Alterations in neonatal neurosteroids affect exploration during adolescence and prepulse inhibition in adulthood

Psychoneuroendocrinology (2010) 35, 525—535

a v a i l a b l e a t w w w. s c i e n c e d i r e c t . c o m

j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / p s y n e u e n

Alterations in neonatal neurosteroids affect exploration during adolescence and prepulse inhibition in adulthood `s * So `nia Darbra, Marc Pallare `ncies de la Salut, Institut de Neurocie `ncies, Departament de Psicobiologia i Metodologia en Cie Universitat Auto `noma de Barcelona, 08193 Bellaterra, Barcelona, Spain Received 5 May 2009; received in revised form 31 July 2009; accepted 27 August 2009

KEYWORDS Neurosteroids; Allopregnanolone; Neurodevelopment; Prepulse inhibition; Novelty exploration; Midazolam

Summary Allopregnanolone (AlloP) is a neurosteroid that plays an important role during neural development. Alterations of endogenous neonatal allopregnanolone levels alter the localisation and function of GABA neurons in the adult brain and affect behaviour in adulthood. We have carried out research into the effects of an increase (AlloP administration) or a decrease (administration of finasteride, inhibitor of the AlloP synthesis) of neonatal AlloP levels during the fifth to ninth postnatal days in male Wistar rats on the novelty exploration (Boissier test) at adolescent ages (40 and 60 days old), and on the prepulse inhibition achievement in adulthood (85 days). We also investigated the role of a GABAA modulator (midazolam, 1, 1.75 or 2.5 mg/kg body weight) in the long-lasting behavioural changes in adulthood (85 days). Results indicate that neonatal finasteride decreases both novelty-exploration (head-dipping and locomotion) and anxiety-relevant scores (the distance travelled in and the number of entries into the central zone) at adolescent age, along with a reduction in body weight and general locomotion. Also, neonatal AlloP administration decreases prepulse inhibition in adulthood. Prepulse inhibition disruption was only partially reproduced decreasing the neonatal AlloP levels by means of finasteride administration. Although there was no interaction between neonatal neurosteroid manipulation and adult benzodiazepine treatments, the effects of midazolam were dosedependent: the lowest dose of midazolam increased whereas the highest disrupted the expected progressive reduction of the startle response (and the consequent improvement of the PPI percentage) after the gradual increase in prepulse intensity. Reduced prepulse inhibition of startle provides evidence of deficient sensorimotor gating in several disorders, including schizophrenia. Alterations of AlloP levels during maturation could partly explain the inter-individual differences shown by adult subjects in response to novelty (exploration) and in the sensorimotor gating and prepulse inhibition. Also, abrupt changes in neonatal levels of AlloP could be related to a susceptibility to neurodevelopmental disorders. # 2009 Elsevier Ltd. All rights reserved.

* Corresponding author. Tel.: +34 93 581 25 42; fax: +34 93 581 20 01. E-mail address: [email protected] (M. Pallare `s). 0306-4530/$ — see front matter # 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.psyneuen.2009.08.020

526

1. Introduction Neurosteroids are a subclass of steroids that can be synthesised de novo in the brain (Baulieu, 1981). Under physiological conditions, neurons are exposed to neurosteroids for long periods of time, but also to changes in neurosteroid levels related to stress, pregnancy, menstrual cycle or menopause (Mellon and Griffin, 2002; Frye et al., 2006). Changes in the central or peripheral production of progesterone and the consequent fluctuations in central allopregnanolone (AlloP) concentrations could contribute to the regulation of the GABAA activity and also to the regulation of emotional states (Mellon and Griffin, 2002; Follesa et al., 2004). In this way, anxiety disorders, classically related to GABAA function are also candidates to be mediated by alterations in neurosteroid concentrations (Engel and Grant, 2001; Reddy, 2006). It is important to note that the neurosteroids acting as positive GABAA receptor modulators, such as AlloP, show anxiolytic (Lambert et al., 1995; Akwa et al., 1999; Martin-Garcia and Pallares, 2005a) and anticonvulsant (Lambert et al., 1995; Finn et al., 2004; Martin-Garcia and Pallares, 2005b) properties. Recent findings indicate that neurosteroids act as important factors during brain development (Mameli et al., 2005): (a) the enzymes necessary for neurosteroidogenesis are expressed in the immature brain; (b) in vitro treatment of neuronal cells with neurosteroids produces trophic effects; (c) in vivo or in vitro treatments with progesterone, the main AlloP precursor, induce dendritic outgrowth in Purkinje cells; (d) exposure of rat pups to AlloP alters the distribution of interneurons in the adult prefrontal cortex; (e) AlloP promotes the establishment of neuronal circuitry and supports the survival of developing neurons. In the rat brain, the cortical levels of AlloP show a first prenatal peak followed by low levels during birth, and a second peak in the second week of life (Grobin et al., 2003). A single AlloP administration (10 mg/kg) on the fifth postnatal day altered the localisation and function of prefrontal (Grobin et al., 2003) and dorsal thalamic (Gizerian et al., 2004) GABAergic neurons in the adult rat brain, data which is consistent with a change in the distribution of a set of interneurons in response to neonatal neurosteroid exposure. Perinatal neurosteroid administration can also alter striatal and cortical dopaminergic activity (Muneoka and Takigawa, 2002; Muneoka et al., 2003), and the normal development of the hippocampus (Cooper et al., 1999; Mtchedlishvili et al., 2003; Shirayama et al., 2005; Mellon, 2007). At a behavioural level, it has been reported that acute neonatal AlloP administration (fifth postnatal day) increases cocaine-induced motor activity and decreases prepulse inhibition (PPI) in adulthood, effects that were partially reverted by clozapine administration (Gizerian et al., 2006). Recent experiments carried out in our laboratory have shown that: (1) sub-chronic neonatal administration (from the fifth to the ninth day of life) of finasteride, an inhibitor of AlloP synthesis (Azzolina et al., 1997), induces an anxiogenic-like profile in the elevated plus maze and deteriorates passive avoidance in adulthood after intrahippocampal neurosteroid administration (Martin-Garcia et al., 2008); (2) acute neonatal (fifth postnatal day) administration of AlloP increases novelty-directed locomotion in adulthood measured in the open field and decreases the anxiolytic effects of the benzodiazepine lorazepam measured in the elevated plus maze test (Darbra and Pallare `s,

