovery and blanks containing buffer, N A D P H 2 and tracers were subtracted from the final values. Four metabolites were quantified: estradiol-17fl (E2), 5a-dihydrotestosterone (5a-DHT), 5fl-dihydrotestosterone (5fl-DHT) and 5fl-androstane-3a, 17fl diol (5fl,3ct-diol). The identity of these 4 metabolites has been confirmed for the hypothalamus of quail chicks by recrystallizations 31. Results are expressed as fmol/mg FW/h. The protein content of the different samples was ca. 6% as determined by the method of Bradford 14 and there were no differences in protein content between different brain areas. The mean weight of brain samples was 4.3 + 0.2 mg (X + S.E.M.). Homogenates from the same brain areas of one adult male were included in both experiments to measure interassay variability. Levels of enzymatic activities (aromatase, 5a-reductase and 5fl-reductase) for this male were almost identical to those obtained in a previous experiment for 8 sexually mature males 57. As a consequence, we can compare enzymatic activities measured in the present experiments in chicks to those measured in a previous experiment in adult males sT.

scribed and validated 24. Steroids were extracted from serum on extrelute columns and then separated on celite:ethylene glycol:propylene glycol columns (2:1:1, w/v/v) with different mixtures of iso-octaneethyl acetate (85:15, v/v for T).

Hormone assays

Behavioral observations

Blood samples were collected from chicks of Expt. 2 during sacrifice. The concentrations of T were measured in samples pooled from at least 3 chicks. The radioimmunoassay used in this study has been de-

The following behavior patterns were recorded: neck-grabs (NG), mount-attempts (MA), mounts (M) and cloacal contact movements (CCM; for a description of these behaviors, see ref. 37). During

Experimental procedures In the first experiment, 10 chicks of either sex (6 males and 4 females) were castrated 3 days after hatching. The day after castration, 6 chicks were implanted with 10 mm of Silastic implants filled with T (T-chicks) and 4 others received empty implants (Cchicks). Birds were tested for male sexual behavior on days 7 and 8. They were killed on day 10, 6 days after hormone treatment. At sacrifice, the presence of implants and the success of castration were checked. In Expt. 2, male and female chicks were gonadectomized or sham-operated at the age of 3 days and were implanted with 2 × 10 mm of T-filled or empty Silastic implants 1 day after surgery. The larger amount of T (10 mm in Expt. 1) was chosen because of the low behavioral effectiveness of 10 mm implants. Recently, it was shown that 20 mm of T-implants are required to induce full copulation in quail chicks n. To obtain sufficient chicks to compare the sexes, experimental results came from two hatches. Birds from the two hatches were distributed in a random order between the following 8 experimental groups: intact (I) and castrated (C) male and female chicks treated with empty or T-filled (T) Silastic implants (I-males (n = 7), I-females (n = 11); I + T males (n = 8), I+T-females (n = 13); C-males (n = 10), C-females (n = 10); C+T-males (n = 16) and C+T-females (n = 13)). Chicks were observed for sexual behavior during the morning of days 9 and 10. When chicks were sampied, each chick was weighed, the cloacal gland was measured, blood samples were taken and the presence of implants was checked. The size of the gonads of I-chicks was also measured. There was no gonadal regeneration in experimental chicks.

66 Expt. 1, chicks were isolated over-night in individual boxes at 40 °C on days 7 and 8 and were tested with a female chick for sexual behavior for 5 min during the following morning. Crowing activity was measured for 10 min following isolation. During Expt. 2, chicks were observed in the morning of days 9 and 10 for the occurrence of male sexual behavior in their floorpen. Afterwards, chicks were isolated for 10 min for the observation of crowing. Chicks which crowed or which showed male sexual behavior (M and/or CCM) were marked.

Statistics The amount of metabolites formed by individual brain samples in the different experimental groups were compared within one sex by A N O V A S followed by Newman-Keuls tests. Sexes were compared using two-tailed Student's t-tests. Behavioral data were analyzed by Fisher's exact probability tests. RESULTS

Effect of testosterone on metabolite formation (Expt. 1)

E2

50t-DHT

50_ ~0_ i'-"

4F 41,l-

30_

U-

E

l= w,-

20_ 10_ 0

C T 5,O-DHT

C T

5/]',3c~-DIOL

1200 _ 1000_ 8~ _ O r

600_ LL.

