Journal of Neuroendocrinology, 1999, Vol. 11, 839–847

Mating Stimuli Influence Endogenous Variations in the Neurosteroids3a,5a-THP and 3a-Diol

C. A. Frye and L. E. BayonDepartment of Psychology, The University at Albany-SUNY, Albany, NY, USA.

Key words: allopregnanolone, non-genomic, lordosis, midbrain, hypothalamus, reproductive behaviour.

Abstract

Progesterone facilitates and 5a-dihydrotestosterone (DHT) inhibits female sexual behaviour inrodents; their metabolites, 5a-pregnan-3a-ol-20-one (3a,5a-THP) and 5a-androstane-3a-17b-Diol (3a-Diol), may influence the onset and termination of lordosis. Changes in these and related steroids inhormonal states associated with differences in receptivity were investigated. Rats were assigned tooestrus, metoestrus, dioestrus, pro-oestrus, mated, gestational days 5–7, 12–14, 18–20, or post-partum conditions; rats 9+ months of age were considered older. Pro-oestrus rats were exposed tothe mating arena, sight and smell of a male with no tactile contact, artificial vaginocervicalstimulation, standard mating, or no mating. Progesterone, 5a-pregnane-3,20-dione, 3a,5a-THP,oestradiol, testosterone, DHT, 3a-Diol, dehydroepiandrosterone, and corticosterone were measuredin plasma and whole brain, midbrain, hypothalamus, cortex, amygdala, hippocampus. 3a,5a-THP and3a-Diol changed with reproductive state and mating stimuli. Plasma and whole brain 3a,5a-THP and3a-Diol were significantly increased in pro-oestrus versus dioestrus rats; plasma 3a,5a-THP wasdecreased and 3a-Diol increased in mated versus pro-oestrus rats. The midbrain and hypothalamushad the most evident changes in 3a,5a-THP and 3a-Diol between dioestrus versus pro-oestrus andpro-oestrus versus mated rats. 3a,5a-THP and 3a-Diol were altered differently in response to matingstimuli. 3a,5a-THP was greater in the midbrain of mated versus pro-oestrus rats; other mating-relevant stimuli decreased 3a,5a-THP. Midbrain 3a-Diol was increased with exposure to a male<VCS< mating. 3a,5a-THP was increased and 3a-Diol was decreased in the hypothalamus ofmated versus pro-oestrus rats; exposure to the various mating stimuli decreased 3a,5a-THP and 3a-Diol. The neurosteroids, 3a,5a-THP and 3a-Diol, vary with mating in the hypothalamus and midbrainof rats.

Sex steroids, synthesized in the gonads and placenta, are 9). This suggests that progesterone and its metabolites mayhave important effects in the mediation of the onset andtraditionally understood to mediate the onset and duration

of sexual receptivity in rodents; however, their metabolites duration of lordosis behavior in female rodents.Oestrus cessation, mediated by androgens, is as importantmay also have important effects on sexual receptivity. In

general, progestins facilitate mating behaviour ( lordosis to successful mating as is facilitation and maintenance ofreproductive behaviour by progestins. Dihydrotestosteronebehavior) by acting on the ventromedial hypothalamus (1)

of oestradiol-primed rodents. Within the ventral medial hypo- (DHT) inhibits oestradiol-induced sexual receptivity inrodents (10–12) by actions in the VMH and POA (13–17).thalamus (VMH), progesterone’s genomic actions at

intracellular oestradiol-induced progestin receptors (PRs) are There is strong evidence that DHT’s inhibitory effects onsexual receptivity are mediated by actions of its metabolitecritical for the manifestation of mating behaviour (2, 3);

however, administration of progesterone metabolites which 5a-androstan-17b-ol-3-one (3a-Diol ), which lacks a highaffinity for intracellular androgen receptors (ARs; 18,19).are devoid of activity at PRs (4) to the VMH, preoptic area

(5, 6), or midbrain (7) also enhances female sexual behaviour. Blocking DHT’s metabolism to 3a-Diol by administering a5a-reductase inhibitor disrupts oestrus cessation in femaleConversely, interfering with progesterone’s metabolism dis-

rupts progesterone-facilitated sexual receptivity in rodents (8, rats (11). This suggests that testosterone or its metabolites

Correspondence to: Cheryl A. Frye, Department of Psychology, University at Albany, State University of New York, 1400 Washington Avenue,Albany, NY 12222, USA (e-mail: cafrye@cnsunix.albany.edu).

© 1999 Blackwell Science Ltd

840 Neurosteroids and mating

months) were considered older rats. Rats were killed by rapid decapitationmay have important effects in mediating oestrus cessation inin the particular condition. Trunk blood was collected and plasma was storedfemale rodents.at −20 °C for experiment 1; whole brains (n=44; yielding four/group;

The progestin 3a,5a-THP, which facilitates, and the andro- experiment 1) were rapidly removed from the skull (where appropriate,gen 3a-Diol, which inhibits lordosis, are metabolized from midbrain, hypothalamus, cortex, hippocampus, amygdala where dissected

(n=268; yielding 24 tissues/brain site/group; yielding eight observations/groupprogesterone and testosterone by the 5a-reductase andfor experiment 2) and tissue was stored at −70 °C until assayed for progester-3-hydroxysteroid enzymes, which have been localized to theone, 5a-pregnane-3,20-dione (DHP), 3a-hydroxy-5a-pregnane-20-one (3a,5a-hypothalamus and midbrain (20). In addition to being metab- THP), testosterone, 5a-androstan-17b-ol-3-one (DHT ), 5a-androstane-

olized from gonadal progesterone and testosterone, 3a,5a- 3a,17b-diol (3a-Diol ),17b-estradiol (oestradiol ), dehydroepiandrosterone(DHEA), and corticosterone.THP and 3a-Diol are synthesized de novo from cholesterol in

the central nervous system. Like other neurosteroids, 3a,5a-Experiment 3: Variations in central concentrations of 3a,5a-THP and 3a-Diol

THP (21) and 3a-Diol (16, 22) do not bind well to intracellu- in response to different mating relevant stimulilar steroid receptors (4, 18, 19) but are potent modulators of

Vaginal smears were obtained daily as outlined above for experiments 1 andc-aminobutyric acid (GABA)A/benzodiazepine receptor com- 2 from rats to determine the day of the oestrous cycle. Normally cycling ratsplexes (GBRs). were assigned to one of the following groups on proestrus: pro-oestrus (home

cage), exposure to the mating arena, exposure to sight and smell of a maleA peak in plasma concentrations of 3a,5a-THP on(with no tactile contact), exogenous vaginocervical stimulation (VCS; 5–10proestrus/early oestrus and a precipitous decline later inintromissions were administered with a glass eyedropper), and standard

estrus (23–25) are followed by a similar pattern of plasma mating. As described above, rats (n=120; resulting in 40 tissues per group,3a-Diol (26–29). Hence, there seems to be a temporal correla- which yielded eight observations/group) were killed via rapid decapitation,

plasma collected, and brain sites dissected according to the landmarkstion between plasma peaks in 3a,5a-THP and facilitation ofpreviously described (40) for later radioimmunoassay of 3a,5a-THP andsexual behaviour, and peaks in 3a-Diol and facilitation3a-Diol.followed by inhibition of oestrous behaviour. Concentrations

of progesterone and 3a-Diol are increased in response to Radioimmunoassaymating (30), although exposure to stressful environmental