S. Darbra, M. Pallare `s 2009). All these results mentioned above indicate that neurosteroid levels are important for the correct development of the brain. The mechanisms by which neurosteroids produce these effects, however, are not fully understood, and it has yet to be determined whether endogenous neurosteroids have any physiological and/or pathophysiological roles in neurodevelopment (Mameli et al., 2005). In this sense, some neurosteroids can act as retrograde messengers that induce plasticity in immature synapses during development (Mameli et al., 2005). Early life exposure to other positive modulators of GABAergic function such as diazepam, a benzodiazepine agonist, leads to distinct changes in GABA-mediated functions in adulthood, also including altered locomotion and exploration (Gruen et al., 1990; Kellogg, 1999). The present study focuses on the long-lasting effects of developmentally altered AlloP levels on exploratory adolescent behaviour and PPI in adulthood. In rodents, the term ‘adolescence’ covers the whole postnatal period ranging from weaning (pnd 21) to adulthood (pnd 60), that is to no longer be considered infancy, but not yet adulthood (Laviola et al., 2003). Furthermore, to provide a more detailed characterisation (Laviola et al., 2003; Adriani et al., 2004; Adriani and Laviola, 2004), an adolescent rodent has been classified by the use of three age-intervals, namely early adolescence (prepubescent or juvenile, postnatal (pnd) 21-to-34), middle adolescence (periadolescent, pnd 34-to-46), and late adolescence (young adult, pnd 46-to-60). The validity of such an animal model for the purpose of comparison or extrapolation to the human case has been recently endorsed by Spear (2000). The present study was designed to characterise the spontaneous behaviour expressed by rats at different phases of adolescence (mid-adolescent (pnd 40) and post-adolescent (pnd 60)) and the achievement of the prepulse inhibition as adults. Thus, the same subjects were behaviourally evaluated at different ages. We also investigated the role of a GABAA modulator (midazolam) in the long-lasting behavioural changes in adulthood. For this purpose we have sub-chronically (between pnd 5 and pnd 9) administered: (1) AlloP; (2) the inhibitor of the AlloP synthesis finasteride (Azzolina et al., 1997; Mukai et al., 2008); (3) Control solution. To measure adolescent behaviour we have chosen the Boissier test that involves exploration in a situation of novelty. To modulate GABAA receptors in adulthood we have achieved a dose—response curve with three doses of midazolam, a benzodiazepine full agonist that acts like a non-selective (i.e. alpha1, alpha2, alpha3 and alpha5) receptor subtype modulator (Iversen, 2004). The novelty exploration has not been measured in adult animals in order not to interfere with PPI measurements. Also, PPI, which is tested after midazolam injections, was performed only in adult animals in order to avoid (1) test repetition and possible learning or habituation effects and (2) benzodiazepine injections in adolescence that could have altered the normal adult behaviour.

2. Methods 2.1. Animals The subjects were 111 male Wistar rats derived from ten pairings raised in an in-house colony (Laboratori de Psicobiologia, Universitat Auto `noma de Barcelona) and

527

8 10 12 8 10 11 8 9 9 32 37 42

Vehicle Total

7 4

1

10

2

Adult drug

MID 1.0

MID 75

MID 2.50

allowed food and water ad libitum. The rats were housed in a temperature-controlled animal room (22—24 8C) on a 12-h light/dark cycle. Experimental sessions were run during the light portion of the cycle (lights on at 0800 h). The male breeders were separated from the females after 48 h. Pregnant females were closely monitored, and on the day of birth (called day 0), mothers were removed from the cage and litters were culled to 10 pups. In order to avoid any cohort effects, each litter of the same colony was assigned to different neonatal treatment groups (AlloP, finasteride or Control solution, see Table 1 for a detailed neonatal treatment assignment according pairs and litters). After weaning (pnd 21), males were separated into groups of brothers (with a maximum of five subjects per cage), and females were killed. All animals were obtained, housed, and killed in accordance with protocols approved by the Animal Care and Use Committee in the Autonomous University of Barcelona and the Department of the Environment of the Generalitat de Catalunya (Regional Government), and with guidelines approved by the European Council Directive (86/609/ECC) for the Care and Use of Laboratory Animals.

8 8 10

Neonatal neurosteroids levels on adolescent and adult behavior

5 7

1

6

2 6

7

1

4 4 8 —

5 4 5 5 8

1

4

Litter

AlloP Finasteride Control

7

2

1

2

1

2

8

1

2

1

2

7

1

2

1

2

8 7 6 5 4 3 2 1 Pair

A square wooden arena (58 cm  58 cm  58 cm) with 16 equidistant holes (5 cm in diameter) was used for the Boissier test (adapted from Boissier and Simon, 1967). The apparatus was situated in a room lit by a bright light (300 lx mean). This test measures activity and provides a relatively reliable measurement of stimulus-directed exploratory behaviour (File and Wardill, 1975). It was administered for 5 min and was evaluated by means of an activity monitoring system (SMART, Letica, Barcelona, Spain). This system is based on the automated analysis of real-time video-images, recorded by a video camera which is suspended from the ceiling over the arena. The distance moved

Neonatal treatment assignment and final composition of the experimental groups.

2.3. Boissier exploration test

Table 1

Pups (males and females) were SC injected with the 5alphareductase inhibitor finasteride (50 mg/kg, n = 37), AlloP (10 mg/kg, s.c., n = 32), or Control solution (n = 42), once per day from the fifth to the ninth day after birth (pnd 5 to pnd P9). All pups were injected in order to avoid possible differences in maternal care. All drugs were dissolved in 0.9% NaCl by sonication for 10 min and suspended in 20% b-cyclodextrin. Like the Control solution, 20% b-cyclodextrin dissolved in 0.9% NaCl was used. The injection volume was 0.1 ml/10 g body weight. After injections the pups were immediately returned to the home cage with their mother. For injections, each litter was removed from the dam at once and was returned to their home cage in less than 12 min. The dose of AlloP was chosen based on previous experiments (Grobin et al., 2003; Martin-Garcia et al., 2008; Darbra and Pallare `s, 2009). On the day of behavioural testing in adulthood (pnd 85), animals were randomly injected (i.p.) with midazolam (MID, 1, 1.75 or 2.5 mg/kg) or vehicle (VEH, 0.9% NaCl) 20 min before testing. MID was obtained from Roche (Dormicum1). The rest of the chemicals were obtained from Sigma (Deisenhofen, Germany). The final experimental group size is indicated in Table 1.

