Prenatal stress and neonatal rat brain development

Neuroscience 137 (2006) 145–155 PRENATAL STRESS AND NEONATAL RAT BRAIN DEVELOPMENT D. L. A. VAN DEN HOVE,a,b* H. W. M. STEINBUSCH,b A. SCHEEPENS,a,b W. D. J. VAN DE BERG,a,b L. A. M. KOOIMAN,b B. J. G. BOOSTEN,b J. PRICKAERTSb AND C. E. BLANCOa 1999). In animal studies PS has been associated with reduced birth weight and increased preweaning mortality (Cabrera et al., 1999), disturbances in the hypothalamo– pituitary–adrenal (HPA) axis (Koehl et al., 1999; Szuran et al., 2000; Welberg and Seckl, 2001), increased anxiety (Archer and Blackman, 1971; Vallee et al., 1997), persistent paradoxical sleep alterations (Dugovic et al., 1999), learning and memory deficits (Hayashi et al., 1998; Vallee et al., 1999; Szuran et al., 2000; Gue et al., 2004), and increased depressive-like behavior in later life (Secoli and Teixeira, 1998; Alonso et al., 1999; Morley-Fletcher et al., 2003). The mechanisms accounting for the impact of PS on postnatal life are not fully understood. Investigations on the putative mechanisms involved have focused mainly on the HPA axis, the regulation of which is thought to be impaired in prenatally stressed subjects. As a result, prenatally stressed offspring may be unable to react appropriately to stressful life events. In other words, PS may lead to a brain being permanently sensitized to subsequent stressful situations. This disturbed reactivity of the HPA axis following PS is probably due to an enhanced release of maternal and placental stress hormones such as corticotropin-releasing factor (CRF) and cortisol/corticosterone, which subsequently enter the fetal circulation (Weinstock, 2001; Huizink et al., 2004) and influence fetal development. PS may also have an impact on cellular and synaptic plasticity. For example, PS has been shown to inhibit cell proliferation in the dentate gyrus (DG) of adult rats (Lemaire et al., 2000; Koo et al., 2003) and juvenile nonhuman primates (Coe et al., 2003). In the latter investigation, prenatally stressed rhesus monkeys also showed a reduced hippocampal volume. Other investigators have shown a reduced number of granule cells in the hippocampus of adult female, but not male, rats due to prenatal restraint stress (Schmitz et al., 2002). PS resulted in a decrease in synaptic density in both the hippocampus and cortex of adult rats, whereas brainderived neurotrophic factor (BDNF) protein content was diminished in the cortex, but not in the hippocampus (Koo et al., 2003). However, little is known about the effect of PS on cellular and synaptic plasticity during early postnatal development. We therefore studied the effects of PS on both fetal growth and stress-induced corticosterone secretion, brain cell proliferation, caspase-3-like activity, and BDNF content in neonatal Fischer 344 rats. a Department of Pediatrics, Research Institute Growth and Development, Faculty of Medicine, Maastricht University, P. Debyelaan 25, P. O. Box 5800, 6202 AZ, Maastricht, The Netherlands b Department of Psychiatry and Neuropsychology, Division of Neuroscience, European Graduate School of Neuroscience, Faculty of Medicine, Maastricht University, P.O. Box 616, 6200 MD Maastricht, The Netherlands Abstract—Chronic or repeated stress during human fetal brain development has been associated with various learning, behavioral, and/or mood disorders, including depression in later life. The mechanisms accounting for these effects of prenatal stress are not fully understood. The aim of this study was to investigate the effects of prenatal stress on early postnatal brain development, a disturbance of which may contribute to this increased vulnerability to psychopathology. We studied the effects of prenatal stress on fetal growth, stress-induced corticosterone secretion, brain cell proliferation, caspase-3-like activity and brain-derived neurotrophic factor protein content in newborn Fischer 344 rats. In addition to a slight reduction in birth weight, prenatal stress was associated with elevated corticosterone levels (33.8%) after 1 h of maternal deprivation on postnatal day 1, whereas by postnatal day 8 this pattern was reversed (ⴚ46.5%). Further, prenatal stress resulted in an approximately 50% decrease in brain cell proliferation just after birth in both genders with a concomitant increase in caspase3-like activity within the hippocampus at postnatal day 1 (36.1%) and at postnatal day 5 (females only; 20.1%). Finally, brainderived neurotrophic factor protein content was reduced in both the olfactory bulbs (ⴚ24.6%) and hippocampus (ⴚ28.2%) of prenatally stressed male offspring at postnatal days 1 and 5, respectively. These detrimental central changes observed may partly explain the increased susceptibility of prenatally stressed subjects to mood disorders including depression in later life. © 2005 Published by Elsevier Ltd on behalf of IBRO. Key words: pregnancy, hypothalamo–pituitary–adrenal axis, cell proliferation, caspase-3, brain derived neurotrophic factor, depression. Accumulating evidence suggests that exposure of a pregnant woman to physical and/or psychological stress (prenatal stress; PS) might affect her offspring by promoting the development of various learning, behavioral and/or mood disorders in later life (Stott, 1973; Huttunen and Niskanen, 1978; Meijer, 1985; Ward, 1991; Watson et al., *Corresponding author. Tel: ⫹31-43-388-4120 or ⫹31-62-557-0872; fax: ⫹31-43-367-1096. E-mail address: d.vandenhove@np.unimaas.nl (D. L. A. Van den Hove). Abbreviations: ANOVA, analysis of variance; BDNF, brain-derived neurotrophic factor; E, embryonic day, e.g. embryonic day 0, E0; HPA, axis