Prenatal exposure to heroin in mice elicits memory deficits that can be attributed to neuronal apoptosis

Neuroscience 160 (2009) 330 –338

PRENATAL EXPOSURE TO HEROIN IN MICE ELICITS MEMORY DEFICITS THAT CAN BE ATTRIBUTED TO NEURONAL APOPTOSIS Y. WANG AND T.-Z. HAN*

long been known to increase the risk of a variety of neurobehavioral defects in the offspring (Hutchings, 1982; Wilson, 1989), yet the neuronal mechanisms underlying this effect remain unclear. Heroin, also known as diacetylmorphine, is a highly addictive drug that is derived from morphine, but is four to eight times more potent. It can rapidly cross the blood– brain barrier due to the presence of acetyl groups, which render it more lipid-soluble than morphine itself (Tschacher et al., 2003). Heroin can act as an exogenous agonist of mu opioid receptors, and can negatively impact hippocampal function by decreasing adult hippocampal neurogenesis (Harburg et al., 2007). There is evidence that heroin targets proteins involved in signaling cascades that are shared by multiple types of receptor and hormonal input, thus eliciting heterogeneous changes in the responses to a wide variety of neurotransmitters in the hippocampus (Shahak et al., 2003; Steingart et al., 2000; Yanai et al., 1992). The hippocampus plays crucial roles in cognitive functions such as learning and memory, and is known to be more vulnerable than other brain regions to stress and other detrimental stimuli. It has also been reported that the hippocampus is particularly susceptible to teratogens during neurogenesis (Emeterio et al., 2006). Apoptosis (programmed cell death) is a crucial physiological determinant of embryonic and neonatal development. Research indicates that there are two main apoptotic pathways: the death receptor pathway and the mitochondrial pathway (Danial and Korsmeyer, 2004; Ghobrial et al., 2005). However, there is now evidence that these two pathways are linked and that the molecules in one pathway can influence those in the other (Igney and Krammer, 2002). The main function of developmental apoptosis in neurons appears to be the adjustment of neuron numbers and the establishment of a precise match between neural circuits by matching the number of innervating neurons to the size of their target cell population (Lee et al., 1994; Vogel, 1994). A previous study on the neurodegeneration of PC12 cells, a dopaminergic cell line, showed that drugs of abuse can induce apoptotic features by elevating the activity of caspase-3 (Oliveira et al., 2003). Opioid drugs (heroin and morphine) are more toxic than stimulant drugs (d-amphetamine and cocaine) in this regard (Oliveira et al., 2003). Although heroin-induced apoptosis in neurons has been reported in a several studies, there is no information in the literature regarding the involvement of apoptosis in the mechanisms underlying the memory deficits observed in prenatally opioid-exposed offspring. We propose that prenatal exposure to heroin induces alterations in the apoptotic pathways that participate in the

Department of Physiology and Pathophysiology, Key Laboratory of Environment and Genes Related to Diseases, Ministry of Education, School of Medicine, Xi’an Jiaotong University, Zhuque Dalie, 205, Xi’an, Shannxi 710061, PR China

Abstract—Maternal heroin abuse has been shown to result in teratogenic neurobehavioral defects in the offspring, but the underlying mechanisms remain largely unknown. This study was designed to explore the role of neuronal apoptosis in the heroin-induced neurobehavioral defects of learning and memory. Pregnant BALB/c mice were treated with either heroin or saline. The animals in the heroin group received heroin subcutaneously at a dosage of 10 mg/kg/day on embryonic days (E) 9 –18, while those in the saline group were treated as drug-naive. Offspring were grouped as prenatal heroin exposure (HER), prenatal saline exposure (SAL), and control (CON) groups, according to the maternal treatment regimen. Some of the mice were killed and their hippocampus harvested on postnatal day (P) 14, and the tissue subjected to reverse transcription polymerase chain reaction, Western blotting, and immunohistochemistry to reveal the mRNA and protein expressions of caspase-3, Bcl-2, and Bax. The Morris water maze was applied to assess the learning and memory capability of the mice at P30; poor maze performances were observed for the animals in the HER group. The results also showed that the mRNA and protein expressions of caspase-3 and Bax were significantly increased, while that of Bcl-2 was markedly decreased in the HER group compared with both the SAL and CON groups. The immunohistochemistry revealed significant caspase-3 immunoreactivity in the dentate gyrus and cornu ammonis (CA) 1 subareas of the hippocampal formation, whereas, no significant changes were seen in subarea CA3. These findings suggest that prenatal heroin exposure during the E9 –18 period enhances neuronal apoptosis by altering the expressions of caspase-3, Bcl-2, and Bax in the mouse hippocampus, and leads to impairment in hippocampus-dependent learning and memory. © 2009 IBRO. Published by Elsevier Ltd. All rights reserved. Key words: heroin, caspase-3, Bcl-2, Bax, prenatal exposure.

