OUP user menu

Gestational nicotine treatment modulates cell death/survival-related pathways in the brains of adolescent female rats

Jinxue Wei, Ju Wang, Jennifer B. Dwyer, Jamie Mangold, Junran Cao, Frances M. Leslie, Ming D. Li
DOI: http://dx.doi.org/10.1017/S1461145710000416 91-106 First published online: 1 February 2011


Gestational exposure to nicotine affects brain development, leading to numerous behavioural and physiological deficits in the offspring during adolescence. To analyse the molecular mechanisms underlying these effects, a pathway-focused oligonucleotide microarray was used to determine gene expression profiles in five brain regions (i.e. amygdala, prefrontal cortex, nucleus accumbens, periventricular nucleus of the hypothalamus, and caudate putamen CPu) of adolescent rats that received nicotine or saline during gestation. Following appropriate statistical and Gene Set Enrichment Analyses, 24 cell death/survival-related pathways were found to be significantly modulated by gestational nicotine. On the basis of their biological functions, these pathways can be classified into three categories: growth factor, death receptor, and kinase cascade. We employed a quantitative real-time PCR array to verify the findings by measuring the expression of 29 genes involved in cell death/survival-related pathways. Together, our findings indicate that gestational nicotine exposure has significant effects on gene expression in cell death/survival-related pathways in the brains of adolescent offspring. Such effects appear to be brain region-specific and are realized through regulation of the expression of growth factors and receptors, caspases, kinases, and transcription factors. On the basis of these findings, we offer a hypothetical model to explain how gestational nicotine exposure may affect cell death and survival in the brains of adolescent offspring by regulating the balance between growth-factor and death-receptor pathways.

Key words
  • Brain development
  • cell death
  • cell survival
  • fetus
  • gene expression
  • gestational nicotine exposure
  • rat


Maternal smoking during pregnancy has been linked to physiological and neurobehavioural deficits in the offspring (Dwyer et al. 2008), as well as a higher fetal mortality rate, low birth weight, and sudden infant death syndrome (Dwyer et al. 2008; Leonardi-Bee et al. 2008). Maternal smoking also correlates highly with neurobehavioural disorders that manifest in the offspring during childhood and adolescence. The late onset of these disorders, including cognitive dysfunction, conduct disorder, and greater risk of substance abuse, indicates that prenatal smoking has long-lasting effects, and highlights the need to determine how maternal smoking alters the brain during adolescence (Dwyer et al. 2008; Franke et al. 2007, 2008; Leonardi-Bee et al. 2008).

Gestational nicotine exposure in animals is the approach most commonly used to model maternal smoking in humans (Coleman, 2008; Dwyer et al. 2008; Franke et al. 2007). Depending on the dose, gestational exposure in rodents can result in low birth weight, impaired cognitive performance, and enhanced locomotor activity, similar to the deficits associated with prenatal exposure to tobacco smoke in humans (Eppolito & Smith, 2006; Franke et al. 2007). This animal model also shows extensive neurochemical deficits that manifest during adolescence (Franke et al. 2007; Slotkin, 2008), paralleling the time-course of clinical deficits associated with maternal smoking. In rats, during adolescence, conservatively defined as postnatal days 28–42 (PD 28–42), the central nervous system undergoes extensive cell growth and pruning (Spear, 2000). These findings strongly indicate that gestational exposure to nicotine disturbs prenatal brain development, with some consequences emerging only as the impacted neural circuitry matures.