hypothalamo–pituitary–adrenal axis; LSD, least significant difference; OB, olfactory bulb; P, postnatal day, e.g. postnatal day 2, P2; PS, prenatal stress; SVZ, subventricular zone; 3H-Thy, [3H]thymidine. EXPERIMENTAL PROCEDURES Animals and procedures These animal studies were all approved by the Animal Ethics Board of the Maastricht University, The Netherlands. All proce- 0306-4522/06$30.00⫹0.00 © 2005 Published by Elsevier Ltd on behalf of IBRO. doi:10.1016/j.neuroscience.2005.08.060 145 146 D. L. A. Van den Hove et al. / Neuroscience 137 (2006) 145–155 dures were carried out in compliance with the EC Directive 86/ 609/EEC and with the Dutch law regulating experiments on animals. Number of animals used and their suffering were minimized as much as possible. Acclimatized time-pregnant normal Fischer 344 rats (Charles River, Maastricht, The Netherlands) were used because the Fischer 344 strain is a pure breeding inbred strain with low heterogeneity and is known to be reasonably stress responsive (Izumi et al., 1997). The animals were housed individually within a temperature-controlled environment (21⫾1 °C) with 12-h light/dark and had access to standard rat chow and water ad libitum. Pregnancy was determined by the presence of vaginal plugs (embryonic day 0, E0). PS was performed daily during the last week of pregnancy (E14 –E21). Pregnant female rats (n⫽15) were individually restrained three times a day (at approximately 9:00 AM, 1:00 PM, and 5:00 PM) for 45 min in transparent plastic cylinders (Instrument Development Engineering & Evaluation, Maastricht, The Netherlands) while being exposed to bright light (Ward and Weisz, 1984). Control pregnant females (n⫽16) were left undisturbed in their home cages. Day of birth was defined as postnatal (P) day 0 and as soon as the last pup of a litter had been born, litter size was defined and pup gender and individual body weights were determined. Only litters of eight or more pups were included in this study. No more than two male and/or two female pups per litter were examined at each particular time point to prevent litter effects (Chapman and Stern, 1978). Body weight of the dams was measured at E0, E21, and postpartum day 21 (weaning). For each group seven to eight pups were investigated per gender per age. To determine the degree of cell proliferation, we used the [3H]thymidine (3H-Thy) incorporation method (see below). For this purpose, on the assigned day, either P1, P2, P5, P8 or P15, the pups received a single dose of 3H-Thy (5 ␮Ci/gm body weight) by s.c. injection into the nape of the neck with an ultrafine (30 gauge, fine-tipped, Teflon coated) needle to minimize injury and discomfort. Following 3H-Thy infusion the pups were placed in a pediatric incubator set at 34 °C and 75% humidity to prevent heat loss. Exactly 1 h after 3H-Thy infusion the pups were decapitated and a blood sample was taken from the neck stump for corticosterone determination by radioimmunoassay. The brains were then carefully removed and microdissected using the method of Wagner et al. (1999). We isolated the olfactory bulb (OB), the cerebellum, the hippocampus, and the subventricular zone (SVZ) contained within the rostral forebrain. The microdissected regions were weighed and quickly snap frozen in liquid nitrogen after which they were stored at ⫺75 °C until further analysis. All dissections were performed by the same investigator (D. L. A. Van den Hove) to preserve consistency. For the purpose of a control experiment (see below) an extra group of 16 pregnant dams was delivered to our department at E4. At P1 and P8 prenatally stressed and control offspring were either decapitated immediately after taking them away from their mother or injected with 0.9% NaCl instead of 3H-Thy and placed in a pediatric incubator for one hour beforehand, as described above. Blood was taken from the neck stump and plasma was subsequently used for corticosterone determination. Cell proliferation rate The dissected brain regions were homogenized using a Bead Beater (Biospec Products, OK, USA) for 3⫻30 s in 350 ␮l (for OB and hippocampus) or 1 ml (for cerebellum and SVZ) of ice-cold lysis buffer (137 mM NaCl, 20 mM Tris–HCl (pH 8.0), 1% NP-40, 10% glycerol and a complete protease inhibitor tablet; Roche, The Netherlands). The 3H-Thy incorporation method, as described previously (Tao et al., 1997; Wagner et al., 1999; Scheepens et al., 2003a), uses the DNA synthesis rate in specific brain regions as an index of mitotic activity. Briefly, an aliquot of the homogenate was used to extract all cellular DNA using a standard tri-chloracetic acid precipitation protocol in order to determine the amount of 3H-Thy which was taken up by proliferating cells and incorporated into newly synthesized DNA during the 1 h exposure. Essentially, this measure represents the product of the number of S phase cells within the sample times the DNA synthetic rate of these S phase cells. However, the incorporation of 3H-Thy into DNA depends on the amount of precursor taken up by the tissue. As a consequence, differences in e.g. blood flow between experimental groups might result in differences in 3H-Thy incorporation which do not reflect differences in mitotic activity. For this purpose, another aliquot was used to measure total amount of radioactivity in the tissue fraction. Homogenized DNA and tissue samples were solubilized in 1 ml Soluene-350 (Packard Instruments, The Netherlands) at 50 °C for 2–24 h until the samples were completely dissolved. After this, 5 ml of Hionic-Fluor scintillation cocktail (Packard Instruments) was added and the samples were read for 20 min on a Wallac