The 2005 National Survey of Drug Use and Health reported that heroin is one of the most common opioid drugs used by pregnant women. Heroin abuse in pregnancy has *Corresponding author. Tel: ⫹86-29-13110401960; fax: ⫹86-2982655274. E-mail address: [email protected] (T.-Z. Han). Abbreviations: ANOVA, analysis of variance; CA, cornu ammonis; CON, control group; DG, dentate gyrus; E, embryonic day; ECL, enhanced chemiluminescence; HER, prenatal heroin exposure group; HRP, horseradish peroxidase; MWM, Morris water maze; P, postnatal day; PBS, phosphate-buffered saline; PFA, paraformaldehyde; RTPCR, everse transcription polymerase chain reaction; SAL, prenatal saline exposure group; SDS, sodium dodecylsulfate; TBST, Tris-buffered saline–Tween 20.

0306-4522/09 $ - see front matter © 2009 IBRO. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.neuroscience.2009.02.058

330

Y. Wang and T.-Z. Han / Neuroscience 160 (2009) 330 –338

subsequent neurobehavioral defects of learning and memory observed in adolescence. We therefore investigated the effects of prenatal heroin exposure on hippocampusdependent memory, and apoptosis in the mouse hippocampus. The Morris water maze (MWM) was applied to explore the effects of prenatal heroin exposure on spatial learning and memory. Reverse transcription polymerase chain reaction (RT-PCR), Western blotting, and immunohistochemistry were applied to detect changes in the expressions of caspase-3, Bcl-2, and Bax in the hippocampus.

EXPERIMENTAL PROCEDURES Animals and drug BALB/c, 60 nulliparous female (19 –22 g) and 20 male (24 –27 g) mice were obtained from the Fourth Military Medical University (Xi’an, China). Animals were housed in groups of four to five per cage in a room that was maintained at a constant temperature (25 °C) and humidity (40%– 60%). Mice were kept on a 12-h light/dark cycle, with lights on at 8:00 AM, and with free access to food and water. All procedures were carried out in accordance with the National Institutes of Health Guidelines for the Care and Use of Laboratory Animals, and were approved by the local Committee of Animal Use and Protection, and all efforts were made to minimize animal suffering and reduce experimental animal numbers. The drug used in this experiment was heroin hydrochloride (purity ⬎98.5%, product ID No.171206-200614, National Institute for the Control of Pharmaceutical and Biological Products, Beijing, China). The heroin was dissolved in physiological saline at a concentration of 0.5 mg/ml and was administered subcutaneously at a dose of 10 mg/kg.

Establishment of the animal model The establishment of an appropriate animal model is necessary to elucidate the mechanism underlying heroin teratogenicity. This particular model focuses on the specific memory defects caused by heroin exposure during early development. During breeding, one male was housed with four females. The presence of a vaginal plug was designated as embryonic day (E) 0. The pregnant mice in the heroin group received heroin subcutaneously at a dosage of 10 mg/kg/day on E9 –18, while the mice in the saline group were treated with same volume of physiological saline. No necrotic skin lesions were observed at the injection sites. The single heroin dose of 10 mg/kg represents the maximum tolerated exposure that produces neurobehavioral deficits, and is commensurate with continuation of pregnancy, lack of fetal resorption, and postnatal survival, and there was no maternal mortality and the maternal weights did not differ significantly between heroin-injected group and saline-injected group (Shahak et al., 2003). Dams were allowed to deliver naturally. Offspring were grouped as prenatal heroin exposure (HER), prenatal saline exposure (SAL), or control (CON) according to the maternal treatment regimen. There were about 50 to 60 offspring mice in each group, eight of each group left for MWM testing at postnatal day (P) 30, the remaining were killed on P14 for RT-PCR, Western blotting and immunohistochemistry experiment.

MWM testing Mice (n⫽8 from each group) were subjected to a place navigation task at P30 using the MWM. The apparatus consisted of a circular, stainless steel, black-painted swimming pool (122 cm in diameter⫻50 cm high) filled with water at 25–26 °C to a depth of 40 cm. The pool was located in a testing room, on the walls of which were

331

hung some objects that the mice could use as geometric or landmark cues for spatial orientation. All of these prominent extramaze cues remained unchanged throughout the entire testing period. The pool was divided into four equal quadrants, labeled I–IV. A hidden black platform (9 cm in diameter) was placed in the center of quadrant I submerged 1.0 cm below the water’s surface. Before orientation training, animals were habituated to the swimming pool environment. They were placed individually on the platform for 20 s and in the pool for 60 s and allowed to swim. In orientation training, mice were trained to escape from drowning by climbing onto the hidden platform; they were allowed to swim for a maximum trial duration of 120 s, and were placed on the platform for 10 s (reinforcement) if they failed to find it themselves. Mice were given four trials per day for 4 consecutive days. In each trail, mice were gently placed into the pool at the middle of the circular edge in a randomly selected quadrant, with the nose pointing toward the wall. Each training session comprised four trials, with an intertrial interval of 60 s, and was performed routinely between 10:00 AM and 17:00 PM. the intersession interval was ⬎2 h. The escape latency (i.e. the time required for the mouse to find and climb onto the platform) was recorded for each trial. The average of the four trials per training day was recorded. If a mouse failed to find the platform within the testing time or if it stayed on the platform for less than 3 s (and thus considered to be continuing its search for the target), a score latency of “120 s” was awarded. The mice applied four strategies to find the platform: (1) marginal: swimming along the pool edge; (2) random: swimming randomly; (3) taxis: swimming around but toward the platform area; (4) straight: swimming straight toward the platform. The straight and taxis methods are more efficient ways of finding the platform, while marginal and random are inferior strategies that are less efficient for escaping from drowning. The strategy applied by the mice during orientation was recorded. In probe testing, the platform was removed from the pool and each mouse was placed into the pool from the opposite quadrant. The number of times the mouse crossed the platform was also recorded.