Depending on the dose, developmental stage, and tissue or cell type, nicotine can be both pro- and anti-apoptotic (Dwyer et al. 2008; Zeidler et al. 2007). In animals, nicotine exerts its long-term effects on brain development by modulating certain aspects of neurite growth (Dwyer et al. 2008), switching neuronal proliferation to differentiation, and regulating cell death (Slotkin et al. 1987). In vitro, nicotine promotes cell survival and inhibits multiple stimuli-induced death in neurons (Zeidler et al. 2007). Additionally, nicotine can inhibit apoptosis in malignant, immune, endothelial, and epithelial cells, indicating its role in regulating processes related to cancer, the immune response, and angiogenesis (Egleton et al. 2008; Zeidler et al. 2007). Furthermore, both human and animal studies have revealed an inverse correlation between nicotine intake and some neurodegenerative diseases, including Parkinson's disease and Alzheimer's disease, which is indicative of the neuroprotective and anti-apoptotic effects of nicotine (Picciotto & Zoli, 2008; Zeidler et al. 2007). Conversely, pro-apoptotic effects of nicotine have been observed both in vivo and in vitro, especially in developing brain (Abreu-Villaca et al. 2004; Qiao et al. 2003; Roy et al. 1998, 2002). Prenatal nicotine exposure induces extensive apoptosis in the brain of the offspring during exposure (Abreu-Villaca et al. 2004; Roy et al. 1998). Although previous studies have provided evidence that nicotine regulates cell death/survival processes by modulating expression or activation of nicotinic acetylcholine receptors (nAChRs), growth factors, kinases, bcl2 family members, and other cell cycle- and apoptosis-related genes (Zeidler et al. 2007), the molecular mechanisms underlying the pro- and anti-apoptosis effects are largely unknown. Neuronal proliferation and death are inherent in brain development during adolescence, which is thought to be necessary for the transition to adult-like behaviour (Spear, 2000). Despite the clinical prevalence of maternal smoking-related deficits during this critical phase, no thorough analysis of the effects of gestational nicotine on cell survival and cell death pathways in the adolescent brain has been reported.

DNA microarray technology, a high-throughput approach for gene expression profiling, has been used to study nicotine effects on expression in both humans and animals (Li et al. 2002; Zeidler et al. 2007). Moreover, this approach can be used to investigate how a group of functionally related genes respond to a treatment (Kane et al. 2004). Previously, through this approach, we identified several biochemical pathways, such as phosphatidylinositol (PI) signalling (Li et al. 2002) and protein modification and degradation (Kane et al. 2004), that are modulated significantly by nicotine. Moreover, we showed that regulation of these pathways by nicotine depends on the brain region and the duration of nicotine treatment (Kane et al. 2004; Konu et al. 2001; Li et al. 2004; Wang et al. 2007, 2008).

To gain a systematic insight into the apoptotic effects of gestational nicotine exposure, we employed a newly developed pathway-focused oligonucleotide microarray to measure gene expression in five brain regions of gestational saline- or nicotine-treated adolescent rats. Because of the known complicated sexual dimorphism, we used female offspring only. We found that pathways related to death receptors, growth factors, and kinase cascades are among the major adolescent targets of prenatal nicotine exposure.

Materials and methods

Animals and nicotine treatment

Animals were maintained in a temperature-controlled (21°C) and humidity-controlled (50%) room on a 12-h light/dark cycle (lights on 07:00 hours) with food and water available ad libitum. They were housed in an AAALAC-accredited vivarium maintained by UCI University Laboratory Animals Resources personnel. Twenty timed pregnant Sprague–Dawley rats (Charles River, USA) arrived on gestational day (GD) 2 as pairs and were randomly assigned to either saline or nicotine treatment. On GD 4, the pregnant dams were implanted subcutaneously with osmotic minipumps (Alzet model 2002, flow rate 51 µl/d) containing either nicotine (3 mg base/kg.d) or saline as previously described (Park et al. 2006). Blood concentrations resulting from this dose of nicotine are equivalent to those found in humans who smoke about 1½ packs of cigarettes per day (Matta & Elberger, 2007). All dams were weighed daily; weight gain was not significantly altered by nicotine. The minipump remained until pup delivery on GD 22, with drug delivery estimated to end on GD 18–20. To minimize any effects of nicotine withdrawal on maternal behaviour, newborn litters were culled to ten and cross-fostered on treatment-naive surrogate dams.

Pups were weighed daily to ensure proper growth and development. Consistent with our previous report, the body weights of pups in the nicotine group was not significantly different from those in the control group, indicating nicotine did not retard growth (Franke et al. 2007). Animals were weaned on PD 21 and housed with same-sex littermates. As previously noted, in order to simplify the experimental procedure and related analysis, only female pups were used.