Extractionsituations, such as cold water swim stress or CO2 exposure,For steroid hormone measurements, plasma was extracted with diethyl etheralso increases plasma concentrations of 3a,5a-THP (31–34)and trace amounts of 3H ligand; ether was evaporated, and the pellets were

and 3a-Diol (35). Administration of neurosteroids to males reconstituted in phosphate assay buffer (pH=7.4).enhances their preference for the odour of a receptive female For brain measurements of progesterone, testosterone, DHT, 3a-Diol,

oestradiol, and DHEA, brain tissue (one whole brain or pools of three brain(36) and exposure to a female increases 3a,5a-THP in thesites) was homogenized with a glass/Teflon homogenizer in distilled water.olfactory bulb of male rats (37).Steroids were extracted from the homogenate with diethyl ether, and driedTo investigate the relationship between progestins anddown in a savant. The pellet was reconstituted in Trimethyl Pentane (TMP)

androgens during reproductive states, levels of 3a,5a-THP to half the homogenate volume (5 ml of reconstituted extract was set asideand 3a-Diol, as well as their parent compounds, progesterone for assay for DHEA). Reconstituted extracts were separated using Celite

column chromatography. Solvents of increasing polarity were used to eluteand dihydroprogesterone (DHP), and testosterone and DHT,the steroid: progesterone (100% TMP), DHT (5% ethyl acetate/TMP),were examined in plasma, whole brain, and tissues pooledtestosterone and 3a-Diol (15% ethyl acetate/TMP), and estradiol (40% ethyl/from midbrain, hypothalamus, cortex, amygdala, and hippo- acetate TMP). Fractions were dried in a Savant and then reconstituted in

campus of dioestrus, pro-oestrus, oestrus, metoestrus, mated, phosphate assay buffer.For DHP, 3a,5a-THP, and corticosterone brains (one whole brain or poolsfirst trimester, second trimester, third trimester, lactating, and

of three brain sites) were homogenized in 50% MeOH, 1% acetic acid using9-month-old, intact rats. DHEA was also examined becausea glass/glass tissue homogenizer. Steroids were extracted from the homogen-this neurosteroid changes in males exposed to a receptiveized brain samples in 50% MeOH, 1% acetic acid, through a series of

female (37, 38). Corticosterone was assessed to ascertain centrifugation and filtrations/washes with increasing concentrations of MeOH.whether variations in hormone levels of 3a,5a-THP and 3a- The final filtrate was dried in a Savant, and reconstituted in phosphate assay

buffer assay buffer. 300 ml of 0.1 M phosphate assay buffer (pH=7.4).Diol may be due to stressful environmental stimuli.

Assay

Plasma and brain samples were assayed according to previously publishedMaterials and methodsmethods (30, 41, 42). All assays were performed using the following tritiated

Subjects and housing steroids and antibodies. The tritiated steroids were progesterone: NET- 208:specific activity=47.5 ci/mmol; 3a,5a-THP (used for 3a,5a-THP and DHP):Female Long-Evans rats (n=428), were obtained from Charles RiverNET-1047: specific activity=ci/mmol; testosterone: NET-387: specific activ-Laboratories ( Kingston, NY, USA) at #55 days of age and housed inity=51.0 ci/mmol; DHT: NET-302: specific activity=43.5 ci/mmol; 3a-Diol:hanging stainless steel cages (24×18×19 cm) in a temperature-controlledNET-806: specific activity=41.00 ci/mmol; DHEA: NET-814: specific activity:room (21±1 °C). Rats were maintained on a 12 h light:12 h dark cycle ( lights92.5 ci/mmol; corticosterone: NET 182: specific activity=48.2 ci/mmol (alloff at 8.00 am) with continuous access to Purina Rat Chow and water.purchased from New England Nuclear, Boston, MA, USA). The antibodies

Tissue collection were progesterone (Dr G.D. Niswender, Colorado State University, Boulder,CO: no. 337): 1530 000; DHP and 3a,5a-THP (Dr Robert Purdy, Veteran’s

Experiments 1 & 2: Variations in steroid hormone concentrations in plasma and Medical Center, La Jolla, CA, USA: no. X-947; no. 921412-5): 155000;brain testosterone (Endocrine Sciences, Tarzana, CA, USA CA: no. T3-125):

1520 000; DHT (Endocrine Sciences: no. DT3-351): 1510 000; 3a-Diol (DrVaginal smears (39) were obtained daily from rats to determine the day ofthe oestrous cycle. After two normal cycles (4–5 days/cycle), animals were P.N. Rao, Southwestern Medical Center, San Antonio, TX, USA no. X-144):

1520 000; corticosterone (Endocrine Sciences: no. B3-136): 1525; DHEArandomly assigned to one of the following endocrine groups, which areassociated with differences in sexual receptivity: dioestrus 1, dioestrus 2, pro- (Endocrine Sciences: D7-421): 1525.

All standard curves were prepared in duplicate; progesterone (range=oestrus, oestrus, metoestrus, mated, first trimester pregnancy (gestational day(GD) 5–7), second trimester pregnancy (GD 12–14), third trimester preg- 50 pg/ml–8000 pg/ml; DHP (range 50 pg/ml–8000 pg/ml ); 3a,5a-THP

(range=10 pg/ml–4000 pg/ml; testosterone, DHT, and 3a-Diol (range=nancy (GD 18–20), or post-partum ( lactating). An additional 28 rats (>9

© 1999 Blackwell Science Ltd, Journal of Neuroendocrinology, 11, 839–847

Neurosteroids and mating 841

50 pg/ml–2000 pg/ml ); DHEA (range=0.01 ng/ml–4.0 ng/ml ); corticos-terone (range=0.01 ng/ml–4.0 ng/ml ). The standards were added to BSAassay buffer, followed by addition of the appropriate antibody and [3H ]steroid.

Assay tubes were vortexed and incubated at 4 °C (with the exception of3a-Diol which was incubated at room temperature) for 24 h. Separation ofbound and free was done by the rapid addition of dextran-coated charcoal.Following incubation with charcoal, samples were centrifuged at 1200 g; thesupernatant was pipetted into a glass scintillation vial with scintillationcocktail. Sample tube concentrations were calculated using the logit-logmethod, interpolation of the standards and correction for recovery. The intra-assay and inter-assay coefficients of variance for progesterone were 9% and10%; DHP and 3a,5a-THP, 12% and 15%; testosterone, 5% and 5%; DHT,2% and 15%; 3a-Diol, 9% and 10%; DHEA, 7% and 11%; corticosterone, 6%and 10%.