E

~00_

o

E

w.-,

2~_ 0

C

C T

Fig. 1. In vitro production of E2, 5a-DHT, 5fl-DHTand 5fl,3adiol from [la,2a-3H]T by the preoptic area + anterior hypothalamus (PA) of castrated (C, white columns) and castrated + T-treated (T, dark columns) chicks. *** P <~ 0.001 by twotailed t-test when compared to C-chicks.

Expt. 1 examined the effects of T on the formation of E2 in relation to the activity of other metabolic pathways in chicks. The metabolism of T was studied in the PA of these chicks, which is known to be a target area for the action of androgen 19'29'39 and has higher aromatase activity than adjacent areas 33,6°. At the age of 10 days, the mean weight of the chicks was 20 + 0.9 g (X + S.E.M.) and there were no differences in weight between the two groups. Testosterone-induced cloacal gland development (C-chicks, 12.7 + 1.5; T-chicks, 53 + 8.7 mm 2, X + S.E.M., P ~< 0.001 by two-tailed t-tests) and crowing (0/4 C-chicks and 5/6 T-chicks; P ~< 0.05 by Fisher's exact probability test). Only one T-chick performed neck-grabs during the two behavioral tests. Fig. 1 shows that E2, 5a- and 5fl-DHT and 5fl,3a-diol were produced by the PA of C-chicks and T-chicks. Testosterone treatment increased aromatase activity in the PA of chicks to adult levels 57 (T-chicks, 28.8 + 2; adult males, 34.4 + 3.8 fmol/mg FW/h, X + S.E.M., P > 0.2 by two-tailed t-test). However, 5aand 5fl-reductase activities were not affected by the androgen treatment. Therefore, the relationship between enzyme activities is altered by T. The production of 5fl-DHT in PA was 6 times higher, and the production of 5fl,3a-diol 7 times higher in chicks than in adult males 57 (5fl-DHT: T-chicks, 1020 + 140; adult males, 175 + 32 fmol/mg/FW/h, ,X + S.E.M., P < 0.001 by two-tailed t-test. 5fl,3a-diol: T-chicks, 224 + 36; adult males, 32 _+ 5 fmol/mg FW/h, X + S.E.M., P < 0.01 by two-tailed t-test).

Effects of testosterone on aromatase in male and female chicks (Expt. 2) Expt. 1 showed that T increases aromatase activity

67 in the P A of male and female chicks, but does not affect the activity of the o t h e r T-metabolizing enzymes. F o r this reason, we limited our study to the aromatization of T to E2. Expt. 2 was p e r f o r m e d to see if: (1) T induces a r o m a t a s e activity to the same extent in male and female chicks; and (2) the induction effect is limited to h y p o t h a l a m i c areas. As in Expt. 1, m e t a b o l i s m of T was studied in homogenates from P A , but also from H T (containing inducible a r o m a t a s e and involved in the control of gonadotropin secretion 21,57) and from VN (as control region) of 6 I-birds, 6 I + T - b i r d s , 6 C-birds and of 7 C + T - b i r d s of both sexes. All the T - t r e a t e d male chicks used for the study of T - m e t a b o l i s m showed mounting or cloacal contact movements. In view of the large n u m b e r of chicks required for Expt. 2, it was necessary to store brain tissue from the first hatch (see methods) for 14 days so that all samples could be incubated together for the aromatase assay. To d e t e r m i n e w h e t h e r storage at - 7 8 °C affected brain a r o m a t a s e activity, P A h o m o g e n a t e s incubated under standard conditions with 18.8 nM [3H]T for 15 min were carried out 2 h or 14 days after chicks were killed. Samples (n = 5 for each group) were stored on dry-ice ( - 7 8 °C). The result showed that chick P A can be stored on dry-ice without affecting a r o m a t a s e activity for 2 weeks. Yield of E2 after 2 h (10.5 + 0.7 fmol/mg FW/h; X + S . E . M . ) was similar to that of samples stored for 14 days (9.8 +