RT-PCR After each mouse was anesthetized and decapitated, the hippocampus was dissected from the brain and snap-frozen in liquid nitrogen for later use. Total RNA was extracted by homogenization in TRIzol reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s protocol, and treated with DNase I (Promega, Madison, WI, USA) to remove genomic DNA. Each RNA sample was qualitatively evaluated by agarose gel electrophoresis. A 1.6-␮g sample of total RNA was reverse-transcribed into cDNA and then amplified using the reagents and the protocol of the SuperScript one-Step RT-PCR with platinum Taq (Invitrogen, Carlsbad, CA, USA). The RT reaction and PCR amplification was performed with a GeneAmp PCR System 2400 (PerkinElmer, Boston, MA, USA). The amplification program was set as follows: the first cycle was 5 min at 60 °C and 30 min at 42 °C, followed by 30 cycles of denaturation at 94 °C for 30 s, annealing at 55 °C for 30 s, elongation at 72 °C for 1 min, with a final extension at 72 °C for 10 min, then immediate storage at 4 °C. For the PCR reaction, the primers for caspase-3 were 5=-ggagctggactgtggcattga-3= (forward) and 5=-cagttctttcgtgagcatgga-3= (reverse), 322 bp; the Bcl-2 primers were 5=-ggtgcagcgatttcgtacc-3= (forward) and 5=-aagaggatgagcagtcagagg-3= (reverse), 206 bp; the Bax primers were 5=-tcccacataactccctcgaca-3= (forward) and 5=-ggcgaagccagcgagaagtccc-3= (reverse), 228 bp; and the ␤-actin primers were 5=-gtgggccgctctaggcaccaa-3= (forward) and 5=-ctctttgatgtcacgcacgatttc-3= (reverse), 540 bp. PCR was performed following a standard protocol. Samples were amplified for 30 cycles at an annealing temperature of 55 °C for caspase-3, Bcl-2, Bax, and ␤-actin.

332

Y. Wang and T.-Z. Han / Neuroscience 160 (2009) 330 –338

The PCR products (5 ml) were separated on 2% agarose gels with 1⫻ Tris-acetate–EDTA buffer. The amounts of RT-PCR products in the ethidium-bromide-stained gel were quantified by analyzing the band intensity with a GDS-8000 System BioImaging System (UVP, Upland, CA, USA). The mRNA levels of caspase-3, Bcl-2, and Bax products were related to ␤-actin mRNA values and normalized as a percentage of those of the control mouse (which was taken as 100%). All experiments were repeated three times.

Western blotting Tissue samples from the hippocampus were homogenized in a buffer containing 20 mM Tris–HCl, pH 6.8, 1 mM EDTA, 1% sodium dodecylsulfate (SDS), 1 mM phenylmethylsulfonyl fluoride, and 1⫻ protease inhibitor cocktail (Roche, Beijing, China). Equal amounts of whole-cell protein from each sample were loaded onto a 15% SDS–polyacrylamide gel, which was then subjected to electrophoresis, and transferred to a polyvinylidene difluoride membrane (Bio-Rad Laboratories, Beijing, China) using a mini trans-blot electrophoresis transfer cell (Bio-Rad). Nonspecific binding of antibodies was prevented by incubating membranes in Tris-buffered saline–Tween 20 (TBST) buffer (10 mM Tris–HCl, pH 7.6; 150 mM NaCl; and 0.05% Tween 20) containing 5% nonfat milk powder. The blots were incubated overnight at 4 °C with the primary antibody diluted in TBST containing 2% nonfat milk powder. The following antibodies were used: caspase-3 p17 (diluted 1:1000; Santa Cruz Biotechnology, CA, USA), Bcl-2 (diluted 1:800, Santa Cruz Biotechnology, CA, USA), and Bax (diluted 1:1000, Santa Cruz Biotechnology, CA, USA). After extensive washing with TBST, the blots were incubated with a horseradish peroxidase (HRP)–labeled secondary antibody for 1 h at room temperature. After washing, immunoreactivity was detected with an enhanced chemiluminescence (ECL) system. For immunoblot detection, immunoreactive bands were visualized with the Super Signal West Pico ECL kit (Pierce, Rockford, IL, USA). Band intensities were quantified using a densitometer analysis system (Flurochem 9900 –50, Alpha Innotech, San Leandro, CA, USA). Equal protein loading was confirmed by staining the gel with Coomassie Blue and probing with ␤-actin antibody (Biosynthesis Biotechnology, Beijing, China).