ELISA

Oestradiol was measured using a competitive ELISA (Enzyme LinkedImmunosorbant Assay) kit, with all necessary reagents (Oxford BiomedicalResearch, Oxford, MI, USA: EA70). Standard curves were prepared induplicate (range=0.0–2 ng/ml ). Fifty ml of the standards or extracted hor-mone followed by enzyme conjugate was added to the ELISA plate andallowed to incubate at room temperature; the plate was then washed withbuffer, followed by addition of tetramethylbenzidine for visualization. Thereaction was stopped by the rapid addition of 1.0 M HCL. The plate wasthen read at 450 nm in a microplate reader (Bio-Tek Instruments Inc.,Winooski, VT, USA). Sample concentrations were extrapolated from thestandard curve.

Results

Experiment 1: Plasma and whole brain steroid hormoneconcentrations vary with reproductive state

Plasma hormone concentrations significantly differed as afunction of hormonal and reproductive state. Oestrous cyclevariations in plasma steroid concentrations, particularly theneurosteroids, were characterized by pronounced differencesbetween receptive, pro-oestrus rats and non-receptive, dioes-

Oestrus

Met-oestrus

Dioestrus1 & 2

Pro-oestrus

Wh

ole

bra

in c

on

cen

trat

ion

s (n

g/g

)

Mated

G D5-7

G D12-14

G D18-20

Post-partum

Older

605550454035302520151050

Oestrus

Met-oestrus

Dioestrus1 & 2

Pro-oestrus

Pla

sma

con

cen

trat

ion

s (n

g/m

l)

Mated

G D5-7

G D12-14

G D18-20

Post-partum

Older

605550454035302520151050

DihydrotestosteroneTestosteroneTHPDihydroprogesteroneProgesterone

Diol

trus rats. The increase in plasma progestin concentrationsF. 1. Significant variations in plasma (top) and whole brain (bottom)from dioestrus to pro-oestrus/oestrus were greatest for pro-progesterone (F (10, 44)=3.72, P∏0.001), dihydroprogesterone (F (10,gesterone >3a,5a-THP>DHP. Testosterone was slightly, 44)=4.06, P∏0.001), 3a,5a-THP (F (10, 44)=5.08, P∏0.001), testoster-

and 3a-Diol more saliently, increased from diestrus to pro- one (F (10, 44)=7.84, P∏0.001), dihydrotestosterone (F (10, 44)=2.55,estrus; whereas, DHT was reduced. Notably, plasma concen- P∏0.05), and 3a-Diol (F (10, 44)=3.38, P∏0.001) over hormonal and

reproductive states (n=4 observations per group).trations of 3a,5a-THP and 3a-Diol were increased twofoldfrom non-receptive to receptive rats (Fig. 1, top). As haspreviously been shown, oestradiol and corticosterone levels increased during pregnancy: testosterone was highest in the

third trimester, whereas DHT and 3a-Diol peaked during thewere increased from dioestrus to pro-oestrus and DHEA wasreduced (see Table 1). second trimester. Androgen levels were modestly reduced

post-partum compared to peak pregnancy levels. PlasmaConsistent with these differences in 3a,5a-THP and 3a-Diol in non-receptive and receptive rats, there were significant concentrations of oestradiol and corticosterone were

increased throughout pregnancy, and were precipitouslychanges in these neurosteroids in response to mating. Plasma3a,5a-THP decreased (P∏0.008) and a plasma 3a-Diol reduced post-partum (Table 1). DHEA levels remained stable

during pregnancy and were slightly decreased post-parturitionincreased (P∏0.01) in mated rats compared to receptive,pro-oestrus rats (Fig. 1, top). Alterations in these neurostero- compared to pregnancy levels (Table 1). Interestingly, 3a,5a-

THP levels showed the greatest per cent decrease (343%) ofids were specific; DHEA levels were not significantly changedand alterations in 3a,5a-THP and 3a-Diol levels were not due all progestins from peak levels in pregnancy to post partum,

in contrast, 3a-Diol levels showed a very modest change.to stress, as corticosterone was not different between proestrusand mated animals (Table 1). Overall, older rats had moderate steroid hormone concentra-

tions; plasma concentrations of older animals reflected aver-Progestin and androgen levels were higher during preg-nancy than immediately after, and concentrations in older age oestrous cycle concentrations of younger rats (Fig. 1,

top; Table 1).animals were similar to mean oestrous cycle concentrations.Plasma progestins were highest during the third trimester of Variations in concentrations of progestins, androgens,

DHEA, and corticosterone in whole brain were similar topregnancy; progesterone <DHP<3a,5a-THP were reducedpost parturition from peak pregnancy levels. Androgens those seen in plasma (Fig. 1, bottom; Table 1).

© 1999 Blackwell Science Ltd, Journal of Neuroendocrinology, 11, 839–847

842 Neurosteroids and mating

Experiment 2: Steroid hormone concentrations vary in themidbrain and hypothalamus as a function of reproductive state

Plasma and whole brain concentrations of steroid hormonesrevealed that levels of the neurosteroids, 3a,5a-THP and 3a-Diol were increased in receptive compared to non-receptiveanimals and were altered by mating. To further elucidate thecentral sites involved in neurosteroids’ mediation of repro-ductive behaviours, the midbrain, hypothalamus, cortex, hip-pocampus, and amygdala were examined. No statisticallysignificant differences in progesterone, testosterone, or estra-diol were found among CNS sites. Dihydrotesterone<DHP<corticosterone<DHEA had a tendency to vary bysite. However, site differences in the neurosteroids 3a,5a-THPand 3a-Diol were the most robust (Figs 2, 3; Table 1).

Plasma and whole brain 3a,5a-THP and 3a-Diol concentra-

Oestrus

Met-oestrus

Dioestrus1 & 2

Pro-oestrus

Bra

in c

on

cen

trat

ion

s (n

g/g

)

Mated

G D5-7

G D12-14

G D18-20

Post-partum

Older

605550454035302520151050

Oestrus

Met-oestrus

Dioestrus1 & 2

Pro-oestrus

Bra

in c

on

cen

trat

ion

s (n

g/g

)

Mated

G D5-7

G D12-14

G D18-20

Post-partum

Older

605550454035302520151050

DihydrotestosteroneTestosteroneTHPDihydroprogesteroneProgesterone

Diol

Hypothalamus

Midbrain

F. 2. The top and bottom panels depict midbrain and hypothalamicvariations in steroids. There were no statistically significant differencesin progesterone (F(4,165)=1.79, P∏0.134) or testosterone (F(4,165)=0.901, P∏0.47). Dihydrotestosterone (F (4,165)=2.18, P∏0.08) andDHP (F(4,165)=3.16, P∏0.01) had a tendency to vary by site. Therewere significant differences in 3a,5a-THP (F(4,165)=7.88, P∏0.001) and3a-Diol (F(4,165)=8.39, P∏0.001) in the midbrain and hypothalamus(n=8 observations per group).T

1.M

ean

(±SE

M)

Pla

sma

and

Who

leB

rain

Oes

trad

iol,

Cor

tico

ster

one,

and

DH

EA

;O

estr

adio

lA

cros

sC

NS

Site

sis

Als

oSh

own.