WinSpectral 1414 liquid scintillation counter. To convert the sample counts per minute (c.p.m.) to disintegrations per minute (d.p.m.) the appropriate quench curves were produced by adding tritium standards to homogenized brain tissue before reading. The d.p.m. measures were corrected for the wet weight of tissue (mg) and the percentage of 3H-Thy incorporated into DNA relative to the total amount of radioactivity counted in the appropriate tissue was calculated. Caspase-3-like activity Caspase-3-like activity was measured using the method as described previously (Puka-Sundvall et al., 2000; Scheepens et al., 2003a). In short, aliquots of homogenized brain samples (30 ␮l) were mixed on a 96 well microtiter plate (Dynex, USA) with 70 ␮l of extraction buffer I (50 mM Tris–HCl (pH 7.3), 100 mM NaCl, 5 mM EDTA, 1 mM EGTA, 3 mM NaN3, 1 mM PMSF, 1 ␮g/ml pepstatin, 2.5 ␮g/ml leupeptin, 10 ␮g/ml aprotinin and 0.2% CHAPS). After incubation for 15 min at room temperature, 100 ␮l of the extraction buffer II was added (extraction buffer I without protease inhibitors and CHAPS but including 4 mM DTT and 50 ␮M peptide substrate Ac-DEVDAMC; Biomol, Germany). Cleavage of the Ac-DEVD-AMC was measured at 37 °C using a Spectramax Gemini microplate fluorometer (Molecular Devices, USA) using an excitation wavelength of 380 nm and an emission wavelength of 460 nm. Cleavage was followed starting 15 min after adding extraction buffer II at 2-min intervals for 3 h. Fluorometric measures were expressed as nmol AMC produced per milligram wet weight of brain tissue per minute. To check the specificity of the assay, an additional 5 ␮l of 5 mg/ml DEVD-CHO (Biomol), a selective caspase-3 inhibitor, was added to one sample of each region at each time point. Blood sampling and corticosterone radioimmunoassay Blood samples taken from the neck stump into heparinized blood collection tubes (Microvette® CB300, Sarstedt, Germany) were kept on ice and subsequently centrifuged at 3000⫻g for 5 min at 4 °C after which the plasma was frozen down to ⫺75 °C for subsequent determination. For the determination of the plasma corticosterone concentrations, 50 ␮l of plasma was extracted with 3 ml dichloromethane and vortexed for one minute. The corticosterone was subsequently measured directly on 1 ml dried dichloromethane and extracted for radioimmunoassay using corticosterone-125I. The radioimmunological reaction was performed overnight at 4 °C, after which a second antibody system was used to separate bound and unbound steroid as previously described in detail. In the original experiment, trunk blood samples were taken D. L. A. Van den Hove et al. / Neuroscience 137 (2006) 145–155 from the pups 1 h after separating them from their mother, which is a known stressor. Therefore, these data do not represent basal corticosterone levels (see below). BDNF content BDNF content was measured in the brain homogenates using the BDNF Emax® ImmunoAssay system from Promega (Madison, MI, USA). Corning high affinity EIA/RIA 96-well plates (Corning, NY, USA) were used and the resultant absorbency was read using a microplate reader at 450 nm (Biorad, CA, USA). We used the acid treatment as described previously (Scheepens et al., 2003b). Tissue samples were diluted within range of the standard curve, either five times (P1 and P2 animals) or 20 times (P5, P8, and P15 animals). Further, the manufacturer’s recommendations were followed. All samples were measured in duplicate. Only tissues from male pups could be used. Statistics Differences in dam weights, litter sizes and male BDNF protein contents were tested using Student’s t-test. Pre-weaning mortality was tested using the Fisher exact test. In the control experiment corticosterone data were evaluated using a threeway analysis of variance (ANOVA) (experimental group⫻ gender⫻basal/maternal deprivation). In all other cases, the data were evaluated with a two-way ANOVA (experimental group⫻gender). Effects were analyzed in more detail with least significant difference (LSD) post hoc tests (P⬍0.05). Correlation analysis was performed using Pearson’s correlation coefficient (rp). Note that both the degree of cell proliferation, caspase-3-like activity and BDNF protein content were determined using the same homogenate. All statistics were carried out using SPSS software version 11.5 (SPSS Inc, USA). 147 RESULTS Dam weights Whereas starting dam weights were the same in both groups, PS dam weights were 10.8% lower at E21 compared with controls (P⬍0.001; see Fig. 1), corresponding to 28% less weight gain over gestation. At the time of weaning (postpartum day 21), the stressed dams still had lower body weights relative to control dams (⫺7.8%; P⬍0.01). Gestational length, litter size, birth weight and pre-weaning mortality PS had no effect on gestational length (all dams delivered at E22; data not shown). There was no difference in litter size between the groups (10.0⫾0.6 and 9.9⫾0.6 pups/litter for controls and PS, respectively). We found a slight overall reduction in birth weight of PS pups (PS effect: ⫺3.0%; P⬍0.001; see Fig. 2). Post hoc analysis showed that both male and female PS birth weights were lower than their corresponding controls (see Fig. 2). In addition, male pups were heavier than female pups (gender effect: 7.2%; P⬍0.001). There were no significant differences in preweaning mortality (0.38⫾0.15 and 0.73⫾0.23 dead pups/ litter for controls and PS, respectively). Stress-induced corticosterone secretion The corticosterone response in the normal developing rat is shown in Fig. 3. In general corticosterone levels changed with age, with low and consistently decreasing levels up to an age of 8 days. At P15, corticosterone levels were much higher again. Fig. 1. Dam weights over gestation, i.e. E0 and E21, and at postpartum day 21 (PP21), in both stressed (PS; n⫽15) and control (C; n⫽16) dams. Bars represent means⫾S.E.M. Whereas initial dam weights (E0) were the same, restraint stress resulted in a reduction in body weight at E21 (PS effect; *** P⬍0.001; t-test). At PP21, PS dam weights were still lower compared with the C (PS effect; *** P⬍0.001; t-test). 