0.4; P > 0.05 by two-tailed t-test). Serum testosterone levels were similar in T-males (2.04 + 0.2) and T-females (2.01 + 0.3 ng/ml, X + S . E . M . ) and c o r r e s p o n d e d to those o b s e r v e d in adult males 23'24. Intact and castrated chicks had very low levels of T (0.35 + 0.12 ng/ml, X + S . E . M . ) . T h e r e were no differences in b o d y weight b e t w e e n the different groups of males. C + T - f e m a l e s were heavier than C-females ( C + T - f e m a l e s , 29.7 + 1; C-females, 25.9 + 0.9 g, X + S . E . M . , P ~< 0.05 by two-tailed t-tests). Testosterone induced the d e v e l o p m e n t of the cloacal gland in C- and 1-chicks (Table I). I + T - f e m a l e s had smaller glands than C + T - f e m a l e s and I + T males. Testis sizes were similar in I- and I + T - m a l e s (2.8 + 0.09 mg, X + S . E . M . ) , but I + T - f e m a l e s had larger ovaries than 1-females (I.+T-females, 11.2 + 0.8; I-females, 8.7 + 0.7 m m 2, ,X + S . E . M . , P ~< 0.05 by two-tailed t-tests). T e s t o s t e r o n e activated crowing in male and female chicks, but only male chicks p e r f o r m e d mounts and cloacal contact movements in response to the h o r m o n e t r e a t m e n t (Table I). The activity of VN a r o m a t a s e was at the limit of sensitivity of our assay ( < 2 fmol/mg FW/h). However, T induced a small but significant increase in VN formation of E2 in castrated chicks of b o t h sexes, but not in intact females. A r o m a t a s e activity was similar in I-males and in I + T - m a l e s and was decreased by

TABLE I Aromatase activity, crowing, mounting and cloacal gland development in male and female quail chicks

A: production of E2 by the preoptic area + anterior hypothalamus (PA) and by the posterior hypothalamus + tuberal area (HT) of castrated (CX) and intact (I) male and female chicks treated with 20 mm of empty or T-filled (+T) Silastic capsules. Results are expressed as the percentage of E2 produced in comparison with 8 adult males (see Materials and Methods and ref. 57). B: percentage of chicks which crowed. C: percentage of chicks which showed male sexual behavior. D: cloacal gland area in mm2. Males

(A) Aromatase activity (%) of adult PA HT (B) Crowing (%) (C) Mounting (%) (D) Cloacal gland (mm 2) (X + S.E.M.)

Females

CX

I

CX+ T

I+ T

CX

I

CX+ T

I+ T

30 27 0 0

29 28 0 0

118 155 75 75***

110 144 62 62**

29 33 0 0

25 34 0 0

99 134 77 0

113 137 77 0

13 + 1.9

10 + 1.1

64 + 2.6

67 + 2.5*** 14 _+ 1.5

8 + 1.2

60 ___1.9a

52 __+1.9

* P < 0.05; ** P < 0.01; *** P < 0.001 by two-tailed t-tests for the cloacal glands and by Fisher's exact probability tests for behaviors when compared to the corresponding female group. P < 0.01 when compared to I + T-females by Newman-Keuls tests.

68

VN Males

0 m-"

Femates

l=:

-h

0 E wm,--

C I

C I C I

C I

,4.

4-

÷

T

T

T T

4.

PA Mates

50_

Females

b

~,0_

b

b _L_k

a

C I

C

0 ,,,(=

3=

b

30_

U-

l::

0 l: W,-

20_ 10_ 0 C

I

4.

I

.4-

C I ÷

T T

.4.

T T

H_3 Males 0 ~b

50_

L,O_

Females b b

_L

castration (Fig. 2). Testosterone had a marked effect in increasing the P A aromatase activity in both intact and gonadectomized males and females. Yield of E2 in P A was equivalent to that of adult, sexually active males (Table I and ref. 57). There were no significant differences between T-induced aromatase activity in P A and H T of males or females, but T-induced increase in aromatase activity of VN differed significantly from both P A and H T in males and females (P ~< 0.001 by A N O V A ) .

Effect of substrate concentration on aromatase activity (Expt. 3) Expt. 3 determined the effect of T on kinetic properties of P A aromatase. Hypothalami from 9 Cchicks and 9 C+T-chicks were pooled separately and homogenized in STMM-buffer (final tissue concentration for both homogenates: 5 mg FW/100 #1 of STMM). One hundred/~1 fractions of the two homogenates were incubated with different amounts of [la,2a-3H]T ranging from 9.3 to 75.4 nM in the same conditions as used in the other experiments. Fig. 3 shows the effect of substrate concentration of [3H]T on the rate of E2 production by the hypothalamus of C-chicks and C+T-chicks. The production of E2 was much higher in chicks treated with T than in castrated chicks. The corresponding doublereciprocal plots show that the maximum velocity (Wmax)of aromatase was increased 4 times by T-treatment whereas the apparent Michaelis-Menten constant (Km) was not significantly altered (C-birds, 5.5 × 108 M; T-birds, 4.7 × 10 -s M).