9

2

2.2. Administration of drugs

528

S. Darbra, M. Pallare `s

was recorded for the total arena, as well as a virtual 29 cm 29 cm centre zone. In addition, the number of entries into, and the time spent in, the centre zone was also measured as locomotor activity. Moreover, the number of head-dips (into the holes up to the eye line) was recorded as an exploratory measure. After each trial, the apparatus was cleaned with a water solution containing ethanol (5%, v/v) in order to prevent any olfactory-induced behavioural modifications. Rats were tested individually at pnd 40 and pnd 60, at about 0900—1100 h.

2.4. Prepulse inhibition of the acoustic startle response PPI of the acoustic startle response was tested when animals were 85 days old in a StartFear system (Panlab, Cornella `, Barcelona, Spain) that allows the signal generated by the animal movement to be recorded and analysed through a high sensitivity weight transducer. The analogical signal is transmitted to the Startle software module (Panlab, Cornella `, Barcelona, Spain) through the load cell unit for recording purposes and later analysis in terms of startle response characterisation. An additional interface associated with corresponding hardware allows the stimuli (sounds) to be controlled from the Startle software module. The experimental chamber is made with black methacrylate walls and a transparent front door (25 mm  25 mm  25 mm). Animals were held in a Plexiglas cylinder (20 cm in length  6.5 cm internal diameter) mounted on a frame and held in position on a base unit by two plastic screws. Movement of the rat within the cylinder was detected and transduced by a piezoelectric accelerometer attached below the frame. A loudspeaker mounted 18.5 cm above the cylinder provided background white noise, acoustic noise bursts and acoustic prepulses. The whole apparatus was housed in a ventilated, sound-attenuated enclosure (67 cm  53 cm  55 cm) which was diffusely illuminated. Throughout the startle session a background level of 70 dB was maintained. Presentation of acoustic pulse and prepulse stimuli were controlled by the Startle software and interface system, which also digitised, rectified and recorded the responses from the accelerometer. For calibration purposes, the loudspeaker was tested daily with a digital sound level meter (2240, Bru ¨el & Kjaer Sound & Vib, Naerum, Denmark) with the microphone placed inside the Plexiglas cylinder. The startle session started with a 5-min habituation period in the startle chamber, followed by 10 blocks of five trials to measure PPI. Each block consisted of one startle trial (120 dB, 20 ms broad band burst, which was delivered to measure basal startle responsiveness), one no-stimulus condition, and three different prepulse—startle pairings administered pseudo randomly. In these pairings the prepulse was 3, 5, or 10 dB above background. These prepulses were always 20 ms broadband burst and given 100 ms before the startle pulse. The interval between two trials was between 10 and 20 s. The startle amplitude was calculated as the mean of 10 delivered startle trials. The degree of PPI (in percentage) was calculated according to the formula:

100

mean of all startle amplitudes on prepulse trials  100 basal startle amplitude

2.5. Statistical analysis We used the STATISTICA package (StatSoft, Tulsa, USA) for data analyses. The normality of the data was assessed by means of the Kolmogorov—Smirnov test. Data from the Boissier test across time were analysed using a mixed analysis of variance (ANOVA) with NST (neonatal neurosteroid administration, three levels: AlloP, finasteride or control) as the between-subject factor and AGE (two levels: 40 and 60 days) as the within-subject factor for each measure recorded. Data from the PPI test were analysed using a two-way analysis of variance (ANOVA) for repeated measures with NST and DRUG (four levels: VEH and three doses of MID) as independent variables, and prepulse intensity (PULSE, three levels: 3, 5 and 10 dB above background) as repeated measure, followed by separated ANOVA for repeated measures or by the Newman—Keuls post hoc test when appropriated. In order to ascertain that the NST effects were not influenced by expressing the data as a percentage of the PPI, data were also analysed as absolute startle values, according to previous studies (Swerdlow et al., 2000, 2004; van den Buuse, 2003; Gogos and van den Buuse, 2007). The basal startle amplitude was analysed using a two-way ANOVA with NST and DRUG as independent variables. Moreover, body weight development was analysed by means of a mixed analysis of variance with NST as the between-subject factor and AGE (40, 60 and 85 days) as the within-subject factor. The criterion for significance was set at P < 0.05.

3. Results 3.1. Body weight The ANOVA of body weight data revealed significant main effects of NST [F(2,110) = 3.002, P = 0.05] and AGE [F(2,220) = 6054.9, P < 0.001]. No interaction effect between NST and AGE was found. Globally, finasteride-treated animals tended to weigh less than Control animals (N-K: P = 0.058). However, unifactorial analyses of variance showed that finasteride-treated animals differed at statistical level only at 40 days (see Table 2).

3.2. Exploration activity in Boissier test A NST effect was found in the total distance moved in the Boissier test [F(2,108) = 10.164, P < 0.001]. Compared to Controls and AlloP, finasteride-treated rats showed a decrease in locomotor activity (N-K: P < 0.01 and

Table 2 Body weight development. *P < 0.05 vs. Control and vs. AlloP-treated animals at 40 old days. Neonatal treatment

Age 40

60

90

AlloP 174.03  2.44 308.14  2.91 389.17  3.37 Control 174.84  3.13 313.95  4.03 401.62  5.76 Finasteride 164.75  2.06 * 304.40  3.04 389.27  4.15 Body weight (g) development (mean  sem).

Neonatal neurosteroids levels on adolescent and adult behavior

529

Figure 1 Neonatal Finasteride effects on activity and exploration in the Boissier test. (A) Activity expressed as distance travelled in total arena ***P < 0.001 vs. Control and +++P < 0.001 vs. AlloP-treated animals. (B) Activity expressed as number of entries in the central zone +P < 0.05 vs. AlloP-treated animals and **P < 0.01 vs. Control. (C) Activity expressed as distance travelled in the outer portion of the apparatus ++P < 0.01 vs. AlloP-treated animals. (D) Exploration measured as head-dips ***P < 0.001 vs. Control and +++ P < 0.001 vs. AlloP-treated animals. Solid bars: 40 old days; open bars: 60 old days.