148 D. L. A. Van den Hove et al. / Neuroscience 137 (2006) 145–155 Fig. 2. Birth weights of both male (M) and female (F) prenatally stressed (PS) and control (C) pups; n⬇70 per group. Bars represent means⫾S.E.M. Both M and F PS birth weights were lower than the corresponding C (* P⬍0.05; LSD tests). Overall, males were heavier than females (gender effect; P⬍0.001). With respect to the effects of PS, corticosterone levels at P1 were increased in PS pups (PS effect: 33.8%; P⬍0.05). Over time this pattern reversed, with lower corticosterone levels in the PS group at P8 (PS effect: ⫺46.5%; P⬍0.001). By P15, there were no differences between PS and control pups. Control experiment: basal HPA axis activity To investigate the nature of the changed corticosterone levels after 1 h of maternal deprivation seen with PS a control experiment was performed in which the original experiment, in terms of maternal deprivation, was repeated. In addition, an equal number of pups were decapitated immediately after taking them away from the mother. These experiments were performed at both P1 and P8. Results are shown in Table 1. At P1 basal corticosterone levels tend to be higher in PS pups as compared with controls (PS effect: 31.6%; P⫽0.065). As with the original experiment this difference was more profound within the maternal deprivation group (PS effect: 46.3%; P⬍0.01). In addition, overall, basal corticosterone levels were higher as compared with levels after maternal deprivation (34.9%; P⬍0.01). Post hoc analysis showed that these effects were more profound in female offspring (Table 1). At P8, a significant effect of gender was observed, with female offspring showing higher levels of plasma corticosterone as compared with males (gender effect: 22.6%; P⬍0.001). In addition, overall, levels after maternal deprivation were much higher as compared with basal levels (159.4%; P⬍0.001). Cell proliferation rate The pattern of cell proliferation within the neonatal brain is shown in Fig. 4. Cell proliferation levels at P1 were lower in all brain regions compared with those at P2. From P2 onwards, all structures showed a decrease in the degree of cell proliferation over time. PS induced a strong decrease in cell proliferation at P1 in all regions (all PS effects; OB: ⫺55.8%; P⬍0.001; cerebellum: ⫺40.2%; P⬍0.001; hippocampus ⫺56.5%; P⬍0.001; SVZ: ⫺53.0%; P⬍0.001; Fig. 4). At P8, PS pups showed a slight increase in cell proliferation within the cerebellum (PS effect: 4.1%; P⬍0.01). Importantly, the total amount of Fig. 3. The effect of PS on maternal deprivation-induced plasma corticosterone secretion during neonatal development (P1 to P15); n⫽7– 8/group/ age. Bars represent means⫾S.E.M. PS was associated with an overall increase in corticosterone levels at P1 (PS effect [males⫹females]; # P⬍0.05). At P8, both male and female PS rats had decreased corticosterone levels compared with control pups (* P⬍0.05; LSD tests). D. L. A. Van den Hove et al. / Neuroscience 137 (2006) 145–155 149 Table 1. Control experiment P1 P8 M F M F 92.62⫾10.18 74.89⫾11.12 106.49⫾11.56 63.95⫾10.27# 24.72⫾2.36 52.39⫾2.83# 21.74⫾3.05 69.96⫾5.71#,$ 116.29⫾19.13 98.51⫾8.51 145.73⫾19.05* 104.56⫾9.35# 19.87⫾1.04 47.94⫾3.85# 23.42⫾1.68 62.60⫾3.32#,$ C B MD PS B MD The effects of PS on basal (B) and maternal deprivation-induced (MD) plasma corticosterone levels at P1 and P8; M, males; F, females; n⫽7– 8/group/age. Bars represent means⫾S.E.M. At P1 basal corticosterone levels tend to be higher in PS pups as compared to controls (PS effect [males⫹females]; P⫽0.065). This difference was more profound within the maternal deprivation group (PS effect [males⫹females]; P⬍0.01). In addition, overall, basal corticosterone levels were higher as compared to levels after maternal deprivation (P⬍0.01) at P1. At P8, in contrast, levels after maternal deprivation were higher as compared to basal levels (P⬍0.001). LSD tests: * P⬍0.05 prenatal stress compared with control; # P⬍0.05 maternal deprivation compared with basal; $ P⬍0.05 females compared with males. For more information, see text. 3 H-Thy taken up by the tissue, i.e. tracer availability, did not differ between PS and control offspring in any of these cases, except for the SVZ at P1, where tracer availability was actually higher in PS pups as compared with controls (PS effect: 33.5%; P⬍0.01; data not shown). No significant differences were found in any of the other cases. In addition, no gender differences were found. Of note, no differences in brain region weights or total brain weights were observed between PS and control offspring (data not shown). Caspase-3-like activity Overall, caspase-3-like activity was relatively low in the OB and hippocampus compared with the SVZ and cerebellum (Fig. 5). In the OB and hippocampus caspase-3-like activity declined after P1, whereas it increased in the cerebellum and SVZ after P1, peaking at P8 and P5, respectively. Regarding the effects of PS, we found an increase in caspase-3-like activity within the hippocampus at P1 (PS effect: 36.1%; P⬍0.05) and at P5. The PS effect at P5 was however confirmed with post hoc analysis only for the females (20.1%; P⬍0.05; LSD