0

30_

DISCUSSION

u_

s 0 E

~-

20_

The present study shows that T enhances the activity of aromatase in the hypothalamus of quail chicks to adult levels. To our knowledge, this is the first

10_ 0

C

I

C I 4.

÷

T T

C I

C I 4.

.4.

T T

Fig. 2. The production of E2 by the preoptic area + anterior hypothalamus (PA), the posterior hypothalamus + tuberal area (HT) and the neostriatum intermediale + hyperstriatum ventrale (VN) of male (white columns) and female (dark columns) chicks. Within one sex, columns with different letters are different at the P ~<0.05 level for the VN and at the P ~<0.01 level for the PA and HT by ANOVA followed by Newman-Keuls tests.

demonstration of induction of a T metabolizing enzyme in developing brain. The preoptic area is a known target for the behavioral effects of T in adult quail 54'57 and doves 33'62. A similar area in the quail chick brain (PA) contains an active aromatase complex which is clearly influenced by circulating T, as is the case in adult quail 57. Testosterone has only a weak effect in the neostriatum intermediale, suggesting that intracellular receptors for the inductive effect of T are mainly localized to hypothalamic areas

69 60.

CX÷T

50 r..

g 2~ u-

o

l=

l,O 30 20 10 0

CX

f I

I

I

I

25

50

75

100

S(nM) CX

300-

I.L

200. "6

100

CX÷T

I

I

I

I

25

50

75

100

1/S (lluM)

Fig. 3. Effect of substrate concentration of [la,2a-3H]T on the rate of E2 production by the hypothalamus of castrated (CX) and castrated + T-treated (CX+T) chicks (above) and the corresponding double-reciprocal plots (below). One hundred/A fractions of two homogenates from 9 C-chicks and 9 C+Tchicks were incubated at 40 °C for 15 min (linear conditions) with 1.2 mM of NADPH2 (saturating amount) and different amounts of [la,2a-3H]T ranging from 9.3 to 75.4 nM. (Each point represents X of duplicates.) CX-chicks: Vmax = 21 fmol/mg FW/h; Km = 5.5 × 10-8 M. CX+T-chicks: Vmax = 91 fmol/mgFW/h; Km= 4.7 × 10-8 M.

as in adult quail and doves 57'62. Our results indicate that induction of PA aromatase does not depend on the presence of the gonads and their secreted hormones, because there was no difference in E2 yield between gonadectomized and intact chicks treated with T. Evidently, the capacity for induction of the enzyme is not maintained by gonadal hormones in the developing brain. The effect of exogenous T on the PA is, however, exerted specifically on aromatase activity; neither 5a- nor 5fl-reductase activity is affected. Therefore, othe relationship between aromatase and other enzyme activities in both preoptic and hypothalamic areas is altered by T. Analysis of the double reciprocal plots for the rate

of E2 production by the hypothalamus of C-chicks and C+T-chicks suggests that the increase in aromatase activity by T-treatment corresponds to an induction of the enzyme: the Vmax was increased by Ttreatment whereas the value of the K m was not altered. The kinetic data indicate that T increases concentration of the hypothalamic enzyme without affecting affinity for T as substrate. These observations are in agreement with a previous study in adult doves 62. The Km for hypothalamic aromatase in CX-chicks (5.5 x 10-8 M) and CX+T-chicks (4.7 x 10-8 M) is similar to that observed in adult quail (1.5 x 10-8 M) 58 and adult doves (2.75 x 10-8 M) 62. There are no data as yet from mammalian species to suggest that POA aromatase is influenced by hormonal conditions during early development. Studies of ferrets 63 and rats 64 during the perinatal period indicate that the brain aromatase is not influenced by T. However, hypothalamic E2 levels have been shown to be consistently higher in neonatal male than female rats 15'49. The postnatal rise in hypothalamic E2 of male rats is correlated with a rise in postnatal T, and is prevented by neonatal castration t5'49. This is in agreement with our own studies of the quail chick showing that a rise in circulating androgens around hatching 46'47 is accompanied by increased PA aromatase activity59. In early development, birds are less sensitive to the behavioral effects of T than during adulthood 13'52. This has been related to the high inactivation of T in the hypothalamus of chicks through its 5fl-reduction 13. During post-hatching development, there is a major decrease in hypothalamic 5fl-reductase activity in cockerels 42, quail 9 and doves 35 suggesting that changing 5fl-reductase activity is important in brain maturation. The observation that treatment of chicks with T does not reduce the high hypothalamic 5fl-reductase activity of day 4 chicks suggests that rapid inactivation of T is intrinsic to the developing brain, and does not depend on low androgen levels. If high 5fl-reductase activity was a consequence of low brain concentrations of T, then T treatment should decrease 5fl-reduction. This does not happen, as is shown by the results of Expt. 1. As with morphological (cloacal gland) and behavioral (crowing) features which are hormone-sensitive in early development, the capacity for induction of brain aromatase by T is not sexually differentiated in