Pla

sma

Who

lebr

ain

Oes

trad

iol

Oes

trad

iol

Cor

tico

ster

one

DH

EA

Oes

trad

iol

Cor

tico

ster

one

DH

EA

Mid

brai

nH

ypot

hala

mus

Am

ygda

laH

ippo

cam

pus

Cor

tex

Con

diti

on(p

g/m

l)(m

g/dl

)(n

g/m

l)(n

g/m

g)(n

g/g)

(ng/

g)(n

g/m

g)(n

g/m

g)(n

g/m

g)(n

g/m

g)(n

g/m

g)

Dio

estr

us1

8.1±

0.6

5.7±

1.6

0.2±

0.1

15.8±

0.1

0.1±

0.1

0.1±

0.1

0.8±

0.4

0.6±

0.4

2.4±

0.4

0.5±

0.4

3.5±

0.4

Dio

estr

us2

16.9±

0.4

6.0±

1.4

0.1±

0.1

23.8±

0.1

0.1±

0.1

0.2±

0.1

0.6±

0.3

0.7±

0.4

3.9±

0.4

1.9±

0.4

6.2±

0.4

Pro

-oes

trus

33.7±

9.9

15.9±

2.3

0.2±

0.1

20.4±

0.1

0.1±

0.1

0.3±

0.1

2.0±

0.4

2.4±

0.4

4.1±

0.4

0.9±

0.3

2.7±

0.9

Oes

trus

20.3±

4.7

10.7±

0.7

0.1±

0.1

19.1±

0.1

0.1±

0.1

0.2±

0.1

2.3±

0.4

1.1±

0.4

2.7±

0.4

1.0±

0.4

3.5±

0.4

Met

oest

rus

15.8±

4.3

11.6±

4.5

0.2±

0.1

17.7±

0.1

0.1±

0.1

0.2±

0.1

0.7±

0.1

8.2±

0.4

0.7±

0.1

1.0±

0.4

1.4±

0.4

Pro

-oes

trus

mat

ed17

.2±

5.8

15.2±

4.6

0.2±

0.1

19.7±

0.1

0.1±

0.1

0.2±

0.1

4.0±

0.4

1.0±

0.2

0.4±

0.2

1.4±

0.4

0.6±

0.4

Fir

sttr

imes

ter

17.3±

2.1

2.5±

0.3

0.2±

0.1

29.3±

0.1

0.1±

0.1

0.2±

0.1

5.1±

1.0

1.4±

0.4

1.9±

0.4

12.7±

2.4

10.6±

0.5

Seco

ndtr

imes

ter

35.9±

7.9

8.6±

1.3

0.1±

0.1

38.9±

0.1

0.1±

0.1

0.2±

0.1

18.0±

0.4

22.7±

0.4

29.6±

0.4

5.5±

0.1

29.6±

0.4

Thi

rdtr

imes

ter

49.0±

6.7

10.8±

1.4

0.2±

0.1

95.6±

0.1

0.1±

0.1

0.3±

0.1

30.9±

2.4

10.4±

0.1

1.0±

0.4

7.3±

0.4

5.6±

0.4

Lac

tati

ng7.

5±4.

01.

9±0.

30.

1±0.

110

.2±

0.1

0.1±

0.1

0.2±

0.1

29.1±

0.3

25.9±

0.4

9.4±

0.4

28.4±

0.4

19.4±

0.4

Age

d15

.9±

2.7

3.2±

1.1

0.2±

0.1

64.6±

0.1

0.1±

0.1

0.2±

0.1

58.4±

0.4

25.6±

0.4

9.0±

0.4

11.4±

0.4

19.0±

0.4

Pla

sma

and

who

lebr

ain

oest

radi

ol(F

(10,

44)=

5.72

,P∏

0.00

1),

DH

EA

(F(1

0,44

)=4.

05,

P∏

0.00

1),

and

CO

RT

(F(1

0,44

)=4.

58,

P∏

0.00

1)

diff

ered

asa

func

tion

ofen

docr

ine/

repr

oduc

tive

stat

us.T

here

wer

eno

stat

isti

cally

sign

ifica

ntdi

ffer

ence

sin

oest

radi

ol(F

(4,1

65)=

0.72

,P∏

0.50

)am

ong

CN

Ssi

tes;

how

ever

;dat

ais

show

nbe

caus

eof

the

influ

ence

ofoe

stra

diol

onpr

oges

tero

nean

dte

stos

tero

nem

etab

olis

m.

Cor

tico

ster

one

(F(4

,165

)=3.

16,

P∏

0.05

)an

dD

HE

A(F

(4,1

65)=

6.49

,P∏

0.01

)in

the

diff

eren

tbr

ain

regi

ons

wer

esi

mila

rto

plas

ma

and

who

lebr

ain

conc

entr

atio

ns(d

ata

not

show

n).

© 1999 Blackwell Science Ltd, Journal of Neuroendocrinology, 11, 839–847

Neurosteroids and mating 843

tions showed the greatest increases during pro-oestrus/oestrus significantly higher in proestrus rats compared to rats exposedto the mating arena<exogenous VCS<mated<exposed to ain the various brain regions compared to dioestrus of the

steroids examined. These changes were most evident in the male (Fig. 4, bottom).midbrain and hypothalamus. Specifically, levels of 3a,5a-THP were increased in midbrain>cortex>hippocampus> Discussionhypothalamus>amygdala in proestrus compared to dioestrusanimals. For 3a-Diol, the greatest increase from dioestrus- The hypothesis that levels of the neurosteroids 3a,5a-THP

and 3a-Diol varied as a function of reproductive state andto pro-oestrus/oestrus concentrations were seen in thehypothalamus>midbrain>hippocampus. There were no sexual experience was supported by the following findings.