P < 0.001, respectively, see Fig. 1A). A NST effect was also found both in the distance moved in and the number of entries into the central zone [F(2,108) = 3.192, P < 0.05 and F(2,108) = 6.596, P < 0.001, respectively]. Globally, the distance travelled in and the number of entries into the central zone was lower in the finasteride-treated group than in the Control group (N-K: P < 0.057 and P < 0.01) or AlloP group (N-K: P < 0.058 and P < 0.05) (see Fig. 1B). Although a NST effect was observed in the distance moved in the outermost portion of the apparatus (close to the wall; [F(2,108) = 4.362, P < 0.05]), post hoc comparisons showed that the finasteride-treated rats did not differ from controls but they showed lower locomotor activity than the AlloP ones (see Fig. 1C). Moreover, the ANOVA of head-dipping data revealed an effect of NST [F(2,108) = 9.423, P < 0.001]. As can be seen in Fig. 1D, lower head-dipping scores were observed in the finasteride-treated animals at both 40 and 60 days. On the other hand, an AGE effect was found both in the distance moved in and the number of entries into the central zone [F(1,108) = 17.612, P < 0.001 and F(1,108) = 17.569, P < 0.001]. Both the distance travelled within and the number of entries into the central zone was higher at 60 days than at 40 days old (N-K, P < 0.001 for both) in all animals. Also, the distance moved in the outer most portions of the apparatus and head-dipping scores decreased across time (from 40

to 60 days old) in all animals [F(1,108) = 16.539, P < 0.001 and F(1,108) = 46.155, P < 0.001]. Because finasteride-treated animals weighed less than the rest of the animals at 40 days (see above), body weight at 40 days was included in the analysis as a covariate (in order to determine the independence of changes). Similar effects were obtained when the total distance moved, the number of entries into the central zone and the number of head-dips were covariated by the body weight at 40 days: NST [F(2,107) = 8.518, P < 0.001]; NST [F(2,107) = 4.757, P < 0.01] and AGE [F(1,108) = 18.203, P < 0.001]; NST [F(2,107) = 6.993, P < 0.01] and AGE [F(1,108) = 46.154, P < 0.001], respectively. Although the distance moved in the inner zone in finasteride-treated rats was also lower than in the rest of the animals, this difference did not reach statistical significance when data was covariated by the body weight at 40 days.

3.3. Prepulse inhibition The analysis of the basal startle amplitude data only revealed a NSTeffect, which just failed to reach statistical significance [F(2,99) = 2.579, P = 0.08]. Post hoc Newman—Keuls tests showed that postnatal AlloP non-significantly decreased basal startle amplitude (N-K: P = 0.06, see Fig. 2A). Neither

530

S. Darbra, M. Pallare `s

Figure 2 Neonatal AlloP deteriorates the prepulse inhibition of the acoustic startle response. (A) PPI expressed as percentage of inhibition startle amplitude. ***P < 0.001 vs. 3 dB prepulse and vs. 5 dB prepulse; * P < 0.05 vs. 3 dB prepulse. (B) PPI expressed as absolute startle amplitude. ***P < 0.001 vs. 3 dB prepulse and vs. 5 dB prepulse; *P < 0.01 vs. 3 dB prepulse. In PULSE ALONE: NST main effect (&P < 0.06 vs. Saline). AVG: percentage of prepulse inhibition in the trials with presence of prepulse (i.e. the arithmetic mean of the prepulse inhibition recorded for each prepulse intensity).

DRUG nor interaction between NST and DRUG effects were found in basal startle amplitude.

3.4. Effects of neonatal neurosteroids administration 3.4.1. PPI expressed as absolute startle amplitude Data analysis showed a significant effect of PULSE [F(2,198) = 17.66, P < 0.001], and a significant interaction between NST and PULSE [F(4,198) = 3.45, P < 0.01], indicating that the effect of neonatal treatment was different according to the specific prepulse intensity. Subsequently, the analysis was split upper neonatal treatment. Further analysis indicated a significant effect of PULSE in all Control and finasteride animals [F(2,50) = 3.93, P < 0.05 and F(2,72) = 5.87, P < 0.01, respectively], reflecting the expected progressive reduction of startle response related

to the increase in prepulse intensity. In contrast, there were no significant effects of PULSE in AlloP-treated animals [F(2,56) = 0.99, NS; see Fig. 2B]. 3.4.2. PPI expressed as a percentage ANOVA showed a significant main effect of PULSE [F(2,198) = 18.32, P < 0.001], and a significant interaction between NST and PULSE [F(4,198) = 2.48, P < 0.05]. Subsequently, the analysis was split upper neonatal treatment. In all Control and finasteride animals, a significant effect of PULSE were observed [F(1,62) = 18.33, P < 0.001; and F(1,72) = 4.219, P < 0.01, respectively], reflecting the expected progressive reduction of startle response after the increase in prepulse intensity. However, in all AlloP subjects the effect of PULSE was not observed [F(1,62) = 2.66, NS]. Moreover, additional post hoc analysis showed that Control rats presented greater PPI percentages than AlloP (N-K: P < 0.05) and

Neonatal neurosteroids levels on adolescent and adult behavior finasteride (N-K: P < 0.05) rats at the highest prepulse intensity (10 dB above background, see Fig. 2A). 3.4.3. Effects of adult midazolam administration Although high doses of MID altered PPI percentages, the interaction between DRUG and PULSE just failed to reach statistical significance [F(6,198) = 1.956, P = 0.07]. In order to elucidate the sense of this possible effect, additional analyses were split upper drug doses. At 0 (VEH), 1 and 1.75 MID doses (mg), the analysis of simple effects revealed a significant main effect of PULSE [F(2,46) = 4.4, P < 0.01; F(2,52) = 10.498, P < 0.001; and F(2,54) = 7.606, P < 0.001, respectively], reflecting the expected progressive increase in prepulse inhibition after the increase in prepulse intensity. Instead, at doses of 2.5 mg of MID, significant PULSE effects were not observed

531 [F(2,46) = 1.172, NS, see Fig. 3A]. So, 2.5 mg of MID disrupted the normal PPI increase across the increasing prepulse values. Also, the analysis split upper prepulse intensities revealed a MID effect only at 10 dB of prepulse intensity [F(3,99) = 2.752, P = 0.05]. Globally and at this prepulse intensity, the animals that received 1 mg/kg of MID presented a significant higher percentage of prepulse inhibition than the animals that received VEH (N-K: P < 0.05) and a non-significant higher percentage than the animals that received 2.5 mg of MID (NK: P = 0.059), as it can be seen in Fig. 3B.