On PD 35, after reaching their mid-adolescence, the young female animals were sacrificed via rapid decapitation, and their brains were removed immediately. Using a rat brain matrix, 2-mm slices were cut that contained the prefrontal cortex (PFC), caudate putamen (CPu), nucleus accumbens (NAc), periventricular nucleus of the hypothalamus (PVN), and the amygdala. Using a 1-mm punch, tissue was collected bilaterally from each of the five brain regions, immediately frozen on dry ice, and stored at −80°C until use. Tissue from five saline- and five nicotine-treated pups was used for the microarray experiment. For reverse transcription–polymerase chain reaction (RT–PCR), we used 20 animals from 10 litters. To obtain sufficient RNA, total mRNA from each brain region of two animals per litter was pooled to yield five samples in each experimental group. All animal-related experiments were performed in accordance with the guidelines of the Institutional Animal Care and Use Committee at the University of California, Irvine, and in compliance with Federal guidelines.

Microarray production

A pathway-focused oligoarray designed specifically for drug addiction and brain-related research was used. Briefly, 3565 genes implicated in regulation of the central nervous system activities for the maintenance of neuronal homeostasis, as well as those associated with the neuron response to addictive substances, were selected on the basis of an earlier version of a pathway-focused cDNA microarray (Konu et al. 2004) and a literature survey. These genes cover most of the major pathways related to cell metabolism, genetic information processing, cellular signalling transduction, neuron-related disease, and cell communication. OligoWiz (http://www.cbs.dtu.dk/services/OligoWiz/) was used to design the oligonucleotide for each gene. The final oligonucleotide length was 59.2±3.8 (mean±s.d.) bp, with a GC content of 0.53±0.05 and a Tm 76.4±1.7°C. The oligonucleotides and 10 control clones were synthesized and spotted at a concentration of 40 µm in 3× SSC and 1.5 m betaine buffer onto CMT-GAPS II slides (Corning, USA) using an OmniGrid MicroArrayer OGR-03 (GeneMachines, USA).

RNA isolation, amplification and cDNA labelling

Total RNA was isolated from each brain region using Trizol (Invitrogen, USA). Two micrograms of total RNA of each sample was used for reverse transcription in a final volume of 20 µl containing 4 µl of 5× first-strand buffer, 1 µl of 10 mm dNTP, 1 µl of 50 µm random hexamer primer, 2 µl of 0.1 m DTT, 1 µl RNaseOUT (Invitrogen, USA), and 1 µl of Superscript II™ reverse transcriptase (Invitrogen). The mixture was incubated at 25°C for 10 min, 42°C for 2 h, and 70°C for 15 min. After reverse transcription, we added 30 µl of 5× second-strand buffer, 3 µl of 10 mm dNTP, 4 µl DNA polymerase, 0.5 µl RNase H, 1 µl E. coli DNA ligase, and 92.5 µl H2O to each reaction tube and incubated them at 16°C for 3 h. After purification by phenol:chloroform and precipitation by isopropanol, this double-stranded DNA was dissolved in 8 µl H2O and transcribed using AmpliScribe™ T7 Transcription kits (Epicentre, USA).

Four micrograms of amplified RNA of each sample was labelled with cyanine 3- or 5-dUTP by reverse transcription. Following purification with phenol:chloroform extraction and isopropanol precipitation, cDNA was dissolved in 28 µl H2O, mixed with 4 µl of 10× buffer, 4 µl of 10 mm dTTP-free dNTP, 1 µl of 10 mm dTTP, 2 µl of 1 mm cyanine 3-dUTP or cyanine 5-dUTP (Enzo, USA), and 1 µl of Klenow fragment (50 units/µl) and incubated at 37°C for 3 h prior to purification with a QIAquick PCR kit (Qiagen, USA).

Hybridization to pathways-focused microarray

We mixed cyanine 3-labelled control cDNA probes with the cyanine 5-labelled sample cDNA probe and added 7.5 µl of 20× SSC, 3 µg CotI DNA, 3 µg polyA, and 0.5 µl of 10% SDS in a final volume of 50 µl. The mixture was applied to the pathway-focused oligonucleotide microarray and hybridized overnight at 60°C. Slides were then washed in 1× SSC and 0.2% SDS at 60°C for 5 min followed by washing in 0.1× SSC and 0.2% SDS and in 0.1× SSC at room temperature for 10 min. Hybridized slides were scanned using the ScanArray Gx microarray scanner, and the intensities of each probe were quantified with the ScanArray Express microarray analysis system (PerkinElmer, USA).