First, of the steroids examined, 3a,5a-THP and 3a-Diolapparent changes in the amygdala or cortex (Figs 2, 3).Compared to all other steroids examined in mated versus increased most from diestrus (non-receptive) condition to

proestrus (receptive) condition in plasma and whole brain.pro-oestrus animals, plasma and whole brain 3a,5a-THP and3a-Diol showed the greatest changes. Consistent with changes In plasma and whole brain, 3a,5a-THP and 3a-Diol were

doubled on pro-oestrus compared to dioestrus. 3a,5a-THPin 3a,5a-THP and 3a-Diol across the cycle being greatest inthe midbrain and hypothalamus, central changes in 3a,5a- was highest in the midbrain, and 3a-Diol was increased most

in the hypothalamus of pro-oestrus rats compared to dioes-THP and 3a-Diol, as a result of the mating experience, alsoappeared to be localized to the midbrain and hypothalamus. trus rats. These data suggest that 3a,5a-THP and 3a-Diol are

altered in the plasma, whole brain and in the hypothalamusIn the hypothalamus of proestrus versus mated animals,3a,5a-THP was increased and 3a-Diol was decreased. In and midbrain respectively between stages of the oestrous

cycle, characterized by differences in receptivity. Second,contrast, in the midbrain of mated animals, levels of 3a,5a-THP and 3a-Diol were increased compared to proestrus 3a,5a-THP was reduced and 3a-Diol was increased in mated

as compared to pro-oestrus rats. Mated rats had significantlyrats (Fig. 2).Plasma and whole brain concentrations of 3a,5a-THP and reduced plasma and whole brain concentrations of 3a,5a-

THP, whereas 3a-Diol was significantly increased in plasma3a-Diol were decreased post-partum compared to peak levelsduring pregnancy. Reductions in post-partum 3a,5a-THP and and whole brain of mated compared to pro-oestrus rats. In

the midbrain and hypothalamus, 3a,5a-THP was increased3a-Diol levels were seen in the midbrain and hypothalamus,as well as in other brain areas. Levels of 3a,5a-THP were in response to mating; 3a-Diol was increased in the midbrain

of rats but decreased substantially in the hypothalamus, ofdecreased in hypothalamus>amygdala>cortex>midbrain>hippocampus relative to peak pregnancy values. 3a-Diol mated compared to pro-oestrus rats. This suggests that the

relationship of 3a,5a-THP to 3a-Diol in plasma, whole brain,concentrations were decreased in amygdala>midbrain>hippocampus>hypothalamus>cortex of post-partum com- and/or areas such as the hypothalamus and midbrain may

differ between receptive unmated and mated rats. Third,pared to pregnant animals. Similarly, in older animals 3a,5a-THP was greatest in the hippocampus>cortex> hypothalamic and midbrain 3a,5a-THP and 3a-Diol concen-

trations varied with different mating stimuli. 3a,5a-THP levelsmidbrain>hypothalamus>amygdala. In older animals, 3a-Diol was greatest in the cortex>midbrain>hippocampus> were increased in the hypothalamus and midbrain of mated,

but reduced in response to exposure to the mating arena, theamygdala>hypothalamus (Figs 2, 3).male, or VCS. 3a-Diol concentrations in the midbrain wereincreased in response to mating, VCS, or exposure to theExperiment 3: 3a,5a-THP and 3a-Diol vary in the midbrainmale compared to that seen in the midbrain of pro-oestrusand hypothalamus as a function of environmental stimulirats or to rats exposed to the mating arena. In the hypothal-proximate to matingamus, 3a-Diol was decreased in response to mating, exposureto the mating arena, male, or vaginocervical stimulation. ThisConcentrations of 3a,5a-THP and 3a-Diol were significantly

altered by mating relevant stimuli. Consistent with changes suggests that the different aspects of mating stimuli examinedmay mediate changes in 3a-Diol in the midbrain but notseen in experiment 2, there was an increase in 3a,5a-THP

levels in the midbrain of mated animals compared to proestrus hypothalamus; as well, 3a,5a-THP was not increased in themidbrain or hypothalamus in response to mating relevantanimals. However, 3a,5a-THP levels were reduced in the

midbrain of animals exposed to the mating arena<exposed stimuli. Together these findings suggest that the neurosteroids3a,5a-THP and 3a-Diol vary systemically and in the brain,to a male<exposed to artificial VCS compared to proestrus

animals. 3a-Diol levels were again increased in the midbrain particularly the midbrain and the hypothalamus, in responseto different hormonal states associated with changes in recep-of mated animals, slight increases were also seen in rats

exposed to artificial VCS and rats exposed to a male compared tivity and in response to different mating relevant stimuli.The present findings are consistent with previous researchto proestrus animals. However, 3a-Diol levels showed a

decrease in the midbrain of animals exposed to the mating that has shown endogenous variations in steroids to be afunction of reproductive state, and extend them to showarena (Fig. 4, top).

3a,5a-THP concentrations in the hypothalamus varied in central concentrations of the steroids across various brainregions in different endocrine states. Consistent with previoussimilar fashion to that seen in the midbrain; 3a,5a-THP was

highest in mated rats and reduced in proestrus rats, however, reports, plasma concentrations of progestins (43–46) andandrogens (47–50) increase during pro-oestrus and preg-3a,5a-THP levels were reduced in rats exposed to a

male<exposed to the mating arena<exposed to artificial nancy, and decrease post-partum and in older rats. Thepresent findings demonstrate that whole brain levels tend toVCS compared to proestrus rats. Levels of 3a-Diol were

© 1999 Blackwell Science Ltd, Journal of Neuroendocrinology, 11, 839–847

844 Neurosteroids and mating

Oestrus

Met-oestrus

Dioestrus1 & 2

Pro-oestrus

Bra

in c

on

cen

trat

ion

s (n

g/g

)

Mated

G D5-7

G D12-14

G D18-20

Post-partum

Older

60

55

50

45

40

35

30

25

20

15

10

5

0

Oestrus

Met-oestrus

Dioestrus1 & 2

Pro-oestrus

Bra

in c

on

cen

trat

ion

s (n

g/g

)

Mated

G D5-7

G D12-14

G D18-20

Post-partum

Older

60

55

50

45

40

35

30

25

20

15

10

5

0

Cortex

Hippocampus

Dihydrotestosterone

Testosterone

THP

Dihydroprogesterone

Progesterone

Diol

Oestrus

Met-oestrus

Dioestrus1 & 2

Pro-oestrus

Bra

in c

on

cen

trat

ion

s (n

g/g

)

Mated

G D5-7

G D12-14

G D18-20

Post-partum

Older

60

55

50

45

40

35

30

25

20

15

10

5

0

Amygdala

© 1999 Blackwell Science Ltd, Journal of Neuroendocrinology, 11, 839–847

Neurosteroids and mating 845

Pro-oestrus Arenaonly

ArtificialVCS

Malein

arena

3a,5

a-T

HP

co

nce

ntr

atio

n (

ng

/gra

m)

Mated

10

9

8

7

6

5

4

3

2

1

0

A

BBC

C

D

Pro-oestrus Arenaonly

ArtificialVCS

Malein

arena

3a,5

a-T

HP

co

nce

ntr

atio

n (

ng

/gra

m)