4. Discussion The results obtained in the present experiment indicate that neonatal AlloP manipulation affects behaviour in adolescent

Figure 3 Midazolam administration effects on prepulse inhibition. (A) MID 2.50 interferes the improvement of prepulse inhibition percentage across the increasing prepulse intensity values, independently of neonatal treatment. Within-subjects comparisons: *P < 0.05 vs. 3 dB prepulse; **P < 0.01 vs. 3 dB prepulse; ***P < 0.001 vs. 3 dB prepulse; +P < 0.05 vs. 5 dB prepulse; ++P < 0.01 vs. 5 dB prepulse. (B) MID 1.00 potentiates the percentage of prepulse inhibition of the acoustic startle response at 10 dB prepulse intensity, independently of neonatal treatment. &P < 0.05 vs. MID 0.0 (VEH); #P = 0.059 vs. MID 2.50.

532 and adult ages. Concretely, the inhibition of neonatal AlloP synthesis by means of repeated (5 consecutive days: from pnd 5 to pnd 9) finasteride injections, a 5alpha-reductase inhibitor (Azzolina et al., 1997), decreased the novelty-directed exploratory behaviour during adolescence (at 40 and 60 days of age). This reduction of exploration is pointed out by the decrease in head-dipping behaviour and locomotion recorded in the Boissier test in the neonatal finasteride-treated rats. Exploratory behaviour is the most common response of an animal exposed to novelty, with head-dipping behaviour reflecting the attention paid to novel objects (File and Wardill, 1975). The inhibition of neonatal AlloP synthesis also caused a decrease in the number of entries into, and the distance travelled in, the central zone of the Bossier test. These variables could be considered as anxiety-relevant scores (Voikar et al., 2001; Prut and Belzung, 2003). In this sense, the decrease seen in locomotor activity may also have comprised a component of increased anxiety-related behaviour (Thiel et al., 1999), because finasteride-treated rats entered the central zone less frequently and travelled less distance in it than the Control rats but finasteride-treated animals did not differ in locomotor activity in the outermost portion of the apparatus. Therefore, finasteride-treated subjects demonstrated more anxiety-related behaviour in the Boissier test. The decreased novelty-directed exploratory behaviour shown by finasteride-treated rats may be related to the just-mentioned anxiety-relevant scores in the Boissier test. In this respect, we have previously reported that neonatal finasteride administration (in the same conditions as in the present experiment) induces an anxiogenic-like profile measured in the elevated plus maze (decrease in the open arms time and entries) at 100 days of age (Martin-Garcia et al., 2008). The results of the present experiment also demonstrate that general locomotor activity (measured as the total distance travelled) was stable across age regardless of the neonatal treatment. However, an age effect regardless of the neonatal treatment was observed in exploration, measured by head-dipping in the Boissier test. Namely, headdipping scores were higher at pnd 40 than at pnd 60. Although we cannot completely rule out the possibility that the number of head-dips reflects in part a component of anxiety, it mainly reflects motivation to seek out new experiences (File and Wardill, 1975; Escorihuela et al., 1999; Vaglenova et al., 2004). Adolescent rodents are more responsive to novel stimuli than their adult counterparts. Accordingly, it is reasonable to assume that mid-adolescents (pnd 40) would be more responsive to novel stimuli than post-adolescents (pnd 60). In fact, an enhanced exploratory drive and/or reduced anxiety associated with risky situations during mid-adolescence have been suggested (Laviola et al., 2003; Adriani et al., 2004). On the other hand, at pnd 60 all animals travelled more distance in, and entered into the central zone than at pnd 40. Taking into account that animals were tested twice, i.e. both 40 and 60 days in the same apparatus, the degree of novelty would be not the same because the test arena was not ‘‘completely’’ unfamiliar when animals were tested at pnd 60. Therefore, we cannot rule out the possibility that the subjects became more familiar with the apparatus. Accordingly, environmental stress would be lower in the second test, and as a result, an increase in the number of entries into, and the distance travelled in, the central zone

S. Darbra, M. Pallare `s would be expected in the second exposure to the hole-board apparatus. Recently, it has been reported that repeated exposure of adult subjects to a novel hole-board apparatus greatly affects the behavioural response, and that the neophobic response experienced by subjects during the first exposure to an apparatus apparently declines over further exposures (Brown and Nemes, 2008). Concretely, head-dipping initially decreased across repeated exposures, while time spent in the aversive central area increased, thus in accordance with our data. It is unlikely that the effects of body weight and general locomotion due to finasteride neonatal administration can affect the results obtained on novelty and anxiety-related parameters given that: (1) body weight reduction is only seen at 40 days; (2) similar results were obtained when they were covariated by body weight; (3) finasteride-treated animal did not differ in locomotor activity in the outermost proportion of the apparatus. Nevertheless, it is not possible to rule out the possibility of unspecific body weight reduction eliciting differences in novelty and anxiety-related parameters. In relation to adult behaviour (up to 80 days in the rat), we have found that neonatal administration of AlloP deteriorates the prepulse inhibition of the acoustic startle response at 85 days of age. In particular, the expected progressive reduction of startle response (and the consequent improvement of the PPI percentage) after the gradual increase in prepulse intensity (3, 5 and 10 dB above background) has not been obtained in the neonatal AlloP-treated rats. Also, at 10 dB above background, AlloP animals displayed a lower percentage of PPI than Control animals. Prepulse inhibition represents an index of sensorimotor gating (the ability of a weak sensory event to inhibit — ‘‘gate’’ — the motor response to an intense sensory stimulus), which thought to be governed by central inhibitory mechanisms that strongly influence the structure and cohesiveness of thought (Swerdlow et al., 2001). In all versions, PPI refers to the inhibitory influence of a weak sensory stimulus on the reaction to a startling acoustic or tactile stimulus (Kilts, 2001). Reduced prepulse inhibition of startle provides evidence of deficient sensorimotor gating in several neuropsychiatric disorders, including schizophrenia (Swerdlow et al., 2000; Braff et al., 2001; Kilts, 2001; Swerdlow et al., 2006). It has been consistently reported that perinatal stress, with the consequent rise in corticosterone levels, increases vulnerability to emotional disorders and schizophrenia in adulthood reported in humans and in animal models (Ellenbroek et al., 1998; Dubovicky et al., 1999; Ellenbroek and Cools, 2000, 2002). Furthermore, an important relation between stress and endogenous AlloP levels has been described: AlloP levels increase in response to acute stress (Purdy et al., 1991; Valle ´e et al., 2000) and decrease after long-term stress (Agı´s-Balboa et al., 2007; Pibiri et al., 2008), thus exogenous increases in AlloP may mimic stress-related outcomes (Gizerian et al., 2006). Our results also suggest that the basal startle amplitude could also be affected by neonatal AlloP administration. Although this effect does not reach statistical significance (P = 0.08), it seems that AlloP decreases the response to pulse alone, i.e. the startle response. Animal, and more recently, human investigations have shown that the basic startle response can be potentiated by behavioural manipulations causing fear and anxiety (Grillon et al., 1991, 1993;