test). No significant differences were found in any of the other cases. BDNF content The pattern of BDNF content within the normal and PS male brain is illustrated in Fig. 6. All structures had relatively low levels at P2 compared with the other time points. In contrast to the other regions, cerebellar BDNF levels decreased again after P8, as has been described previously (Das et al., 2001). Male PS BDNF levels all tended to be lower at P1 compared with controls. However, this difference was only significant in the OB (⫺24.6%; P⬍0.05; of note, cerebellum: ⫺50.1%; P⫽0.062). Hippocampal BDNF content at P5 was 28% lower in PS rats compared with controls (P⬍0.05). No significant differences were found in any of the other cases. Correlations Only when an effect of PS was observed, correlation analysis was performed (see Table 2). At P1, there was a Table 2. The effect of PS corticosterone, cell proliferation, caspase-3-like activity and BDNF content and the PS-induced changes and correlations between the different variables Affected variable Age Region Effect PS Corticosterone P1 P8 P1 P1 P1 P1 P8 P1 P5 P1 P5 n.a. n.a. OB CRB HIP SVZ CRB HIP HIP OB HIP 1 2 2 2 2 2 1 1 1 2 2 Cell proliferation Caspase-3-like activity BDNF Corticosterone ⫺0.61*** ⫺0.57*** ⫺0.60*** ⫺0.57*** ⫺0.19 0.17 ⫺0.26 ⫺0.10 0.46 Cell proliferation Caspase-3 BDNF See below See below ⫺0.49 (SVZ)** No correlations ⫺0.33 ⫺0.16 ⫺0.44* 0.12 ⫺0.11 No correlations No correlations 0.63** 0.17 0.15 ⫺0.11 ⫺0.05 ⫺0.06 See above 0.05 0.07 ⫺0.09 ⫺0.13 ⫺0.53P⫽0.052 Depicted are Pearson’s correlation coefficients (rp) representing the associations between the PS-induced changes and the other variables measured in the same plasma and/or tissue samples. Note that BDNF was measured in males only. In addition, caspase-3-like activity in the hippocampus at P5 was only affected in females, which were therefore used for correlation analysis. * P⬍0.05; ** P⬍0.01; *** P⬍0.001; CRB, cerebellum; HIP, hippocampus; n.a., not applicable. 150 D. L. A. Van den Hove et al. / Neuroscience 137 (2006) 145–155 Fig. 4. The effect of PS on developmental cell proliferation rate within the OB, cerebellum, hippocampus, and SVZ, from P1 to P15; n⫽7– 8/group/age. Bars represent means⫾S.E.M. PS induced a strong decrease in cell proliferation at P1 in all regions studied of both male as well as female pups (* P⬍0.05; LSD tests). In contrast, there was a small increase in cerebellar cell proliferation at P8 in both male and female PS pups (* P⬍0.05; LSD tests). negative correlation between cell proliferation values and plasma corticosterone levels (rp⬇⫺0.6 in each region; P⬍0.001 in all cases). In addition, male P1 OB cell proliferation values were positively correlated to BDNF levels in this region (rp⫽0.63; P⬍0.01). Furthermore, hippocampal caspase-3like activity levels were negatively correlated to cell proliferation values within this region at P1 (rp⫽⫺0.44; P⬍0.05). Further, P1 plasma corticosterone levels were negatively correlated to caspase-3-like activity within the SVZ (rp⫽⫺0.49; P⬍0.01). Finally, there was a trend toward a negative correlation between BDNF protein content and the levels of caspase-3-like activity within the hippocampus at P5 (rp⫽⫺0.53; P⫽0.052). DISCUSSION Besides impaired fetal growth, PS was associated with an aberrant HPA axis (re-)activity, a strong decrease in brain cell proliferation, increased caspase-3-like activity, and reduced brain BDNF content just after birth. These developmental abnormalities may have significant long-term detrimental consequences for brain functioning. D. L. A. Van den Hove et al. / Neuroscience 137 (2006) 145–155 151 Fig. 5. Developmental caspase-3-like activity within the OB, cerebellum, hippocampus, and SVZ, from P1 to P15; n⫽7– 8/group/age. Bars represent means⫾S.E.M. At P1, PS resulted in an overall increase in caspase-3-like activity within the hippocampus (PS effect [males⫹females]; # P⬍0.05). At P5, caspase-3-like activity levels were elevated within the hippocampus of prenatally stressed females only (* P⬍0.05; LSD test). Somatic growth Pups of dams exposed to stress during pregnancy weighed less at birth compared with control offspring. Moreover, reduced birth weight has been related to various diseases in adulthood, e.g. coronary heart disease, hypercholesterolemia, hypertension, stroke and non-insulin-dependent diabetes mellitus, referred to as the “Barker hypothesis” or the “fetal origins of adult disease hypothesis” (Barker, 1995). In addition, low birth weight has, more recently, been associated with an increased susceptibility to stress (Nilsson et al., 2001) and depression in later life (Thompson et al., 2001; Gale and Martyn, 2004). The observed reduction in birth weight in prenatally stressed pups may be explained by the reduced weight gain over gestation (⫺28%) of the stressed dams. Unfortunately, we were not able to check whether this was due to a reduction in food and water intake and/or an impaired conversion of dietary calories into maternal weight gain as seen in reaction to stress (Ward and Wainwright, 1988; Hobel and Culhane, 2003). Both phenomena might indirectly impair fetal growth, eventually resulting in a lower birth weight. It has been suggested that a reduced food and water intake by the mother by itself may largely explain the effects of pregnancy stress 152 D. L. A. Van den Hove et al. / Neuroscience 137 (2006) 145–155 Fig. 6. The effect of PS on BDNF protein content within the OB, cerebellum, hippocampus, and SVZ of the neonatal male rat brain, from P1 to P15; n⫽7– 8/group/age. Bars represent means⫾S.E.M. PS resulted in a reduction of BDNF