Immunohistochemistry Mice were anesthetized with pentobarbital sodium and killed by intracardiac perfusion with 0.9% NaCl followed by phosphatebuffered saline (PBS, pH 7.4)– buffered 4% paraformaldehyde (PFA) solution for 12 min. Following perfusion, the tissues were obtained by dissection, and were postfixed in PBS-buffered 4% PFA solution overnight. Tissue blocks were dehydrated in 70% ethanol for at least 12 h at 4 °C before being embedded in paraffin. Coronal sections (7 ␮m thick) were obtained from each block and floated onto glass slides. These sections were deparaffinized and then subjected to an antigen-retrieval protocol, which involved incubating the sections in 10-mM citrate buffer (pH 6.0) in a steam-set rice cooker for 30 min. Sections were then oxidized with 3% H2O2 for 15 min at room temperature to block endogenous peroxidase activity. Antibody labeling was carried out as follows (Majumdar et al., 2008): Sections were incubated first in a blocking buffer containing PBS supplemented with 1% bovine serum albumin, 0.2% evaporated milk, and 0.3% Triton X-100 for 15 min at room temperature, then overnight at 4 °C in the blocking buffer with anti-caspase-3 p17 polyclonal antibody (diluted 1:600; Santa Cruz Biotechnology, CA, USA). After being washed with PBS, sections were incubated in blocking buffer containing HRP-conjugated goat antirabbit antibody (diluted 1:1000, Santa Cruz Biotechnology, CA, USA) for 1 h at room temperature, followed by HRP-coupled streptavidin. The resulting chromogen was detected using diaminobenzidine– hydrogen peroxide (DAB Map kit, Tucson, AZ, USA). Slides were then counterstained using hematox-

ylin, dehydrated through graded alcohols and xylene, and mounted with Permount mounting medium. Sections were examined and photographed, and analyzed with a computer-assisted image-analysis system (Image-ProPlus 5.0, Media Cybernetics, MD, USA). Contrast was enhanced using Adobe Photoshop (Adobe Systems, San Jose, CA, USA). The positive areas were assessed in at least 16 randomly selected tissue sections from each group studied. Sections serving as negative controls were incubated with PBS instead of the primary antiserum. All immunohistochemical labeling was performed at least twice.

Data analysis All data are mean and SEM values. All statistical measurements were carried out using SPSS PC version 13.0 (SPSS, USA). For the MWM experiments, escape latency data were analyzed by repeated-measures analysis of variance (ANOVA) followed by Tukey’s post hoc test. Strategy data were analyzed using the Mann–Whitney nonparametric test, and the number of platform crossings and other normally distributed data were assessed by one-way ANOVA. Comparisons that yielded a P-value of ⬍0.05 were considered statistically significant in all cases.

RESULTS Prenatal heroin exposure leads to deficits in MWM performance All latencies during maze acquisition for all experimental groups are shown in Fig. 1. The escape latencies of all three groups decreased significantly throughout the trials (F(3,112)⫽20.86, P⬍0.05), which indicates that all mice developed a memory of the hidden platform. Further analysis of these data revealed significant effects of treatment condition (F(1,14)⫽34.16, P⬍0.05 in males; F(1,14)⫽22.15, P⬍0.05 in females) but no significant interactions of treatment⫻sex (F(1,15)⫽0.17, P⬎0.05 in males; F(1,15)⫽ 1.84, P⬎0.05 in females). Post hoc group differences are shown in Fig. 1A, B. The latencies were much longer in the HER than in both the SAL and CON throughout the trials, indicating that prenatal heroin treatment leads to poor performance in both sexes. The strategies used in orientation are shown Fig. 1C, D. The heroin-exposed offspring more frequently used the less efficient marginal and random strategies to find the hidden platform (P⬍0.05). Post hoc group differences are shown in Fig. 1C, D. The number of platform crossings made by the mice is shown in Fig. 1E, F. Analysis of the crossings revealed a significant treatment effect on both males (F(1,14)⫽17.97, P⬍0.05) and females (F(1,14)⫽4.96, P⬍0.05). The heroin-exposed mice crossed over the target area significantly less often than did those that had not been exposed. Post hoc group differences are shown in Fig. 1E, F. Prenatal heroin exposure alters the expression of apoptosis genes in the hippocampus The mRNA expressions of caspase-3, Bcl-2, and Bax were detected by RT-PCR. The mRNA expressions of caspase-3 and Bax increased significantly, while that of Bcl-2 mRNA decreased significantly in the HER group, compared to the SAL and CON groups (P⬍0.05, Fig. 2A, B).

Y. Wang and T.-Z. Han / Neuroscience 160 (2009) 330 –338

333

Fig. 1. Poor performance of mice in the MWM after prenatal heroin exposure. (A, B) Escape latency in the four testing sessions during orientation training in the CON (black circles), SAL (white circles), and HER (black triangles) groups (data for males and females are shown in A and B, respectively; one-way repeated-measures ANOVA with a post hoc Tukey test). (C, D) Time for each of the four strategies used during orientation in the CON (black bars), SAL (gray bars), and HER (dark bars) groups (data for males and females are shown in C and D, respectively; Mann–Whitney nonparametric test). (E, F) Number of platform crossings in the probe test in the CON, SAL, and HER groups (data for males and females are shown in E and F, respectively; one-way ANOVA with a post hoc Tukey test). Data are mean and SEM values. * P⬍0.05 as compared with the SAL group; # P⬍0.05 as compared with the CON group.