Neonatal neurosteroids levels on adolescent and adult behavior Davis et al., 1999; Riba et al., 2001). Thus, neonatal alteration of AlloP levels could have altered the animals’ fear levels. It seems that the decrease in neonatal AlloP (finasteride administration) could increase the fear of novelty situations in adolescence (decrease in exploration in the Boissier test), and that the increase in neonatal AlloP levels could decrease this fear in adulthood, as reflected by the decrease in the acoustic startle reflex. Also, we have observed that an acute administration of AlloP in the fifth postnatal day decreases the anxiolytic effects of lorazepam in the elevated plus maze in the adult age (Darbra and Pallare `s, 2009). On the other hand, although neonatal finasteride does not alter the normal evolution of the PPI percentage across the different prepulse sound intensities, we have observed a decrease in PPI in this group compared to the Control group at the 10 dB above background intensity. Thus, it seems that the reduction of the endogenous neonatal levels of AlloP could also be altering, at least partially, PPI in adulthood. These results confirm previous data indicating that an acute AlloP administration (10 mg/kg, s.c.) in the fifth postnatal day can deteriorate prepulse inhibition in adulthood (Grobin et al., 2006). In relation to MID administration in adulthood, results suggest a possible inverse dose—response relation between MID and prepulse inhibition achievement. The lowest dose of MID (1 mg/kg) improves the PPI percentage, although only at 10 dB above background of prepulse intensity, whereas the effects on PPI of 1.75 mg of MID do not differ from the effects of VEH, and the highest MID dose (2.5 mg/kg) disrupts the expected progressive reduction of the startle response (and the consequent improvement of the PPI percentage) after the gradual increase in prepulse intensity. Nevertheless, more experiments with higher prepulse levels of intensity above background would be necessary to elucidate this possible relation. Otherwise, alterations in the localisation and function of prefrontal (Grobin et al., 2003) and dorsal thalamic (Gizerian et al., 2004) GABAergic neurons in the adult brain after a single AlloP administration (10 mg/kg) on the fifth postnatal day have been documented. MID is a benzodiazepine full agonist that acts on GABAA receptors that contain the a1—3 or a5 GABAA receptor subunits (Iversen, 2004). Nevertheless, our results indicate no interaction between neonatal and adult treatments. Thus, the effects of endogenous neonatal AlloP alterations on adult PPI achievement in adulthood have not been modified by adult MID administration at the three doses tested. From our data we cannot rule out a possible alteration of these GABAA receptor subunits related to neonatal manipulation. However, it seems that the adult PPI disruption induced by neonatal AlloP administration is not related to a possible alteration of these receptors. In summary, the neonatal AlloP administration reduces PPI in adulthood, providing evidence of deficient sensorimotor gating. Moreover, the neonatal finasteride administration increases emotional reactivity in situations of stress or conflict in the adolescent age, as reflected by the reduction in exploration of a novelty situation (decrease in novelty-directed activity and head-dips). It is important to remark that the present results have been obtained only in male rates. As there are sexual differences in the effects of neurosteroids on behaviour (Reddy and Kulkarni, 1999; Zimmerberg et al., 1999; Gulinello and Smith, 2003), further studies are needed

533 in order to show if results can be generalized to females. These results indicate the importance of the maturation of the endogenous neurosteroid mechanisms in the brain related to emotionality and the responses to environmental stressors in adulthood. Alterations of AlloP levels during maturation could partly explain the inter-individual differences shown by adolescents in response to novelty (exploration) and in the sensorimotor gating and prepulse inhibition in adults. Also, abrupt changes in neonatal levels of AlloP could be related to susceptibility to neurodevelopmental disorders such as schizophrenia.

Role of funding source Funding for this study was provided by the Spanish Ministry of Education and Science (SMES, Grant SEJ2006-14792); the SMES had no further role in study design; in the collection, analysis and interpretation of data; in the writing of the report; and in the decision to submit the paper for publication.

Conflict of interest All authors declare that they have no conflicts of interest.

Acknowledgments We thank the Spanish Ministry of Education and Science for the financial aid (Grant SEJ2006-14792).

References Adriani, W., Granstrem, O., Macri, S., Izykenova, G., Dambinova, S., Laviola, G., 2004. Behavioral and neurochemical vulnerability during adolescence in mice: studies with nicotine. Neuropsychopharmacology 29 (5), 869—878. Adriani, W., Laviola, G., 2004. Windows of vulnerability to psychopathology and therapeutic strategy in the adolescent rodent model. Behav. Pharmacol. 15 (5—6), 341—352. Agı´s-Balboa, R.C., Pinna, G., Pibiri, F., Kadriu, B., Costa, E., Guidotti, A., 2007. Down-regulation of neurosteroid biosynthesis in corticolimbic circuits mediates social isolation-induced behavior in mice. Proc. Natl. Acad. Sci. U.S.A. 104, 18736—18741. Akwa, Y., Purdy, R.H., Koob, G.F., Britton, K.T., 1999. The amygdala mediates the anxiolytic-like effect of the neurosteroid allopregnanolone in rat. Behav. Brain Res. 106, 119—125. Azzolina, B., Ellsworth, K., Andersson, S., Geissler, W., Bull, H.G., Harris, G.S., 1997. Inhibition of rat a-reductases by finasteride: evidence for isozyme differences in the mechanism of inhibition. J. Steroid. Biochem. Mol. Biol. 61, 55—64. Baulieu, E.E., 1981. Steroid hormones in the bra. In: several mechanisms?, Fuxe, K., Gustafsson, J.A., Weterberg, L. (Eds.), Steroid Hormone Regulation of the Brain. Pergamon Press, Oxford, pp. 3—14. Boissier, J.R., Simon, P., 1967. Automatisation du test de la planche a ` trous. Physiol. Behav. 2, 447—448. Braff, D.L., Geyer, M.A., Swerdlow, N.R., 2001. Human studies of prepulse inhibition of startle: normal subjects patient groups and pharmacological studies. Psychopharmacology 156, 234— 258. Brown, G.R., Nemes, C., 2008. The exploratory behaviour of rats in the hole-board apparatus: is head-dipping a valid measure of neophilia? Behav. Process. 78 (3), 442—448.