Mated

10

9

8

7

6

5

4

3

2

1

0

A

AB

BAB

C

Pro-oestrus Arenaonly

ArtificialVCS

Malein

arena

Mated

10

9

8

7

6

5

4

3

2

1

0

A

B BB B

Pro-oestrus Arenaonly

ArtificialVCS

Malein

arena

Mated

10

9

8

7

6

5

4

3

2

1

0

AB

C C C

3a-D

iol c

on

cen

trat

ion

(n

g/g

ram

)3a

-Dio

l co

nce

ntr

atio

n (

ng

/gra

m)

Hypothalamus

Midbrain

F. 4. The top and bottom panels depict variations (mean±SEM) in 3a,5a-THP and 3a-Diol in the midbrain and hypothalamus respectively offemales in proestrus (exposed to home cage only), exposed to the mating arena (arena only), given 5–10 intromissions with an eyedropper (artificialVCS), exposed to the sight and smell of the male in the mating arena (male in arena), or after a copulatory mating sequence (mated). Concentrationsof 3a,5a-THP (F(4, 75)=23,70, P∏0.001) and 3a-Diol (F(4, 75)=43.89, P∏0.001) were significantly altered by mating relevant stimuli (n=8observations/group).

fluctuate in parallel with plasma concentrations of each of which can enhance the activity of the 5a-reductase enzyme(51)—was also increased in the midbrain and hypothalamus;the steroids examined. In general, steroids derived primarily

from the gonads (progesterone, testosterone, and oestradiol ) as were the parent compounds for 3a,5a-THP, progesteroneand dihydroprogesterone. The levels of 3a-Diol were muchwere present at similar concentrations in all the brain regions

examined. However, steroids that have ovarian, adrenal lower than that of its parent’s compounds as has previouslybeen demonstrated (14).and/or central sources such as DHP, 3a,5a-THP, DHT, 3a-

Diol, DHEA, and corticosterone were more likely to vary Previous reports have suggested that the peripheral andcentral concentrations of 3a,5a-THP and 3a-Diol have a roleacross brain regions, in particular the neurosteroids 3a,5a-

THP and 3a-Diol varied the most across CNS sites. in mediating reproductive behaviour. Variations in circulatingconcentrations of 3a,5a-THP rise on late pro-oestrus, decrease3a,5a-THP and 3a-Diol differed in the midbrain and hypo-

thalamus from pro-oestrus to oestrus; however, the source of on late-oestrus, and are at their lowest point of the oestrouscycle during dioestrus and early oestrus (23–25). Thesethe variations in 3a,5a-THP and 3a-Diol was not determined

in the present study. Both of these steroids can be metabolized variations may underlie correlating changes in lordosis beha-viour. Local application of 3a,5a-THP to the hypothalamusfrom peripheral or central progesterone/dihydroprogesterone

and testosterone/dihydrotestosterone, as well as being pro- (5, 6) or midbrain (52) of oestrogen-primed rodents facilitatesreceptivity. Furthermore, blocking progesterone’s metabolismduced de novo by glial cells in the brain. Notably, oestradiol—

F. 3. The top, middle, and bottom panels respectively depict amygdala, hippocampus, and cortex variations in steroids as a function of hormonaland reproductive state (n=8 observations/group).

© 1999 Blackwell Science Ltd, Journal of Neuroendocrinology, 11, 839–847

846 Neurosteroids and mating

to 3a,5a-THP or synthesis of 3a,5a-THP in the midbrain novel mating arena did not. These findings suggest thatincreases in 3a-Diol in the midbrain may underlie the durationdisrupts progesterone-facilitated sexual receptivity (8, 9). The

present findings extend the past research to demonstrate that or termination of estrus in rats.In summary, concentrations of the neurosteroids 3a,5a-endogenous 3a,5a-THP is increased in areas of the brain

associated with progesterone-facilitated lordosis, e.g. 3a,5a- THP and 3a-Diol vary both in the hypothalamus and themidbrain of females as a function of endocrine states and/orTHP levels in the midbrain of mated rats is increased over

that of pro-oestrus rats which is greater than that in dioestrus mating relevant stimuli. 3a,5a-THP and 3a-Diol have verylow affinities for intracellular progestin (4) and androgenrats. In the hypothalamus, 3a,5a-THP is greater in mated

than in pro-oestrus and dioestrus rats. Hence, these data receptors (18–19). Within the physiological concentrationranges 3a,5a-THP and 3a-Diol were determined to be in, insupport a role for 3a,5a-THP in reproductive behaviour.

3a-Diol is known to have a suppressive effect on the onset the present study, they would certainly be unable to exert‘genomic’ actions through intracellular steroid receptors.of maturation (26, 53). However, following first ovulation,

3a-Diol’s role is less clear, as the ovary shifts from preferen- Alternatively, 3a,5a-THP and 3a-Diol could have ‘non-genomic’ actions by altering GBRs; the concentrations rangestially metabolizing 5a-reduced androgens to progestins

(54–55). Although the mature gonad continues to produce presently determined for 3a,5a-THP and 3a-Diol can enhanceGABA-stimulated chloride influx in GBRs (15, 16, 21, 22).3a-Diol, concentrations are lower than before puberty

(54–55). After puberty, 3a-Diol concentrations vary over the Together these findings suggest that 3a,5a-THP and 3a-Diolmay influence reproductive behavior in part through non-oestrous cycle, which suggest that this neurosteroid has a

functional role in adulthood (26–29). As with 3a,5a-THP, genomic actions at GBRs. Further research is warranted todetermine the functional significance of these variations; asplasma 3a-Diol peaks on proestrus, declines on oestrus, and

reaches basal levels during dioestrus. Administration of 3a- well as the mechanism of action by which changes in 3a,5a-THP and 3a-Diol, as well as their parent compounds, mayDiol, producing circulating concentrations similar to that

seen on pro-oestrus, facilitates lordosis of EB and progester- mediate reproductive behaviour.one-primed rats. In contrast, administration of lower dosagesof 3a-Diol, producing circulating concentrations similar to

Acknowledgmentsthat seen on dioestrus, inhibits lordosis of EB and progesterone-primed rats (16). 3a-Diol concentrations are increased in This research was supported by CAREER grants from the National Scienceresponse to naturalistic, paced mating, which is associated Foundation (IBN 9514463 and IBN 98-96263).with a briefer period of oestrus (30, 56, 57). As well,application of 3a-Diol to the VMH or POA can have an Accepted 9 March 1999inhibitory or facilitatory effect on lordosis, depending uponthe concentration of systemic steroids concurrently adminis-tered (17). These data suggest that elevated concentrations Referencesof 3a-Diol on the evening of pro-oestrus, may facilitate

1 Rubin BS, Barfield RJ. Induction of estrous behavior in ovariectomizedprogestin-induced receptivity, while basal levels across therats by sequential replacement of estrogen and progesterone to the

cycle may tonically inhibit receptivity. Our present findings ventromedial hypothalamus. Neuroendocrinology 1983; 37: 218–224.that plasma, whole brain, and midbrain concentrations of 2 McEwen BS, Coirini H, Schumacher M. Steroid effects on neuronal

activity: when is the genome involved. In: Chadwick D, Widdows K,3a-Diol are higher in mated rats than proestrus or diestruseds. Steroids and Neuronal Activity. Chichester: John Wiley & Sons,rats, and that concentrations of 3a-Diol are decreased in the1990: 3–21.hypothalamus of mated rats but increased in the midbrain, 3 Schumacher M, Coirini H, Frankfurt M, McEwen BS. The neuronal

suggest that oestrus cessation may be mediated by relative membrane: a target for steroid hormone action. In: Balthazart J, ed.Hormones, Brain and Behavior in Vertebrates. Basel: Karger, 1990:concentrations of 3a-Diol in these brain regions, which may91–115.be altered in response to mating relevant stimuli.