levels in the OB at P1 (* P⬍0.05; cerebellum: P⫽0.062). Further, PS resulted in a reduction of BDNF level within the hippocampus of 5-day-old male pups (* P⬍0.05). on offspring development (Ward and Wainwright, 1988). In addition, the transplacental transfer of maternal stress hormones and a reduction in uteroplacental blood flow, both of which are commonly observed in pregnancy stress, may affect fetal growth (Hobel and Culhane, 2003; Huizink et al., 2004). whereas at P15, corticosterone levels were much higher again. Interestingly at P1, we observed higher corticosterone levels after a 1 h maternal deprivation in both male and female PS pups as compared with controls. In contrast, at P8, corticosterone levels were lower in the PS group as compared with the control group. Stress-induced corticosterone secretion Control experiment: basal HPA axis activity The first two weeks of postnatal life in rats are characterized by a hyporesponsiveness to stress (stress-hyporesponsive period) (Henning, 1978; De Kloet et al., 1988). In line with this, we found low and consistently decreasing corticosterone levels up to an age of 8 days in both groups, From the original experiment it was unclear whether the higher corticosterone levels after a 1 h maternal deprivation in PS pups at P1 were the result of a higher basal HPA axis activity or an affected negative feedback function of the HPA axis. The control experiment performed suggests D. L. A. Van den Hove et al. / Neuroscience 137 (2006) 145–155 that basal HPA axis activity is affected by PS immediately after birth. At P8 plasma corticosterone levels were much higher in both PS and control pups after maternal deprivation, a known stressor, as compared with basal levels. Surprisingly, at P1, maternal deprivation resulted in decreased corticosterone levels. The implications of this new and rather unexpected finding are not clear at the moment. One could speculate that since this decrease was observed in both PS and control offspring, a common physiological mechanism is involved, related to parturition, which probably reflects the start of the stress-hyporesponsive period. This awaits further research though. The higher corticosterone levels in the PS group as compared with controls after maternal deprivation at P8 as seen in the original investigation were not found in the control experiment. This might be explained by the fact that for the control experiment, rats were delivered pregnant to our apartment at E4 for practical reasons, which may have had a stressful effect on both groups of dams and their offspring, possible causing a slight shift/change in the pattern. Taken together, these data support the idea that PS perturbs the development and function of the HPA-axis; a disruption that may have important consequences for adult stress-related behavior. Cell proliferation and caspase-3-like activity within the neonatal brain As discussed in the introduction, several studies have shown that PS results in an inhibition of proliferation of future neurons in the adult hippocampus. We now show that already during early neonatal development, at P1, hippocampal cell proliferation is drastically (⫺58%) inhibited by PS. Interestingly, this effect was not restricted to the hippocampus, since cell proliferation was affected to the same extent in all brain regions investigated. Further experiments using 3H-Thy autoradiography or 5-bromo-2deoxyuridine (BrdU) immunohistochemistry and state-ofthe-art stereology should be performed to investigate whether the effects of PS are primarily on neuronal or glial cells, i.e. does PS have consequences for the different types of differentiation of progenitor cells ultimately? Interestingly, cell proliferation values in all brain regions at P1 were negatively correlated to plasma corticosterone levels. It is not clear though, whether both features are indeed linked in terms of cause and effect. In a recent investigation by Yu and colleagues (2004) it was shown that both corticosterone and dexamethasone, a glucocorticoid receptor agonist, were able to suppress the proliferation of cells derived from the hippocampus of 16.5-dayold rat embryos (Yu et al., 2004). This might suggest that elevated levels of corticosterone resulting from HPA-axis (hyper-)activity may play an important role in controlling (hippocampal) cell proliferation. Kippin and colleagues (2004) recently showed that PS in hamsters is associated with a reduction in the number of neural stem cells derived from the subependyma of the lateral ventricle (Kippin et al., 2004). In support of our findings, they showed that this 153 reduction was already present at birth, endured into late adulthood and was accompanied by a reduced cell proliferation in the adult subependyma of the lateral ventricle. Interestingly, they also showed that postnatal handling was able to increase neural stem cell number, thus reversing the effects of PS. At postnatal day 8, we observed a slight increase in the degree of cell proliferation in the cerebellum of PS pups. We think this may represent a compensatory, or catch-up effect to the decrease in cell proliferation seen at P1; a detrimental effect that may also be present even earlier, i.e. prenatally. Another explanation for the higher levels of cell proliferation in the PS group could be the higher corticosterone levels in the control group at this time point. As discussed before, corticosterone may exert a negative influence on brain cell proliferation. In our investigation, the activity of caspase-3—a pivotal mediator of apoptosis—was significantly increased after PS within the hippocampus, both at P1 (males⫹females) and at P5 (females only). In agreement with our results, Ladefoged et al. (2004), using a combination of various prenatal stressors, found a similar increase in caspase-3 activity in the hippocampus of male (females were not included) Wistar pups of 6 days of age. In addition, they also found an increase in cerebellar caspase-3 activity at this age. These data show that both hippocampal cell proliferation and apoptosis are affected by PS during early postnatal development, which is likely to have long-term consequences for adult hippocampal functioning. BDNF content in the neonatal brain We show that PS resulted in a reduction in BDNF levels at P1 in the OB of male rats. Further, there was a positive correlation between the levels of BDNF and cell proliferation within this region at this time point. In addition, PS resulted in a reduction in BDNF level within the hippocampus of 5-day-old pups. In this particular case, BDNF levels showed a trend toward a negative correlation with caspase-3-like activity. BDNF, as a member of the neurotrophin family of regulatory factors, is necessary for the survival and function of neurons, and plays an important role in the modulation of synaptic transmission and synaptic plasticity. In this respect, reduced BDNF function is thought to play an important role in the pathophysiology of mood disorders (see review by Manji et al., 2003). Interestingly, our findings differ from those of Koo et al. (2003), who found a reduction in cortical, but not in hippocampal BDNF level, after PS. Koo and coworkers did, however, not investigate the neonatal, but the young adult rat brain instead. In addition, as shown by others, fetal hypothalamic BDNF protein content was not altered by maternal restraint stress (Fujioka et al., 2003). Altogether, the observed changes in BDNF content may have important consequences for cellular and synaptic plasticity, both during development and in adulthood. 154 D. L. A. Van den Hove et al. / Neuroscience 137 (2006) 145–155 CONCLUSION In conclusion, in this study we show that PS has several important consequences for early postnatal development. In addition to a reduction in birth weight, PS caused a disturbed HPA axis (re-)activity, a strong decrease in brain cell proliferation, increased caspase-3-like activity, and reduced brain BDNF levels just after birth. The observed developmental abnormalities may have significant long-term detrimental consequences for brain functioning and may explain the increased susceptibility to psychopathology (e.g. depression) observed in prenatally stressed subjects. Acknowledgments—The authors would like to thank Mrs. Marjanne Markerink for her assistance and expert advice with the ELISAs. J. Prickaerts is supported by the EU Framework G Integrated Project NEWMOOD (LSHM-CT-2004-503474). REFERENCES Alonso SJ, Castellano MA, Quintero M, Navarro E (1999) Action of antidepressant drugs on maternal stress-induced hypoactivity in female rats. Methods Find Exp Clin Pharmacol 21:291–295. Archer JE, Blackman DE (1971) Prenatal psychological stress and offspring behavior in rats and mice. Dev Psychobiol 4:193–248. Barker DJ (1995) The fetal origins of adult disease. Proc R Soc Lond B Biol Sci 262:37– 43. Cabrera RJ, Rodriguez-Echandia EL, Jatuff AS, Foscolo M (1999) Effects of prenatal exposure to a mild chronic variable stress on body weight, preweaning mortality and rat behavior. Braz J Med Biol Res 32:1229 –1237. Chapman RH, Stern JM (1978) Maternal stress and pituitary-adrenal manipulations during pregnancy in rats: effects on morphology and sexual behavior of male offspring. J Comp Physiol Psychol 92: 1074–1083. Coe CL, Kramer M, Czeh B, Gould E, Reeves AJ, Kirschbaum C, Fuchs E (2003) Prenatal stress diminishes neurogenesis in the dentate gyrus of juvenile rhesus monkeys. Biol Psychiatry 54:1025–1034. Das KP, Chao SL, White LD, Haines WT, Harry GJ, Tilson HA, Barone JS (2001) Differential patterns of nerve growth factor, brain-derived neurotrophic factor and neurotrophin-3 mRNA and protein levels in developing regions of rat brain. Neuroscience 103:739 –761. De Kloet ER, Rosenfeld P, Van Eekelen JA, Sutanto W, Levine S (1988) Stress, glucocorticoids and development. Prog Brain Res 73:101–120. Dugovic C, Maccari S, Weibel L, Turek FW, Van Reeth O (1999) High corticosterone levels in prenatally stressed rats predict persistent paradoxical sleep alterations. J Neurosci 19:8656 – 8664. Fujioka T, Fujioka A, Endoh H, Sakata Y, Furukawa S, Nakamura S (2003) Materno-fetal coordination of stress-induced fos expression in the hypothalamic paraventricular nucleus during pregnancy. Neuroscience 118:409 – 415. Gale CR, Martyn CN (2004) Birth weight and later risk of depression in a national birth cohort. Br J Psychiatry 184:28 –33. Gue M, Bravard A, Meunier J, Veyrier R, Gaillet S, Recasens M, Maurice T (2004) Sex differences in learning deficits induced by prenatal stress in juvenile rats. Behav Brain Res 150:149 –157. Hayashi A, Nagaoka M, Yamada K, Ichitani Y, Miake Y, Okado N (1998) Maternal stress induces synaptic loss and developmental disabilities of offspring. Int J Dev Neurosci 16:209 –216. Henning SJ (1978) Plasma concentrations of total and free corticosterone during development in the rat. Am J Physiol 235:E451–E456. Hobel C, Culhane J (2003) Role of psychosocial and nutritional stress on poor pregnancy outcome. J Nutr 133:1709S–1717S. Huizink AC, Mulder EJ, Buitelaar JK (2004) Prenatal stress and risk for psychopathology: specific effects or induction of general susceptibility? Psychol Bull 130:115–142. Huttunen MO, Niskanen P (1978) Prenatal loss of father and psychiatric