70 4-day-old chicks. In contrast to adult quail, where T is less effective in inducing hypothalamic aromatase activity in females than in males 57, T induces aromatase activity with the same effectiveness in the PA and HT of male and female chicks. The enzyme system is evidently differentiated later during post-hatching life. This provides further evidence, based on a physiological mechanism, that the process of sexual differentiation in quail is not limited to a critical period in embryonic life as suggested by Adkins 2, but is a continuous process extending to the first 4 weeks of life 1°'37'55. Ovarian E2 contributes to the full morphological and behavioral demasculinization of the female quail by acting during the first 4 weeks of life 1°'31. As aromatization of T to E2 in the PA is an important step in the activation of adult male sexual behavior 4, the loss by females of their sensitivity to the behavior activating effects of T 10'37'55 during development may thus correspond to the post-hatching differentiation of the aromatase system in this brain target area for T. The post-hatching decrease in sensitivity of the HT-aromatase to the inductive effects of T may correspond to the sexual differentiation of gonadotropin secretion 55, particularly as the HT is involved in the control of LH release 2°-22. Results of Expts. 1 and 2 show that testosterone induces copulatory behavior in male, but not in female chicks. However, neonatally ovariectomized females do show male sexual behavior in response to T as adults 1°'37'55. These observations are consistent with the view that females are already partly demasculinized at hatching and that they have a higher threshold for the activation of male sexual behavior 1°,55. Adult, neonatally ovariectomized females only show male sexual behavior when tested in optimal conditions 1°'55. Only incipient copulation in post-hatching female chickens is induced by T in response to handtests, suggesting that the female insensitivity to androgens may be partly achieved, but not completed after hatching 5. Induction of hypothalamic aromatase by the T-treatment of quail chicks may also complete the behavioral demasculinization of the females, since only female chicks are sensitive to the demasculinizing actions of estrogens 1°. Systemic testosterone induced the development of the cloacal gland with the same effectiveness in both male and female chicks. The finding that this androgen-dependent structure is not sexually differen-

tiated in chicks fits well with the observation that E2 demasculinizes the gland in females by acting during the first weeks of life 1°'55. In contrast to the hypothalamus, the cloacal gland does not aromatize T into E2. Treatment of chicks with T will not, therefore, cause the demasculinization of this structure by production of E2. We conclude that: (1) in the developing brain, aromatase activity is induced by T to levels seen in adults; (2) the inducible aromatase in the PA is not sexually differentiated in early post-hatching development, whereas it is differentiated in the adult; and (3) sexual differentiation of hypothalamic enzymes occurs after the post-hatching period. The latter conclusion is in agreement with the hypothesis that sexual differentiation is a continuous process in quail extending during the first weeks of life 1°'37'55, but contrasts with the earlier view that female demasculinization is limited to a critical period during embryonic development 3. The developmental decline in sensitivity of the female brain to the inductive effects of T on hypothalamic aromatase may correspond to progressive death of aromatase containing cells. Sex differences in cell death which might contribute to hypothalamic sex dimorphism have been proposed by Gorski 27. Alternatively, aromatase containing cells might be differentiated to become less sensitive in females or more sensitive in males to the inductive effects of androgens. In this latter case, sexual differentiation might correspond to a permanent inhibition or activation of hormone-inducible genomic expression. ACKNOWLEDGEMENTS This work was supported by the British Medical Research Council. M.S. is aspirant du Belgian Fonds National de la Recherche Scientifique (FNRS). His visit to the MRC Unit on the Development and Integration of Behaviour was supported by a grant from the European Science Foundation (ETP) and by a grant from the British Council. We thank Y. Delville for technical assistance during the R I A and are grateful to P.P.G. Bateson for allowing us to use the quail breeding facilities of the Sub-Department, and to J. Owen for his technical expertise in supplying chicks. We thank R.E. Hutchison for comments on the manuscript.