334

Y. Wang and T.-Z. Han / Neuroscience 160 (2009) 330 –338

Total cell density was assessed by counting the number of hematoxylin-stained cell nuclei (Fig. 4); caspase-3positive neurons appeared yellow or brown. The findings are expressed as the ratio of positive caspase-3-stained neurons (pyramidal neurons and dentate granule cells) to total neurons. There was a significant main effect of treatment on the density of caspase-3-reactive neurons (F(1,14)⫽107.35, P⬍0.05). The treatment main effect reflected the consistent increase in the density of caspase3-reactive neurons in HER mice. The expression of activated caspase-3-immunoreactive neurons in the HER group increased significantly in the DG and CA1 areas of the hippocampus, while their expression remained unchanged in area CA3, compared with the SAL and CON groups (P⬍0.05, Fig. 5), it indicated significant effect of heroin exposure can be attributed to the higher density of caspase-3-reactive neurons in the DG and area CA1 of the hippocampus, rather than in area CA3. The extent of the damage (i.e. increase in expression of activated caspase3-immunoreactive neurons) was greater in the DG than in area CA1.

Fig. 2. Effects of prenatal heroin exposure on the expressions of caspase-3, Bcl-2, and Bax mRNA in mice. (A) RT-PCR to detect the mRNA levels of caspase-3, Bcl-2, and Bax in the hippocampus (n⫽5). (B) Changes in caspase-3, Bcl-2, and Bax immunoreactivity in the HER and SAL groups expressed relative to those in the CON group (defined as 100%). Compared with the SAL and CON groups, the expressions of caspase-3 and Bax mRNA increased significantly, while that of Bcl-2 mRNA decreased significantly in the HER group. Data are mean and SEM values. * P⬍0.05 as compared with the SAL group; # P⬍0.05 as compared with the CON group (one-way ANOVA with a post hoc Tukey test).

Prenatal heroin exposure alters the expression of apoptosis proteins in the hippocampus The protein expressions of caspase-3 (17 kDa), Bcl-2, and Bax were determined by Western blot analysis. A higherintensity band of Bax and caspase-3 (17 kDa) protein and a lower-intensity band of Bcl-2 protein were observed in the HER group, compared with the SAL and CON groups. The expressions of Bax and caspase-3 (17 kDa) protein increased significantly, while that of Bcl-2 protein decreased significantly in the HER group (P⬍0.05, Fig. 3A, B). Prenatal heroin exposure induces caspase-3 activation in the hippocampus No caspase-3 neurons with false-positive staining were found in the negative controls (blank sections), which were processed through the immunohistochemistry procedure without the primary anti-caspase-3 antibody. Profiles of caspase-3-reactive neurons displayed typical neuronal morphology with caspase-3-positive cell bodies and processes. Fig. 4 shows a representative caspase-3-stained section of three areas of the hippocampal formation: dentate gyrus (DG, Fig. 4a– c), cornu ammonis (CA) 1 (d–f), and CA3 (g–i).

Fig. 3. Effects of prenatal heroin exposure on the expressions of caspase-3 (17 kDa), Bcl-2, and Bax protein in mice. (A) Western blotting to detect the protein levels of caspase-3 (17 kDa), Bcl-2, and Bax in the hippocampus (n⫽5). (B) Changes in caspase-3, Bcl-2, and Bax immunoreactivity in the HER and SAL groups are expressed relative to those in the CON group (defined as 100%). Compared with the SAL and CON groups, the expressions of caspase-3 (17 kDa) and Bax protein increased significantly, and that of Bcl-2 protein decreased significantly in the HER group. Data are mean and SEM values. * P⬍0.05 as compared with the SAL group; # P⬍0.05 as compared with the CON group (one-way ANOVA with a post hoc Tukey test).

Y. Wang and T.-Z. Han / Neuroscience 160 (2009) 330 –338

335

Fig. 4. Effects of prenatal heroin exposure on the expression of caspase-3-positive neurons in the hippocampus. Immunohistochemistry to detect activated caspase-3-immunoreactive neurons in the DG (a– c), area CA1 (d–f), and area CA3 (g–i) in the CON (a, d, g), SAL (b, e, h), and HER (c, f, i) groups (n⫽5). Caspase-3-positive cells are indicated by the black arrowheads. Scale bar⫽50 ␮m. For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.

DISCUSSION The responses of hippocampal cells to heroin during neurogenesis and their contribution to the pathogenesis of postnatal learning and memory defects were investigated herein. These defects might be at least partly attributable to

Fig. 5. Quantitative analysis of caspase-3-positive neurons in the hippocampus. Changes in caspase-3 immunoreactivity in the HER and SAL groups are expressed relative to that in the CON group (defined as 100%). Compared with the SAL and CON groups, the immunoreactivity of caspase-3 increased significantly in the DG and area CA1 of the hippocampus, while no significant changes were observed in area CA3. Data are mean and SEM values (n⫽5). * P⬍0.05 as compared with the SAL group; # P⬍0.05 as compared with the CON group (one-way ANOVA with a post hoc Tukey test).