4 Iswari S, Colas AE, Karavolas HJ. Binding of 5a-dihydroprogesteroneNeurosteroid levels can be altered as a function of environ-and other progestins to female rat anterior pituitary nuclear extracts.

mental stimuli (58). The most robust increases in neurostero- Steroids 1986; 47: 189–203.ids that have been demonstrated result from exposure to 5 Glaser JH, Etgen AM, Barfield RJ. Intrahypothalamic effects of progestin

agonists on estrous behavior and progestin receptor binding. Physiolstressful stimuli (59); however, more subtle stimuli mayBehav 1985; 34: 871–878.influence neurosteroid concentrations (58). The present find-

6 Beyer C, Gonzalez-Mariscal G, Eguibar JR, Gomora P. Lordosis facilita-ings demonstrate that in females, particular aspects of mating tion in estrogen primed rats by intrabrain injection of pregnanes.can alter central neurosteroid concentrations, independent of Pharmacol Biochem Behav 1989; 31: 919–926.

7 DeBold JF, Frye CA. Progesterone and the neural mechanisms ofperipheral circulating concentration, or concentrations in thehamster sexual behaviour. Psychoneuroendocrinology 1994; 19: 563–579.brain or blood of the parent compounds. The particular

8 Frye CA, Leadbetter EA. 5a-reduced progesterone metabolites are essen-aspects of mating stimuli that may mediate changes in thetial in hamster VTA for sexual receptivity. Life Sci 1994; 54: 653–659.

concentrations of 3a,5a-THP in the midbrain or hypothal- 9 Vongher J, Frye CA. Finasteride and epostane reduce lordosis whenamus were not elucidated. 3a-Diol levels were low in the applied to the midbrain of rats and hamsters. Soc Behav Neuroendocrinol

1997; 29: 178.hypothalamus of all rats, irrespective of mating, exposure to10 Dohanich GP, Clemens LG. Inhibition of estrogen-activated sexualartificial VCS, the sight or smell of a male, or the novel

behavior by androgens. Horm Behav 1983; 17: 366–373.mating arena. In contrast, exposure to the sight and smell of 11 Erskine MS. Effects of an anti-androgen and 5a-reductase inhibitors ona male without tactile contact, artificial VCS, or mating, estrus duration in the cycling female rat. Physiol Behav 1983; 20: 519–524.

12 Erskine MS. Effect of 5a-dihydrostestosterone and flutamide on theincreased 3a-Diol in the midbrain; while, exposure to the

© 1999 Blackwell Science Ltd, Journal of Neuroendocrinology, 11, 839–847

Neurosteroids and mating 847

facilitation of lordosis by LHRH and naloxone in estrogen-primed female 35 Erskine MS, Kornberg E. Stress and ACTH increase circulating concen-trations of 3a-androstanediol in female rats. Life Sci 1992; 51: 2065–2071.rats. Physiol Behav 1989; 43: 753–759.

36 Kavaliers M, Wiebe JP, Galea LAM. Male preference for the odors of13 Erskine MS, Kornberg E, Cherry JA. Paced copulation in rats: effectsestrous female mice is enhanced by the neurosteroid 3a-hydroxy-of intromission frequency and duration on luteal activation and estrus4-pregnen-20-one (3a-HP). Brain Res 1994; 646: 140–144.length. Physiol Behav 1989; 45: 33–39.

37 Corpechot C, Leclerc P, Baulieu E, Brazeau P. Neurosteroids: regulatory14 Erskine MS, Hippensteil M, Kornberg E. Metabolism of dihydrotesto-mechanisms in male rat brain during heterosexual exposure. Steroidssterone to 3a-androstanediol in brain and plasma: effect on behavioral1985; 45: 229–234.activity in female rats. J Endocrinol 1992; 134: 183–195.

38 Corpechot C, Robel P, Axelson M, Sjovall J, Baulieu EE.15 Frye CA, Duncan JE, Basham M, Erskine MS. Behavioral effects of 3a-Characterization and measurement of dehydroepiandrosterone sulfate inAndrostanediol. II: Hypothalamic and preoptic area actions via arat brain. Proc Natl Acad Sci USA 1981; 78: 4704–4707.GABAergic mechanism. Behav Brain Res 1996; 79: 119–130.

39 Long JA, Evans HM. Oestrus cycle in the rat and its associated16 Frye CA, van Keuran KR, Erskine MS. Behavioral effects of 3a-phenomena. Memoirs Univ Cal 1992; 6: 1–146.Androstanediol: I. Modulation of sexual receptivity and promotion of

40 Glowinski J, Iversen LL. Regional studies of catecholamines in the ratGABA-stimulated chloride flux. Behav Brain Res 1996; 79: 109–118.

brain-I. J Neurochem 1966; 13: 655–696.17 Frye CA, Van Keuran KR, Rao PN, Erskine MS. Progesterone and 3a- 41 Frye CA, Scalise TJ, Bayon LE. Finasteride blocks the reduction in ictal

Androstanediol conjugated to bovine serum albumin affects estrous activity produced by exogenous estrous cyclicity. J Neuroendocrinol 1998;behaviour when applied to the MBH and POA. Behav Neurosci 1996; 10: 291–296.96: 603–612. 42 Purdy RH, Moore PH, Narasimha Roa P, Hagino N, Yamaguchi T,

18 Cunningham GR, Tindall DJ, Means AR. Differences in steroid specifi- Schmidt P, Rubinow DR, Morrow AL, Paul SM. Radioimmunoassaycity for rat androgen binding protein and the cytoplasmic receptor. of 3a-hydroxy-5a-pregnan-20-one in rat and human plasma. SteroidsSteroids 1979; 33: 261–276. 1990; 55: 290–296.