534 Cooper, E.J., Johnston, G.A., Edwards, F.A., 1999. Effects of a naturally occurring neurosteroid on GABAA IPSCs during development in rat hippocampal or cerebellar slices. J. Physiol. 521 (Pt2), 437—449. Darbra, S., Pallare `s, M., 2009. Neonatal allopregnanolone increases novelty-directed locomotion and disrupts behavioural responses to GABAA receptors modulators in adulthood. Int. J. Dev. Neurosci. 27 (6), 617—625. Davis, M., Walker, D.L., Lee, Y., 1999. Neurophysiology and neuropharmacology of startle and its affective modulation. In: Dawson, M.E., Schell, A.M., Bo ¨mehlt, A.H. (Eds.), Startle Modification. Implications for Neuroscience, Cognitive Science and Clinical Science. Cambridge University Press, Cambridge, pp. 95—113. Dubovicky, M., Skultetyova, I., Jezova, D., 1999. Neonatal stress alters habituation of exploratory behavior in adult male but not female rats. Pharmacol. Biochem. Behav. 64, 681—686. Ellenbroek, B.A., Cools, A.R., 2000. Animal models for the negative symptoms of schizophrenia. Behav. Pharmacol. 11 (3—4), 223— 233. Ellenbroek, B.A., Cools, A.R., 2002. Early maternal deprivation and prepulse inhibition: the role of the postdeprivation environment. Pharmacol. Biochem. Behav. 73 (1), 177—184. Ellenbroek, B.A., van den Kroonenberg, P.T., Cools, A.R., 1998. Schizophr. Res. 30 (3) 251—260. Engel, S.R., Grant, K.A., 2001. Neurosteroids and behavior. Int. Rev. Neurobiol. 46, 321—348. Escorihuela, R.M., Ferna ´ndez-Teruel, A., Gil, L., Aguilar, R., Toben ˜a, A., Driscoll, P., 1999. Inbred Roman high- and low-avoidance rats: differences in anxiety, novelty-seeking, and shuttlebox behaviors. Physiol. Behav. 67, 19—26 1999. File, S.E., Wardill, A.G., 1975. Validity of head-dipping as a measure of exploration in a modified hole-board. Psychopharmacologia 44 (1), 53—59. Finn, D.A., Ford, M.M., Wiren, K.M., Roselli, C.E., Crabbe, J.C., 2004. The role of pregnane neurosteroids in ethanol withdrawal: behavioral genetic approaches. Pharmacol. Ther. 101, 91—112. Follesa, P., Biggio, F., Caria, S., Gorini, G., Biggio, G., 2004. Modulation of GABA (a) receptor gene expression by allopregnanolone and ethanol. Eur. J. Pharmacol. 500, 413—425. Frye, C.A., Rhodes, M.E., Petraglia, S.M., Walf, A.A., Sumida, K., Edinger, K.L., 2006. 3a-Hydroxy-5a-pregnan-20-one in the midbrain ventral tegmental area mediates social, sexual, and affective behaviors. Neuroscience 138, 1007—1014. Gizerian, S.S., Morrow, A.L., Lieberman, J.A., Grobin, A.C., 2004. Neonatal neurosteroid administration alters parvalbumin expression and neuron number in medial dorsal thalamus of adult rats. Brain Res. 1012, 66—74. Gizerian, S.S., Moy, S.S., Lieberman, J.A., Grobin, A.C., 2006. Neonatal neurosteroid administration results in developmentspecific alterations in prepulse inhibition and locomotor activity: neurosteroids alter prepulse inhibition and locomotor activity. Psychopharmacology 86, 334—342. Gogos, A., van den Buuse, M., 2007. The importance of baseline in identifying 8-OH-DPAT-induced effects on prepulse inhibition in rats. Br. J. Pharmacol. 150 (6), 750—757. Grillon, C., Ameli, R., Foot, M., Davis, M., 1993. Fear-potentiated startle: relationship to the level of state/trait anxiety in healthy subjects. Biol. Psychiatry 33 (8—9), 566—574. Grillon, C., Ameli, R., Woods, S.W., Merikangas, K., Davis, M., 1991. Fear-potentiated startle in humans: effects of anticipatory anxiety on the acoustic blink reflex. Psychophysiology 28 (5), 588— 595. Grobin, A.C., Gizerian, S., Lieberman, J.A., Morrow, A.L., 2006. Perinatal allopregnanolone influences prefrontal cortex structure, connectivity and behavior in adult rats. Neuroscience 138, 809—819. Grobin, A.C., Heenan, E.J., Lieberman, J.A., Morrow, A.L., 2003. Perinatal neurosteroid levels influence gabaergic interneuron