Microarray data analysis and identification of overrepresented pathways

After scanning each array, we obtained the raw hybridization intensities of each element and used the background-subtracted median intensity of each spot for statistical analysis. The two replicates of each gene were analysed separately. To minimize spot variations and reduce experimental error, we discarded spots that were over-saturated or poorly expressed (i.e. 5% of the weakest spots in each replicate of an array). Then we used an intensity-dependent normalization method (locally weighted linear regress; Lowess) to normalize the data for each replicate (Yang et al. 2002). After removing spots with fewer than six valid measurements per experimental group, we averaged two replicates per chip to be used as the measurement of expression of a gene.

Following identification of the genes differentially expressed in the nicotine-treated and saline control groups, a bioinformatics tool called Gene Set Enrichment Analysis (GSEA) (Subramanian et al. 2005) was utilized to determine the pathways showing differences in each brain region. GSEA computationally identifies whether a priori-defined sets of genes (pathways) show statistically significant and concordant differences in two biological states. For each gene set (pathway), a Normalized Enrichment Score (NES) is calculated by considering all the gene sets tested, and a p value is calculated to determine whether this set is enriched among the input genes compared to a random distribution. The pathways included in the GSEA database were collected from multiple public domains (e.g. http://www.sigmaaldrich.com/; http://www.biocarta.com, and http://www.genome.jp/kegg/). The software and curated pathway database was implemented locally in our laboratory.

When the software and database were downloaded (June 2008), 441 pathways were included. For a detailed description, please refer to http://www.broadinstitute.org/gsea/.

Verification of microarray results using quantitative RT–PCR (qRT–PCR) array

Two micrograms of total RNA was reverse transcribed using Superscript II RT. The mixture was incubated at 25°C for 10 min, 42°C for 1.5 h, and 70°C for 15 min. The PCR was performed in a total volume of 20 µl containing 10 µl of 2× Power SYBR Green PCR Master Mix (Applied Biosystems, USA) and 1 µl (5 µm) of combined sense and anti-sense primers of the gene of interest (see below for primer designs and sequences). The RT–PCRs were performed in a 96-well plate using an ABI Prism 7000 Sequence Detection System (Applied Biosystems) with 1 cycle at 50°C for 2 min, initial denaturation at 95°C for 10 min, 40 cycles of denaturation at 95°C for 15 s, and annealing/extension at 60°C for 1 min. Subsequent to the last cycle, a dissociation curve was generated to check for non-specific products. Expression of all genes assayed in the qRT–PCR array was normalized to the expression of glyceraldehyde 3-phosphate dehydrogenase (GAPDH) and then analysed using a comparative Ct method (Winer et al. 1999). As described above, five pooled mRNA samples were to be used for both gestational nicotine-treated and saline control animals, but one sample of each group was discarded because of insufficient mRNA extraction or failure to pass quality control. Thus, the quantitative analyses of RNA expression were based on four pooled samples. All primers were designed using Primer Express Software 3.0 for Real-Time PCR (Applied Biosystems) and synthesized by Fisher Scientific (USA) (Supplementary Table S1, available online).


Pathways regulated by gestational nicotine exposure

A pathway-focused oligonucleotide microarray was used to profile gene expression in five brain regions of adolescent female rats receiving gestational nicotine. The processed expression data were analysed by GSEA (Subramanian et al. 2005), which assigned a valid NES value and p value to 69, 56, 76, 58, and 84 pathways for the amygdala, CPu, NAc, PFC, and PVN, respectively. Too few genes in the remaining pathways were represented on the chip. For each brain region, the pathways involved in different biochemical processes and various functions were significantly enriched (nominal p<0.05; Supplementary Table S2), suggesting nicotine has multiple effects on brain development. We categorized the pathways into cell death/survival, development/plasticity, immune response, and cellular metabolism (Supplementary Table S2).