19 Verhoeven G, Heyns W, De Moor P. Interconversion between 17b- 43 Barraclough CA, Collu R, Massa R, Martini L. Temporal interrela-hydroxy-5a-androstan-3-one (5a-dihydrostestosterone) and 5a-andros- tionships between plasma LH, ovarian secretion rates and peripheraltane-3a,17b-diol: Tissue specificity and role of the microsomal NAD: plasma progestin concentrations in the rat: effects of nembutal and

exogenous gonadotropins. Endocrinology 1971; 88: 1437–1447.3-hydroxysteroid oxidoreductase. J Steroid Biochem 1977; 8: 731–733.44 Butcher RL, Collins WE, Fugo NW. Plasma concentrations of LH,20 Mellon SH. Neurosteroids: biochemistry, modes of action, and clinical

FSH, prolactin, progesterone and estradiol throughout the 4-day estrousrelevance. J Clin Endocrinol Metab 1994; 78: 1003–1008.cycle of the rat. Endocrinology 1974; 94: 1704–1708.21 Majewska MD, Harrison NL, Schwartz RD, Barker JL, Paul SM.

45 Horikoshi H, Suzuki Y. On circulating sex steroids during the estrousSteroid hormone metabolites are barbiturate-like modulators of thecycle and the early pseudopregnancy in the rat with special reference toGABA receptor. Science 1986; 232: 1004–1007.its luteal activation. Endocrinol Jap 1974; 21: 69–79.22 Gee KW. Steroid modulation of the GABA/benzodiazepine receptor-

46 Sanyal MK. Secretion of progesterone during gestation in the rat.linked chloride ionophore. Mol Neurobiol 1988; 2: 291–317.J Endocrinol 1978; 79: 179–190.23 Holzbauer M. Physiological aspects of steroids with anesthetic properties.

47 Belanger A, Cusan L, Barden L, Dupont A. Ovarian progestins, andro-Med Biol 1976; 54: 227–242.gens and estrogen throughout the 4-day estrous cycle in the rat. Biol24 Ichikawa S, Sawada T, Nakamura Y, Morioka H. Ovarian secretion ofReprod 1981; 24: 591–596.

pregnane compounds during the estrous cycle and pregnancy in rats.48 Legrand C, Marie J, Maltier JP. Testosterone, dihydrotestosterone,

Endocrinology 1974; 94: 1615–1620. androstenedione and dehydroepiandrosterone concentrations in placen-25 Paul SM, Purdy RH. Neuroactive steroids. FASEB Journal 1992; 6: tae, ovaries and plasma of the rat in late pregnancy. Acta Endocrinol

2311–2322. 1984; 105: 119–125.26 Ojeda SR, Katz KH, Costa ME, Advis JP. Effect of experimental 49 Dunlap KD, Sridaran R. Administration of gonadotropin-releasing

alterations in serum levels of 5a-androstane-3a, 17b-diol on the timing hormone to pregnant rats inhibits postpartum sexual behavior. Pharmacolof puberty in the female rat. Neuroendocrinology 1984; 39: 19–24. Biochem Behav 1989; 31: 725–728.

27 Belanger A, Cusan L, Caron S, Barden N, Dupont A. Ovarian progestins, 50 Ghanadian R, Lewis JG, Chisholm GD. Serum testosterone and dihy-androgens and estrogen throughout the 4-day estrous cycle of the rat. drotestosterone changes with age in rat. Steroids 1975; 25: 753–762.Horm Behav 1981; 7: 87–104. 51 Cheng Y, Karavolas J. Conversion of progesterone to 5a-

28 Meijs-Roelofs HMA, Kramer P, Cappellen WAV, Woutersen PJA. pregnan-3,20-dione and 3a-hydroxy-5a-pregnan-20-one by rat medialbasal hypothalamus and the effects of estradiol stage of estrous cycle inChanges in serum concentration and ovarian content of 5a-androstane-conversion. Endocrinol 1973; 93: 1157–1162.3a,17b-diol in the female rat approaching first ovulation. Biol Reprod

52 Frye CA, Gardiner SG. Progestins can have a membrane-mediated action1985; 32: 301–308.in rat midbrain for facilitation of sexual receptivity. Horm Behav 1996;29 Meijs-Roelofs HMA, van Kramer P, Gribling-Hegge L, Woute PJA.30: 682–691.Periovulatory changes in serum concentration and of 5a-androstane-

53 Ruf KB. On the significance of 5a-androstane-3a,17b-diol in the peripub-3a,17b-diol in the androgenized rat. Biol Reprod 1986; 35: 890–896.ertal female rat. J Steroid Biochem 1983; 19: 887–890.30 Frye CA, McCormick CM, Coopersmith C, Erskine MS. Effects of

54 Eckstein B, Ravid R. On the mechanisms of the onset of puberty:paced and non-paced mating stimulation on plasma progesterone, 3a-identification and pattern of 5a-androstane-3a,17b-diol and its 3 epimerdiol and corticosterone. Psychoneuroendocrinology 1996; 21: 431–439.in peripheral blood in immature rats. Endocrinol 1974; 94: 224–229.31 Barbaccia M, Roscetti G, Bolacchi F, Concas A, Mostallino MC, Purdy

55 Eckstein B, Mechoulam R, Berstein SH. Identification of 5a-androstane-RH, Biggio G. Stress-induced increase in brain neuroactive steroids:3a,17b-diol as a principal metabolite of pregnenolone in rat ovary at the

antagonism by abecarnil. Pharmacol Biochem Behav 1996; 54: 205–210. onset of puberty. Nature 1970; 228: 866–868.32 Dazzi L, Sanna A, Cagetti E, Concas A, Biggio G. Inhibition by the 56 Erskine MS. Serum 5a-Androstane-3a,17b-diol increases in response to

neurosteroid allopregnanolone of basal and stress-induced acetylcholine paced coital stimulation in cycling female rats. Biol Reprod 1987; 37:release in the brain of freely moving rats. Brain Res 1996; 710: 275–280. 1139–1148.

33 Guo A-L, Petraglia F, Criscuolo M, Ficarra G, Nappi RE, Palumbo 57 Erskine MS. Effects of paced coital stimulation on estrus duration inMA, Trentini GP, Purdy RH, Genazzani AR. Evidence for a role of intact cycling rats and ovariectomized and ovariectomized-neurosteroids in modulation of diurnal changes and acute stress-induced adrenalectomized hormone-primed rats. Biol Reprod 1987; 37: 1139–1148.corticosterone secretion in rats. Gynecol Endocrinol 1995; 9: 1–7. 58 Baulieu EE. Neurosteroids: Of the nervous system, by the nervous

34 Motzo C, Porceddu ML, Maira G, Flore G, Concas A, Dazzi L, system, for the nervous system. Rec Prog Horm Res 1997; 52: 1–33.Biggio G. Inhibition of basal and stress-induced dopamine release in the 59 Purdy RH, Morrow AL, Moore PH, Paul SM. Stress-induced elevationscerebral cortex and nucleus accumbens of freely moving rats by the of c-aminobutyric acid type-A receptor-active steroids in the rat brain.

Proc Nat Acad Sci USA 1991; 88: 4553–4557.neurosteroid allopregnanolone. J Psychopharmacol 1996; 10: 266–272.

© 1999 Blackwell Science Ltd, Journal of Neuroendocrinology, 11, 839–847