S. Darbra, M. Pallare `s localization in adult rat prefrontal cortex. J. Neurosci. 23, 1832—1839. Gruen, R.J., Deutch, A.Y., Roth, R.H., 1990. Perinatal diazepam exposure: alterations in exploratory behavior and mesolimbic dopamine turnover. Pharmacol. Biochem. Behav. 36, 169— 175. Gulinello, M., Smith, S.S., 2003. Anxiogenic effects of neurosteroid exposure: sex differences and altered GABAA receptor pharmacology in adult rats. J. Pharmacol. Exp. Ther. 305 (2), 541— 548. Iversen, L., 2004. GABA pharmacology–—what prospects for the future? Biochem. Pharmacol. 68, 1537—1540. Kellogg, C.K., 1999. Sex differences in long-term consequences of prenatal diazepam exposure: possible underlying mechanisms. Pharmacol. Biochem. Behav. 64, 673—680. Kilts, C.D., 2001. The changing roles and targets for animals models of schizophrenia. Biol. Psychiatry 50, 845—855. Lambert, J.J., Belelli, D., Hill-Venning, C., Peters, J.A., 1995. Neurosteroids and GABAA receptor function. Trends Pharmacol. Sci. 16, 295—303. Laviola, G., Macri, S., Morley-Fletcher, S., Adriani, W., 2003. Risktaking behavior in adolescent mice: psychobiological determinants and early epigenetic influence. Neurosci. Biobehav. Rev. 27, 19—31. Mameli, M., Carta, M., Partridge, L.D., Valenzuela, C.F., 2005. Neurosteroid-induced plasticity of immature synapses via retrograde modulation of presynaptic NMDA receptors. J. Neurosci. 25, 2285—2294. Martin-Garcia, E., Pallares, M., 2005a. Intrahippocampal nicotine and neurosteroids effects on the anxiety-like behaviour in voluntary and chronic alcohol-drinking rats. Behav. Brain Res. 164, 117—127. Martin-Garcia, E., Pallares, M., 2005b. The intrahippocampal administration of the neurosteroid allopregnanolone blocks the audiogenic seizures induced by nicotine. Brain Res. 1062, 144—150. Martin-Garcia, E., Darbra, S., Pallares, M., 2008. Neonatal finasteride induces anxiogenic-like profile and deteriorates passive avoidance in adulthood after intrahippocampal neurosteroid administration. Neuroscience 154, 1497—1505. Mellon, S.H., 2007. Neurosteroid regulation of central nervous system development. Pharmacol. Ther. 116 (1), 107—124. Mellon, S.H., Griffin, L.D., 2002. Neurosteroids: biochemistry and clinical significance. Trends Endocrinol. Metab. 13, 35—43. Mtchedlishvili, Z., Sun, C.S., Harrison, M.B., Kapur, J., 2003. Increased neurosteroid sensitivity of hippocampal GABAa receptors during postnatal development. Neuroscience 118, 655—666. Mukai, Y., Higashi, T., Nagura, Y., Shimada, K., 2008. Studies on neurosteroids. XXV. Influence of a 5alpha-reductase inhibitor, finasteride, on rat brain neurosteroid levels and metabolism. Biol. Pharm. Bull. 31 (9), 1646—1650. Muneoka, K.T., Borella, A., Whitaker-Azmitia, P.M., 2003. Transient expression of s-100beta immunostaining in developing thalamus and somatosensory cortex of rat. Brain Res. Dev. Brain Res. 142, 101—104. Muneoka, K., Takigawa, T.M., 2002. A neuroactive steroid, pregnenolone, alters the striatal dopaminergic tone before and after puberty. Neuroendocrinology 75, 288—295. Pibiri, F., Nelson, M., Guidotti, A., Pinna, G., 2008. Decreased corticolimbic allopregnanolone expression during social isolation enhances contextual fear: a model relevant for posttraumatic stress disorder. Proc. Natl. Acad. Sci. U.S.A. 105, 5567—5572. Prut, L., Belzung, C., 2003. The open field as a paradigm to measure the effects of drugs on anxiety-like behaviors: a review. Eur. J. Pharmacol. 463, 3—33. Purdy, R.H., Morrow, A.L., Moore Jr., P.H., Paul, S.M., 1991. Stressinduced elevations of gamma-aminobutyric acid type A receptoractive steroids in the rat brain. Proc. Natl. Acad. Sci. U.S.A. 88, 4553—4557.

Neonatal neurosteroids levels on adolescent and adult behavior Reddy, D.S., 2006. Physiological role of adrenal deoxycorticosteronederived neuroactive steroids in stress-sensitive conditions. Neuroscience 138, 911—920. Reddy, D.S., Kulkarni, S.K., 1999. Sex and estrous cycle-dependent changes in neurosteroid and benzodiazepine effects on food consumption and plus-maze learning behaviors in rats. Pharmacol. Biochem. Behav. 62 (1), 53—60. Riba, J., Rodrı´guez-Fornells, A., Urbano, G., Morte, A., Antonijoan, R., Barbanoj, M.J., 2001. Differential effects of alprazolam on the baseline and fear-potentiated startle reflex in humans: a dose—response study. Psychopharmacology (Berl.) 157 (4), 358— 367. Shirayama, Y., Muneoka, K.T., Iwata, M., Ishida, H., Hazama, G., Kawahara, R., 2005. Pregnenolone and dehydroepiandrosterone administration in neonatal rats alters the immunoreactivity of hippocampal synapsin I, neuropeptide y and glial fibrillary acidic protein at post-puberty. Neuroscience 133, 147—157. Spear, L.P., 2000. The adolescent brain and age-related behavioral manifestations. Neurosci. Biobehav. Rev. 24, 417—463. Swerdlow, N.R., Martinez, Z.A., Hanlon, F.M., Platten, A., Farid, M., Auerbach, P., Braff, D.L., Geyer, M.A., 2000. Toward understanding the biology of a complex phenotype: rat strain and substrain differences in the sensorimotor gating-disruptive effects of dopamine agonists. J. Neurosci. 20, 4325—4336. Swerdlow, N.R., Geyer, M.A., Braff, D.L., 2001. Neural circuit regulation of prepulse inhibition of startle in the rat: current knowledge and future challenges. Psychopharmacology (Berl.) 156 (2— 3), 194—215. Swerdlow, N.R., Shoemaker, J.M., Crain, S., Goins, J., Onozuka, K., Auerbach, P.P., 2004. Sensitivity to drug effects on prepulse

535 inhibition in inbred and outbred rat strains. Pharmacol. Biochem. Behav. 77 (2), 291—302. Swerdlow, N.R., Light, G.A., Cadenhead, K.S., Sprock, J., Hsieh, M.H., Braff, D.L., 2006. Startle gating deficits in a large cohort of patients with schizophrenia: relationship to medications, symptoms, neurocognition, and level of function. Arch. Gen. Psychiatry 63 (12), 1325—1335. Thiel, C.M., Muller, C.P., Huston, J.P., Schwarting, R.K., 1999. High versus low reactivity to a novel environment: behavioural, pharmacological and neurochemical assessment. Neuroscience 93, 243—251. Vaglenova, J., Birru, S., Pandiella, N.M., Breese, C.R., 2004. An assessment of the long-term developmental and behavioral teratogenicity of prenatal nicotine exposure. Behav. Brain Res. 150, 159—170. Valle ´e, M., Rivera, J.D., Koob, G.F., Purdy, R.H., Fitzgerald, R.L., 2000. Quantification of neurosteroids in rat plasma and brain following swim stress and allopregnanolone administration using negative chemical ionization gas chromatography/mass spectrometry. Anal. Biochem. 287 (1), 153—166. van den Buuse, M., 2003. Deficient prepulse inhibition of acoustic startle in Hooded—Wistar rats compared with Sprague—Dawley rats. Clin. Exp. Pharmacol. Physiol. 30 (4), 254—261. Voikar, V., Koks, S., Vasar, E., Rauvala, H., 2001. Strain and gender differences in the behavior of mouse lines commonly used in transgenic studies. Physiol. Behav. 72, 271—281. Zimmerberg, B., Rackow, S.H., George-Friedman, K.P., 1999. Sexdependent behavioral effects of the neurosteroid allopregnanolone (3alpha, 5alpha-THP) in neonatal and adult rats after postnatal stress. Pharmacol. Biochem. Behav. 64 (4), 717—724.