disturbed apoptotic pathways. Our findings indicate that prenatal heroin exposure during fetal development might promote cell death in the hippocampus by affecting specific proteins in the apoptotic signal-transduction pathways. The ability of heroin to induce apoptosis in neurons can be reflected by the exposure-induced expressions of the key apoptotic regulators caspase-3, Bcl-2 and Bax. In addition, enhanced activation of caspase-3 in the DG and CA1 subareas of the hippocampus was observed, whereas no such enhancement was observed in subarea CA3. It has been reported that prenatal exposure to a variety of drugs is associated with an increased risk of regulatory dysfunction and neuropsychological difficulties (Papageorgiou et al., 2004; Slinning, 2004) and that prenatal exposure to heroin can cause neurobehavioral defects of cognitive functions, including impairment of learning and memory (Yanai et al., 2000). Heroin abuse could damage the fetal hippocampus by altering the circulating levels of maternal hormonal factors and impairing the developing cells in the fetus. These changes in maternal hormones could influence, for example, the weight, nutrition, and development of the offspring. In the present study, no significant differences were found in developmental indices such as weight, eye opening, and tooth eruption among the HER, SAL, and CON groups. We have demonstrated that the deficits of learning and memory, and the alterations in the expression of key apoptotic genes in the hippocampus might be due to direct heroin-induced impairment of the developing cells.

336

Y. Wang and T.-Z. Han / Neuroscience 160 (2009) 330 –338

MWM was first introduced as an instrument designed to test spatial memory, and is particularly sensitive to the effects of hippocampal lesions in rats (Morris, 1984). It has been suggested that hippocampal cells are the primary substrate of spatial memory, which underlies the spatial navigation processes involved in hidden-platform MWM learning (Poucet et al., 2000). In the present study, prenatally heroin-exposed mice (i.e. the HER group) performed poorly in the MWM test with regard to finding the platform, using effective strategies, and crossing the platform, compared to the SAL and CON groups at P30. This suggests that the heroin-induced learning and memory deficits occurred in early adolescence. Yanai et al. (2000) also reported memory defects in adult mice from the same animal model, confirming the long-term influences on offspring from transitory exposure to heroin in utero. Tissue samples were taken at P14 for several reasons. Firstly, studies conducted in rodents have revealed that naturally occurring cell death takes place during the embryonic, fetal, and early postnatal stages. In the hippocampus and cortex of the developing rat brain, this process occurs during the first 2 postnatal weeks (Sandau et al., 2006). By P20, the event has largely subsided to adult levels. Secondly, most synapses develop postnatally; studies of synaptogenesis have demonstrated that the period of early growth of synapses is very important for the onset of cognitive function (Nelson, 2005). In the hippocampus, most synapses occur on dendritic shafts during the first 2 postnatal weeks (Fiala et al., 1998). Moreover, apoptosis is initiated by activating or altering the expression of proapoptotic members of the Bcl-2 family during this period (Sandau et al., 2006). It is well established that hippocampal neurogenesis is important for learning and memory (Kempermann et al., 2004). The major subareas of the hippocampal formation—CA1 and CA3, which contain densely packed pyramidal cells, and the DG, which contains a tightly packed layer of small granule cells—are clearly separate from each other (Bakker et al., 2008). During hippocampal neurogenesis in mice, the pyramidal neurons are generated at E15–16 for CA1 and at E14 –15 for CA3 (Soriano et al., 1989a,b). The genesis of the DG granule cells commences at E20, reaching a peak during the first postnatal week (Altman and Bayer, 1990; Bayer, 1980). The migration during hippocampal neurogenesis of a population of large hippocampal neurons has also been described, suggesting that they are interneurons. Various studies have indicated that most of the interneurons from areas CA1 and CA3 are generated at E12–13, whereas most DG interneurons are generated at E13–14 (Altman and Bayer, 1990; Bayer, 1980; Soriano et al., 1989a,b). These studies suggest that the hippocampus is susceptible to cell damage if it is exposed to heroin during neurogenesis. In our experiment, significant changes in caspase-3 immunoreactivity were found in the hippocampus at P14. There is widespread agreement that the hippocampal system is intimately involved in navigational spatial–relational learning. Lesion experiments have shown that areas CA1 and CA3 are differentially involved in behavior (Jarrard,