71 REFERENCES 1 Adkins, E.K., Hormonal control of sexual differentiation in the Japanese quail, J. Comp. Physiol. Psychol., 89 (1975) 61-71. 2 Adkins, E.K., Effect of embryonic treatment with estradiol or testosterone on sexual differentiation of the quail brain, Neuroendocrinology, 29 (1979) 178-185. 3 Adkins, E.K. and Adler, N.T., Hormonal control of behavior in the Japanese quail, J. Cornp. Physiol. Psychol., 81 (1972) 27-36. 4 Adkins, E.K., Boop, J.J., Koutnik, D.L., Morris, J.B. and Pniewski, E.E., Further evidence that androgen aromatization is essential for the activation of copulation in male quail, Physiol. Behav., 24 (1980) 441-446. 5 Andrew, R.J., Precocious adult behaviour in the young chick, Anim. Behav., 14 (1966) 485-500. 6 Balthazart, J., Hormonal correlates of behavior. In D.S. Farner, J.R. King and K.C. Parkes (Eds.), Avian Biology, Vol. 7, Academic Press, New York, 1983, pp. 221-365. 7 Balthazart, J. and Hirschberg, D., Effect of several androgens on testosterone metabolism in the brain and crest of male chicks, IRCS Med. Sci., 10 (1982) 377-378. 8 Balthazart, J. and Schumacher, M., Testosterone metabolism and sexual differentiation in quail. In J. Balthazart, E. Pr6ve and R. GiUes (Eds.), Hormones and Behaviour in Higher Vertebrates, Springer Verlag, Heidelberg, 1983, pp. 235-260. 9 Balthazart, J. and Schumacher, M., Changes in testosterone metabolism by the brain and cloacal gland during sexual maturation in the Japanese quail (Coturnix coturnix japonica), J. Endocrinol., 100 (1984) 13-18. 10 Balthazart, J. and Schumacher, M., Estradiol contributes to the postnatal demasculinization of female Japanese quail (Coturnix coturnix japonica), Horrn. Behav., 18 (1984) 287-297. 11 Balthazart, J. and Schumacher, M., A two step model for sexual differentation, Proc. N. Y. Acad. Sci., in press. 12 Balthazart, J., Schumacher, M. and Ottinger, M.A., Sexual differences in the Japanese quail: behavior, morphology and intracellular metabolism of testosterone, Gen. Comp. Endocrinol., 51 (1983) 191-207. 13 Balthazart, J., Schumacher, M. and Malacarne, G., Relative potencies of testosterone and 5a-dihydrotestosterone on crowing and cloacal gland growth in the Japanese quail (Coturnix coturnix japonica), J. Endocrinol., 100 (1984) 19-23. 14 Bradford, M.M., A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding, Anal. Biochem., 72 (1976) 248-254. 15 Butte, J.C., Moore, J.A. and Kakihana, R., Brain and plasma levels of testosterone, dihydrotestosterone and estradiol in the one-day-old rat, Life Sci., 24 (1979) 2343-2349. 16 Callard, G.V., Aromatization is cyclic AMP-dependent in cultured brain cells, Brain Research, 204 (1981) 461-464. 17 Callard, G.V., Petro, Z. and Ryan, K.J., Conversion of androgen to estrogen and other steroids in the vertebrate brain, Am. Zool., 18 (1978) 511-523. 18 Daniel, S.A.J. and Armstrong, D.T., Enhancement of follicle-stimulating hormone-induced aromatase activity by andogens in cultured rat granulosa cells, Endocrinology, 107 (1980) 1027-1033.