disorders. Arch Gen Psychiatry 35:429 – 431. Izumi J, Washizuka M, Hayashi-Kuwabara Y, Yoshinaga K, Tanaka Y, Ikeda Y, Kiuchi Y, Oguchi K (1997) Evidence for a depressive-like state induced by repeated saline injections in Fischer 344 rats. Pharmacol Biochem Behav 57:883– 888. Kippin TE, Cain SW, Masum Z, Ralph MR (2004) Neural stem cells show bidirectional experience-dependent plasticity in the perinatal mammalian brain. J Neurosci 24:2832–2836. Koehl M, Darnaudery M, Dulluc J, Van Reeth O, Le Moal M, Maccari S (1999) Prenatal stress alters circadian activity of hypothalamopituitary-adrenal axis and hippocampal corticosteroid receptors in adult rats of both gender. J Neurobiol 40:302–315. Koo JW, Park CH, Choi SH, Kim NJ, Kim HS, Choe JC, Suh YH (2003) The postnatal environment can counteract prenatal effects on cognitive ability, cell proliferation, and synaptic protein expression. FASEB J 17:1556 –1558. Ladefoged O, Hougaard KS, Hass U, Sorensen IK, Lund SP, Svendsen GW, Lam HR (2004) Effects of combined prenatal stress and toluene exposure on apoptotic neurodegeneration in cerebellum and hippocampus of rats. Basic Clin Pharmacol Toxicol 94:169–176. Lemaire V, Koehl M, Le Moal M, Abrous DN (2000) Prenatal stress produces learning deficits associated with an inhibition of neurogenesis in the hippocampus. Proc Natl Acad Sci USA 97:11032– 11037. Manji HK, Quiroz JA, Sporn J, Payne JL, Denicoff KA, Gray N, Zarate J, Carlos A, Charney DS (2003) Enhancing neuronal plasticity and cellular resilience to develop novel, improved therapeutics for difficult-to-treat depression. Biol Psychiatry 53:707–742. Meijer A (1985) Child psychiatric sequelae of maternal war stress. Acta Psychiatr Scand 72:505–511. Morley-Fletcher S, Darnaudery M, Koehl M, Casolini P, Van Reeth O, Maccari S (2003) Prenatal stress in rats predicts immobility behavior in the forced swim test. Effects of a chronic treatment with tianeptine. Brain Res 989:246 –251. Nilsson PM, Nyberg P, Ostergren P-O (2001) Increased susceptibility to stress at a psychological assessment of stress tolerance is associated with impaired fetal growth. Int J Epidemiol 30:75– 80. Puka-Sundvall M, Hallin U, Zhu C, Wang X, Karlsson JO, Blomgren K, Hagberg H (2000) NMDA blockade attenuates caspase-3 activation and DNA fragmentation after neonatal hypoxia-ischemia. Neuroreport 11:2833–2836. Scheepens A, van de Waarenburg M, van den Hove D, Blanco CE (2003a) A single course of prenatal betamethasone in the rat alters postnatal brain cell proliferation but not apoptosis. J Physiol 552: 163–175. Scheepens A, Wassink G, Piersma MJ, Van de Berg WD, Blanco CE (2003b) A delayed increase in hippocampal proliferation following global asphyxia in the neonatal rat. Brain Res Dev Brain Res 142:67–76. Schmitz C, Rhodes ME, Bludau M, Kaplan S, Ong P, Ueffing I, Vehoff J, Korr H, Frye CA (2002) Depression: reduced number of granule cells in the hippocampus of female, but not male, rats due to prenatal restraint stress. Mol Psychiatry 7:810 – 813. Secoli SR, Teixeira NA (1998) Chronic prenatal stress affects development and behavioral depression in rats. Stress 2:273–280. Stott DH (1973) Follow-up study from birth of the effects of prenatal stresses. Dev Med Child Neurol 15:770 –787. Szuran TF, Pliska V, Pokorny J, Welzl H (2000) Prenatal stress in rats: effects on plasma corticosterone, hippocampal glucocorticoid receptors, and maze performance. Physiol Behav 71:353–362. Tao Y, Black IB, DiCicco-Bloom E (1997) In vivo neurogenesis is inhibited by neutralizing antibodies to basic fibroblast growth factor. J Neurobiol 33:289 –296. Thompson C, Syddall H, Rodin I, Osmond C, Barker DJ (2001) Birth weight and the risk of depressive disorder in late life. Br J Psychiatry 179:450 – 455. D. L. A. Van den Hove et al. / Neuroscience 137 (2006) 145–155 Vallee M, MacCari S, Dellu F, Simon H, Le Moal M, Mayo W (1999) Long-term effects of prenatal stress and postnatal handling on age-related glucocorticoid secretion and cognitive performance: a longitudinal study in the rat. Eur J Neurosci 11:2906 –2916. Vallee M, Mayo W, Dellu F, Le Moal M, Simon H, Maccari S (1997) Prenatal stress induces high anxiety and postnatal handling induces low anxiety in adult offspring: correlation with stress-induced corticosterone secretion. J Neurosci 17:2626 –2636. Wagner JP, Black IB, DiCicco-Bloom E (1999) Stimulation of neonatal and adult brain neurogenesis by subcutaneous injection of basic fibroblast growth factor. J Neurosci 19:6006 – 6016. Ward AJ (1991) Prenatal stress and childhood psychopathology. Child Psychiatry Hum Dev 22:97–110. Ward GR, Wainwright PE (1988) Reductions in maternal food and water intake account for prenatal stress effects on neurobehavioral development in B6D2F2 mice. Physiol Behav 44:781–786. 155 Ward IL, Weisz J (1984) Differential effects of maternal stress on circulating levels of corticosterone, progesterone, and testosterone in male and female rat fetuses and their mothers. Endocrinology 114:1635–1644. Watson JB, Mednick SA, Huttunen M, Wang X (1999) Prenatal teratogens and the development of adult mental illness. Dev Psychopathol 11:457– 466. Weinstock M (2001) Alterations induced by gestational stress in brain morphology and behaviour of the offspring. Prog Neurobiol 65: 427–451. Welberg LA, Seckl JR (2001) Prenatal stress, glucocorticoids and the programming of the brain. J Neuroendocrinol 13:113– 128. Yu IT, Lee S-H, Lee Y-S, Son H (2004) Differential effects of corticosterone and dexamethasone on hippocampal neurogenesis in vitro. Biochem Biophys Res Commun 317:484 – 490. (Accepted 28 August 2005) (Available online 20 October 2005)