1993), although some overlapping does exist between these two subareas. In our own animal model, 9 days of prenatal heroin exposure at 10 mg/kg/day resulted in apoptotic activation predominantly in the DG and area CA1, and both of these hippocampal subareas contained a greater abundance of apoptotic cells than area CA3. Therefore, it is reasonable to assume that prenatal heroin exposure results in enhancement of the apoptotic pathways primarily in the DG and CA1 regions. It has been suggested that areas CA1 and CA3, and the DG exhibit differential vulnerability or tolerance to particular circumstances. For example, area CA1 is the most vulnerable to ischemic insult, while area CA3 and the DG are relatively resistant to it (Schmidt et al., 1991); anoxia can lead to certain electrophysiological changes in area CA1, but not in area CA3 (Tropp et al., 2006). Our findings indicate that in mice, the neurons in area CA1 and the DG are more vulnerable than those in area CA3 to heroin exposure during intrauterine development. Data from various studies have shown that chronic exposure to opioid drugs interferes with learning and memory through the activation of apoptotic pathways, by increasing the brain expression of proapoptotic factors (Boronat et al., 2001; Garcia-Fuster et al., 2003; Mao et al., 2002; Tramullas et al., 2008). Our present findings concur with these studies. The levels of mRNA and protein expressions of caspase-3 and Bax increased significantly, while that of Bcl-2 decreased significantly in the HER group compared to the SAL and CON groups. The apoptotic pathway comprises three phases: initiation, control, and effector. The control phase involves crosstalk among various apoptotic regulators that have pro- and antiapoptotic functions, the most important proteins being those that originate from the Bcl-2 family (Karbowski et al., 2006). Members of the Bcl-2 family can be divided into two groups—the repressors (e.g. Bcl-2 and Bcl-XL) and promoters (e.g. Bax and Bad) of apoptosis— depending on their role in its regulation. One of their critical roles is adjusting the number of postmitotic neurons (Motoyama et al., 1995) and neural precursors (Lindsten et al., 2003) during the development of the nervous system. The effector phase of apoptosis is characterized by cascade activation of the caspases, which results in the cleavage of a variety of cellular proteins, leading to the orderly demise of the cell (Widlak and Garrard, 2006). Caspase-3 appears to be the most abundant of the caspases and is involved in the convergence of all caspase-mediated pathways related to apoptosis. When the neurons in some brain regions are removed through apoptosis, caspase-3 is highly upregulated, playing an essential role in this process. Procaspase-3 is highly expressed from E17 to P7, decreasing after P14 in the neonatal rat brain (Kuroso et al., 2004). Thus, caspase-3 is more likely to be affected when the embryo receives a teratogenic stimulus during neurogenesis. The immunocytochemical detection of caspase-3 can reveal the location and degree of apoptotic changes, reflecting the balance of apoptosis occurring in the hippocampal subfields. Our results support these findings. The number of caspase-3-immunoreactive neurons in-

Y. Wang and T.-Z. Han / Neuroscience 160 (2009) 330 –338

creased significantly in target areas of hippocampus when mice were exposed to heroin on E9 –18.

CONCLUSION Our results show that prenatal heroin exposure during E9 –18 can affect spatial learning and memory in adolescent mice, and that this postnatal neurobehavioral defect might be attributable to alterations in the apoptotic pathways in the developing hippocampus. Acknowledgments—The project was supported by the National Natural Science Foundation of China (No. 30500411; 30572089; 30772081).

REFERENCES Altman J, Bayer SA (1990) Migration and distribution of two populations of hippocampal granule cell precursors during the perinatal and postnatal periods. J Comp Neurol 301:365–381. Bakker A, Kirwan CB, Miller M, Stark CE (2008) Pattern separation in the human hippocampal CA3 and dentate gyrus. Science 319:1640 –1642. Bayer SA (1980) Development of the hippocampal region in the rat. I. Neurogenesis examined with 3H-thymidine autoradiography. J Comp Neurol 190:87–114. Boronat MA, Garcia-Fuster MJ, Garcia-Sevilla JA (2001) Chronic morphine induces upregulation of the pro-apoptotic Fas receptor and downregulation of the antiapoptotic Bcl-2 oncoprotein in rat brain. Br J Pharmacol 134:1263–1270. Danial NN, Korsmeyer SJ (2004) Cell death: critical control points. Cell 116:205–219. Emeterio EP, Tramullas M, Hurle MA (2006) Modulation of apoptosis in the mouse brain after morphine treatments and morphine withdrawal. J Neurosci Res 83:1352–1361. Fiala JC, Feinberg M, Popov V, Harris KM (1998) Synaptogenesis via dendritic filopodia in developing hippocampal area CA1. J Neurosci 18:8900 – 8911. Garcia-Fuster MJ, Ferrer-Alcon M, Miralles A, Garcia-Sevilla JA (2003) Modulation of Fas receptor proteins and dynamin during opiate addiction and induction of opiate withdrawal in rat brain. Naunyn Schmiedebergs Arch Pharmacol 368:421– 431. Ghobrial IM, Witzig TE, Adjei AA, Cancer C (2005) Targeting apoptosis pathways in cancer therapy. J Clin 55:178 –194. Harburg GC, Hall FS, Harrist AV, Sora I, Uhl GR, Eisch AJ (2007) Knockout of the mu opioid receptor enhances the survival of adultgenerated hippocampal granule cell neurons. Neuroscience 144:77– 87. Hutchings DE (1982) Methadone and heroin during pregnancy: a review of behavioral effects in human and animal offspring. Neurobehav Toxicol Teratol 4:429 – 434. Igney FH, Krammer PH (2002) Death and anti-death: tumour resistance to apoptosis. Nat Rev Cancer 2:277–288. Jarrard LE (1993) On the role of the hippocampus in learning and memory in the rat. Behav Neural Biol 60:9 –26. Karbowski M, Norris KL, Cleland MM, Jeong SY, Youle RJ (2006) Role of Bax and bak in mitochondrial morphogenesis. Nature 443:658 – 662. Kempermann G, Wiskott L, Gage FH (2004) Functional significance of adult neurogenesis. Curr Opin Neurobiol 14:186 –191. Kuroso K, Saeki M, Kamisaki Y (2004) Formation of high molecular weight caspase-3 complex in neonatal rat brain. Neurochem Int 44:199 –204. Lee KK, Li FC, Yung WT, Kung JL, Ng JL, Cheah KS (1994) Influence of digits, ectoderm, and retinoic acid on chondrogenesis by mouse interdigital mesoderm. Dev Dyn 201:297–309.