19 Davidson, J.M., Activation of the male rats sexual behavior by intracerebral implantation of androgen, Endocrinology, 79 (1966) 783-794. 20 Davies, D.T., The neuroendocrine control of gonadotrophin release in the Japanese quail. III. The role of the tuberal and anterior hypothalamus in the control of ovarian development and ovulation, Proc. Roy. Soc. London, Ser. B., 206 (1980) 421-437. 21 Davies, D.T. and Follett, B.K., The neuroendocrine control of gonadotrophin release in the Japanese quail. I. Role of the tuberal hypothalamus, Proc. R. Soc. London, Ser. B., 191 (1975) 285-301. 22 Davies, D.T. and Follett, B.K., Electrical stimulation of the hypothalamus and luteinizing hormone secretion in Japanese quail, J. Endocrinol., 67 (1975) 431-438. 23 Delville, Y., Sulon, J., Hendrick, J.-C. and Balthazart, J., Effect of the presence of females on the pituitary-testicular activity in male Japanese quail (Coturnix coturnix japonica), Gen. Comp. Endocrinol., 55 (1984) 295-305. 24 Delville, Y., Sulon, J. and Balthazart, J., Hormonal correlates of gonadal regression and spontaneous recovery in Japanese quail exposed to short day-lengths, Arch. Int. Physiol. Biochirn., in press. 25 Dessi-Fulgheri, F. and Lupo, C., Odour of male and female rats changes hypothalamic aromatase and 5a-reductase activity and plasma sex steroid levels in unisexually reared male rats, Physiol. Behav., 28 (1982) 231-235. 26 Gibson, W.R., Follett, B.K. and Gledhill, B., Plasma levels of luteinizing hormone in gonadectomized Japanese quail exposed to short or to long day lengths, J. Endocrinol., 64 (1975) 87-101. 27 Gorski, R.A., Critical role for the medial preoptic area in the sexual differentiation of the brain, Progr. Brain Res., 61 (1984) 129-146. 28 Hillier, S.G. and De Zwart, F.A., Evidence that granulosa cell aromatase induction/activation by follicle-stimulating hormone is an androgen receptor-regulated process in vitro, Endocrinology, 109 (1981) 1303-1305. 29 Hutchison, J.B., Initiation of courtship by hypothalamic implants of testosterone propionate in castrated doves (Streptopelia risoria), Nature (London), 216 (1967) 591-592. 30 Hutchison, J.B., Hypothalamic regulation of male sexual responsiveness to androgen. In J.B. Hutchison (Ed.), Biological Determinants of Sexual Behaviour, Wiley, New York, 1978, pp. 277-319. 31 Hutchison, J.B. and Schumacher, M., Development of testosterone metabolizing pathways in the avian brain: enzyme localization and characteristics, Dev. Brain Res., 25 (1986) 33-42. 32 Hutchison, J.B. and Steimer, Th., Brain 5fl-reductase. A correlate of behavioral sensitivity to androgen, Science, 213 (1981) 244-246. 33 Hutchison, J.B. and Steimer, Th., Hormone-mediated behavioural transitions: a role for brain aromatase. In J. Balthazart, E. Pr6ve and R. Gilles (Eds.), Hormones and Behaviour in Higher Vertebrates, Springer-Verlag, Berlin, 1983, pp. 261-274. 34 Hutchison, J.B. and Steimer, Th., Androgen metabolism in the brain: Behavioral correlates, Prog. Brain Res., 61 (1984) 23-51. 35 Hutchison, J.B. and Hutchison R.E., Phasic effects of hormones in the avian brain during behavioural development. In R. Gilles and J. Balthazart (Eds.), Neurobiology,