Identification of cell death/survival-related pathways modulated by gestational nicotine exposure

For each brain region, multiple pathways related to cell death/survival showed statistically significant, concordant differences between the adolescent offspring exposed to nicotine and saline during gestation (Fig. 1). For the NAc, 5/7 pathways with the most significant p values were cell death/survival related; for the PFC, 2/3 pathways with the most significant p values, as well as 7/19 differentially expressed pathways, were involved in cell death/survival processes; and for the PVN, 12/33 pathways were cell death/survival related. For the other two brain regions, amygdala and CPu, some pathways involved in cell death/survival were identified. Altogether, 24 unique biological pathways related to cell death/survival were significantly modulated by nicotine (Table 1). Among these pathways, 11 were regulated in more than one brain region; e.g. the epidermal growth factor (EGF) pathway was significantly regulated in both the NAc and PVN, while the EGFR_SMRTEPATHWAY was differentially expressed in the amygdala and PVN. The regulation of cell death/survival-related pathways in each brain region, as well as the modulation of some of these pathways in multiple brain regions, suggest that biological processes related to cell death/survival are key to nicotine's effects on neuron survival in adolescent rats.

Fig. 1

Identification of differentially expressed biological pathways in the five brain regions of adolescent rats receiving nicotine exposure during gestation. Gene Set Enrichment Analyses (GSEA) was adopted to detect the statistically significantly modulated pathways (nominal p<0.05). For each brain region, the pathways were plotted in descending order of the negative logarithm of their p values at base 10. The biological pathways related to cell death or cell survival are shown in black columns (■), while the others are shown in light grey (Embedded Image). For each pathway, a short format of its name in the GSEA database is shown in the figure: for amygdala, Gamma_hexachlorocyclohexane and Mitochondrial_fatty_acid are short formats of Gamma_hexachlorocyclohexane_degradation and Mitochondrial_fatty_acid_betaoxidation, respectively; for NAc, Sig_pip3_signaling corresponds to Sig_pip3_signaling_in_cardiac_myoctes; for PFC, St_T_cell_signal and St_B_cell_antigen are short formats of St_T_cell_signal_transduction and St_B_cell_antigen_receptor, respectively; for PVN, Oxidative_phosph, Glycerolipid and Sa_B_cell_receptor are short formats of Oxidative_phosphorylation, Glycerolipid_metabolism and Sa_B_cell_receptor_complexes, respectively; for the other pathways, the word ‘pathway’ has been omitted from their names.

View this table:
Table 1

On the basis of their biological functions and characteristics, we classified the cell death/survival-related pathways into three subgroups, i.e. growth factor, death receptor, and kinase cascade (Table 1). Some pathways have common upstream genes and signalling transduction. For example, EGFPATHWAY demonstrates that EGF induces cell proliferation through activating mitogen-activated protein kinase (MAPK), phosphatidylinositol 3-kinase (PI3K), and calcium signalling, whereas EGFR_SMRTEPATHWAY shows MAPK signalling-mediated suppression of silencing mediator of retinoid and thyroid hormone receptor (SMRT; also known as nuclear receptor co-repressor 2) by EGF, indicating EGFR_SMRTPATHWAY is a branch of EGFPATHWAY. Thus, these 24 cell death/survival-related pathways could be grouped into subcategories according to their signalling (Table 1).

Twelve pathways included in the growth factors category belong to six subcategories, i.e. EGF, erythropoietin (EPO), insulin/insulin-like growth factor (IGF), integrin, platelet-derived growth factor (PDGF), and pituitary adenylate cyclase-activating peptide (PACAP). Three pathways, i.e. CERAMIDEPATHWAY, HIVNEFPATHWAY, and STRESSPATHWAY, are among the death receptor pathways. Nine are kinase cascade pathways, which belong to two subcategories, MAPK and PI3K. Three MAPK pathways, i.e. p38, extracellular signal-regulated kinases (ERK), and c-Jun N-terminal kinases (JNK), were identified (Table 1).

Replication of cell death/survival-related pathways in different brain regions using qRT–PCR

Microarray analysis showed that nicotine exposure regulated multiple cell death/survival-related pathways in each brain region. We selected 29 representative genes for examination with qRT–PCR. They were part of the growth factors and their receptors, death receptors and their ligands, caspase family, kinases, and transcriptional factor pathways (Supplementary Table S3). Table 2 show the changes in gene expression at p<0.05. The following is a summary of the qRT–PCR results in different regions.

View this table:
Table 2


Three genes were significantly changed in this region. PDGFA (1.18, p=0.04), which belongs to growth factors, and TNFSF6 (1.62, p=0.01), which belongs to death receptors, were up-regulated, whereas MEF2C, which belongs to transcription factors (0.80, p=0.03), was down-regulated.