337

Lindsten T, Golden JA, Zong WX, Minarcik J, Harris MH, Thompson CB (2003) The proapoptotic activities of Bax and Bak limit the size of the neural stem cell pool. J Neurosci Res 23:11112–11119. Majumdar D, Maunsbach AB, Shacka JJ, Williams JB, Berger UV, Schultz KP, Harkins LE, Boron WF, Roth KA, Bevensee MO (2008) Localization of electrogenic Na/bicarbonate cotransporter NBCe1 variants in rat brain. Neuroscience 155:818 – 832. Mao J, Sung B, Ji RR, Lim G (2002) Neuronal apoptosis associated with morphine tolerance: evidence for an opioid-induced neurotoxic mechanism. J Neurosci Res 22:7650 –7661. Morris R (1984) Developments of a water-maze procedure for studying spatial learning in the rat. J Neurosci Res Methods 11:47– 60. Motoyama N, Wang F, Roth KA, Sawa H, Nakayama KI, Nakayama K, Negishi I, Senju S, Zhang Q, Fujii S (1995) Massive cell death of immature hematopoietic cells and neurons in Bcl-x-deficient mice. Science 267:1506 –1510. Nelson CA (2005) Neural development and lifelong plasticity. In: Applied developmental science. An advanced textbook (Lerner RM, Jacobs F, Wertlieb D, eds), pp 42– 43. Thousand Oaks, CA: Sage Publications. Oliveira MT, Rego AC, Macedo TR, Oliveira CR (2003) Drugs of abuse induce apoptotic features in PC12 cells. Ann N Y Acad Sci 1010:667– 670. Papageorgiou CC, Liappas IA, Ventouras EM, Nikolaou CC, Kitsonas EN, Uzunoglu NK, Rabavilas AD (2004) Long-term abstinence syndrome in heroin addicts: indices of P300 alterations associated with a short memory task. Prog Neuropsychopharmacol Biol Psychiatry 28:1109 –1115. Poucet B, Save E, Lenck-Santini PP (2000) Sensory and memory properties of hippocampal place cells. Rev Neurosci 11:95–111. Sandau US, Handa RJ (2006) Localization and developmental ontogeny of the pro-apoptotic Bnip3 mRNA in the postnatal rat cortex and hippocampus. Brain Res 1100:55– 63. Schmidt-Kastner R, Freund TF (1991) Selective vulnerability of the hippocampus in brain ischemia. Neuroscience 40:599 – 636. Shahak H, Slotkin TA, Yanai J (2003) Alterations in PKC␥ in the mouse hippocampus after prenatal exposure to heroin: a link from cell signaling to behavioral outcome. Brain Res Dev Brain Res 140:117–125. Slinning K (2004) Foster placed children prenatally exposed to polysubstances—attention-related problems at ages 2 and 4 1/2. Eur Child Adolesc Psychiatry 13:19 –27. Soriano E, Cobas A, Fairen A (1989a) Neurogenesis of glutamic acid decarboxylase immunoreactive cells in the hippocampus of the mouse. I. Regio superior and regio inferior. J Comp Neurol 281:586 – 602. Soriano E, Cobas A, Fairen A (1989b) Neurogenesis of glutamic acid decarboxylase immunoreactive cells in the hippocampus of the mouse. II. Regio superior and regio inferior. J Comp Neurol 281:603– 611. Steingart RA, Silverman WF, Barron S, Slotkin TA, Awad Y, Yanai J (2000) Neural grafting reverses prenatal drug-induced alterations in hippocampal PKC and related behavioral deficits. Brain Res Dev Brain Res 125:9 –19. Tramullas M, Martinez-Cue C, Hurle MA (2008) Chronic administration of heroin to mice produces upregulation of brain apoptosis-related proteins and impairs spatial learning and memory. Neuropharmacology 54:640 – 652. Tropp SJ, Chrobak JJ, Quirk MC, Oler JA, Markus EJ (2006) Differential behavioral state-dependence in the burst properties of CA3 and CA1 neurons. Neuroscience 141:1665–1677. Tschacher W, Haemmig R, Jacobshagen N (2003) Time series modeling of heroin and morphine drug action. Psychopharmacology (Berl) 165:183–193. Vogel KS (1994) Development of trophic interactions in the vertebrate peripheral nervous system. Mol Neurobiol 7:363–382.

338

Y. Wang and T.-Z. Han / Neuroscience 160 (2009) 330 –338

Widlak P, Garrard WT (2006) The apoptotic endonuclease DFF40/ CAD is inhibited by RNA, heparin and other polyanions. Apoptosis 11:1331–1337. Wilson GS (1989) Clinical studies of infants and children exposed prenatally to heroin. Ann N Y Acad Sci 562:183–194. Yanai J, Avraham Y, Levy S, Maslaton J, Pick CG, Rogel-Fuchs Y, Zahalka EA (1992) Alterations in septohippocampal cholin-

ergic innervations and related behaviors after early exposure to heroin and phencyclidine. Dev Brain Res 69:207– 214. Yanai J, Steingart RA, Snapir N, Gvaryahu G, Rozenboim I, Katz A (2000) The relationship between neural alterations and behavioral deficits after prenatal exposure to heroin. Ann N Y Acad Sci 914:402– 411.

(Accepted 26 February 2009) (Available online 9 March 2009)