72 Springer-Verlag, Berlin, 1985, pp. 105-120. 36 Hutchison, J.B., Steimer, Th. and Duncan, M., Behavioural action of androgen in the dove: effects of long-term castration on response specificity and brain aromatization, J. Endocrinol., 90 (1981) 167-178. 37 Hutchison, R.E., Hormonal differentiation of sexual behavior in Japanese quail, Horm. Behav., 11 (1978) 363-387. 38 Larsson, K., S6dersten, P. and Beyer, C., Sexual behavior in male rats treated with estrogen in combination with dihydrotestosterone, Horm. Behav., 4 (1973) 289-299. 39 Lisk, R.D., Neural localization for androgen activation of copulatory behavior in the male rat, Endocrinology, 80 (1967) 754-761. 40 Lupo Di Prisco, C., Lucarini, N. and Dessi-Fulgheri, E., Testosterone aromatization in rat brain is modulated by social environment, Physiol. Behav., 20 (1978) 345-348. 41 MacLusky, N.J. and Naftolin, F., Sexual differentiation of the central nervous system, Science, 211 (1981) 1294-1302. 42 Massa, R. and Sharp, P.J., Conversion of testosterone to 5/3-reduced metabolites in the neuroendocrine tissues of the maturing cockerel, J. Endocrinol., 88 (1981) 263-269. 43 McEwen, B.S., Sexual maturation and differentiation: the role of gonadal steroids, Progr. Brain Res., 48 (1978) 291-307. 44 Mode, A., Gustafsson, J.A., SOdersten, P. and Eneroth, P., Sex differences in behavioural androgen sensitivity: possible role of androgen metabolism, J. Endocrinol., 100 (1984) 245-248. 45 Nock, B. and Feder, H.H., Neurotransmitter modulation of steroid action in target cells that mediate reproduction and reproductive behavior, Neurosci. Behav. Rev., 5 (1981) 437-447. 46 Ottinger, M.A. and Brinkley, H.T., Testosterone and sex related physical characteristics during maturation of the male Japanese quail (Coturnix coturnix japonica), Biol. Reprod., 20 (1979) 905-909. 47 Ottinger, M.A. and Bakst, M.R., Peripheral androgen concentrations and testicular morphology in embryonic and young male Japanese quail, Gen. Comp. Endocrinol., 43 (1981) 170-177. 48 Perez-Palacios, G., Larsson, K. and Beyer, C., Biological significance of the metabolism of androgens in the central nervous system, J. Steroid Biochem., 6 (1975) 999-1006. 49 Rhoda, J., Corbier, P. and Roffi, J., Gonadal steroid concentrations in serum and hypothalamus of the rat at birth: aromatisation of testosterone to 17/3-estradiol, Endocrinology, 114 (1984) 1754-1760. 50 Roselli, C.E., Ellinwood, W.E. and Resko, J.A., Regulation of brain aromatase activity in rats, Endocrinology, 114 (1984) 192-200.

51 Sachs, B.D., Photoperiodic control of the cloacal gland of the Japanese quail, Science, 157 (1967) 201-203. 52 Schleidt, W.M., Precocial sexual behaviour in turkeys (Meleagris Gallopavo L.), Anim. Behav., 18 (1970) 760-761. 53 Schumacher, M. and Balthazart, J., The effects of testosterone and its metabolites on sexual behavior and morphology in male and female Japanese quail, Physiol. Behav., 30 (1983) 335-339. 54 Schumacher, M. and Balthazart, J., Sexual dimorphism in the hypothalamic metabolism of testosterone in the Japanese quail (Coturnix coturnix japonica), Progr. Brain Res., 61 (1984) 51-59. 55 Schumacher, M. and Balthazart, J., The postnatal demasculinization of sexual behavior in the Japanese quail (Coturnix coturnix japonica), Horm. Behav., 18 (1984) 298-312. 56 Schumacher, M. and Balthazart, J., Sexual differentiation is a biphasic process in mammals and birds. In R. Gilles and J. Balthazart (Eds.), Neurobiology, Springer-Verlag, Berlin, 1985, pp. 203-219. 57 Schumacher, M. and Balthazart, J., Testosterone-induced brain aromatase is sexually dimorphic, Brain Research, in press. 58 Schumacher, M., Contenti, E. and Balthazart, J., Partial characterization of testosterone-metabolizing enzymes in the quail brain, Brain Research, 305 (1984) 51-59. 59 Schumacher, M., Hutchison, J.B., Hutchison, R.E. and Delville, Y., Developmental changes in androgen metabolism around hatching in the Japanese quail (Coturnix coturnix japonica), in preparation. 60 Steimer, Th. and Hutchison, J.B., Aromatization of testosterone within a discrete hypothalamic area associated with the behavioral action of androgen in the male dove, Brain Research, 192 (1980) 586-591. 61 Steimer, Th. and Hutchison, J.B., Metabolic control of the behavioural action of androgens in the dove brain: testosterone inactivation by 5fl-reduction, Brain Research, 207 (1981) 1-16. 62 Steimer, Th. and Hutchison, J.B., Androgen increases formation of behaviourally effective oestrogen in dove brain, Nature (London), 292 (1981) 345-347. 63 Tobet, S.A., Shim, J.H., Osiecki, S.T., Baum, M.J. and Canick, J.A., Androgen aromatization and 5a-reduction in ferret brain during perinatal development: effects of sex and testosterone manipulation, Endocrinology, 116 (1985) 1869-1877. 64 Weisz, J., Brown, B.L. and Ward, I.L., Maternal stress decreases steroid aromatase activity in brains of male and female male fetuses, Neuroendocrinology, 35 (1982) 374-379.