Multiple growth factors, namely brain-derived neurotropic factor (BDNF) (2.11, p=0.01), IGF1R (1.29, p=0.01), INS (1.56, p=0.02), and ITGB5 (1.25, p=0.02), were increased significantly by nicotine. Consistent with the up-regulation of growth factors, downstream molecules of these growth factors; i.e. kinases (MAPK8: 1.43, p=0.03; PRKCE: 1.16, p<0.01) and transcription factors (MEF2C: 1.24, p=0.03; TP53: 1.27, p=0.02) were significantly up-regulated. However, no genes in the death receptor and caspase pathways were changed, suggesting nicotine has limited effects on the death receptor pathways in this region.


No growth or transcription factors were significantly changed, and only one gene in the kinase pathway (PRKCE: 2.05, p=0.03) was increased in the PFC. On the other hand, one death receptor, i.e. TNFSF6 (0.78, p=0.05) and two caspase family members, i.e. caspases 2 (0.83, p=0.04) and 3 (0.86, p<0.01), were decreased, indicating nicotine suppresses the caspase cascade through death receptor pathways in the PFC.


Both MAP2K6 (1.24, p=0.03) and FOS (1.54, p=0.03), which belong to the kinases and transcription factors, respectively, were up-regulated. But nicotine regulated the expression of growth factors in different directions. For example, BDNF (0.59, p<0.01) and PDGFA (0.87, p=0.01) were decreased, whereas ITGB5 (1.23, p<0.01) was increased. No death receptors or caspases were modulated significantly.


In the striatum, several growth factors, namely BDNF (1.82, p=0.03), IGF1R (1.37, p<0.01), ITGB5 (1.25, p<0.02), and VEGFA (1.15, 0.04), were increased significantly by nicotine. In the category of kinases and transcription factors, PRKCA (1.16, p=0.02) was up-regulated, and MEF2C (0.89, p<0.01) was down-regulated. These data indicate that nicotine predominantly induces growth factor pathways in the striatum, similar to the NAc. On the other hand, death receptors [tumour necrosis factor (TNF): 1.21, p=0.01; TNFRSF1A: 1.16, p<0.01] and caspases (CASP2: 1.19, p=0.04; CASP3: 1.20, p=0.04; CASP8: 1.56, p=0.01) were increased significantly by nicotine.

Proposed model for nicotine's effects on cell survival in different brain regions

On the basis of our microarray and qRT–PCR results, we present a model of nicotine's involvement in regulating cell survival in different brain regions (Fig. 2). Generally speaking, involvement of nicotine in cell death/survival processes is realized through interaction of growth factor and the death receptor pathway. Activation of growth factor-related pathways as well as downstream kinase cascade and other transcription factors promotes cell survival. On the other hand, death receptors inhibit survival by activating the caspase cascade. Nicotine modulates both growth factor and death receptor pathways by affecting the expression of their genes. The balance of these pathways may determine the final effect of nicotine on cell survival.

Fig. 2

Proposed model of nicotine's anti-apoptotic and pro-apoptotic effects in the brain regions investigated. Because both growth factor and death receptor pathways are regulated by nicotine, the final effect of nicotine on cell survival is the consequence of the interaction between the two pathways. Prenatal nicotine exposure exerts a region-specific effect on each brain region. (a) In the amygdala, nicotine activates the death receptor pathways by inducing the expression of the death receptors. The expression of the growth factors is increased by nicotine; however, expression of the downstream transcription factors is suppressed. Thus, the final effect of nicotine in this region depends on the eventual balance of the two cascades. (b) In the nucleus accumbens (NAc), nicotine activates the growth factor pathway by increasing expression of growth factors, kinases, and transcription factors but has little effect on the death receptor pathway, which probably leads to enhancement of cell survival in this region. (c) In the prefrontal cortex (PFC), nicotine may suppress the death receptor pathway by inhibiting the expression of death receptors and caspases. Although nicotine has limited effects on growth receptors, upregulation of kinases suggests that nicotine probably has protective effects in this region. (d) In the periventricular nucleus of the hypothalamus (PVN), nicotine does not regulate the death receptor pathway obviously. On the other hand, it induces expression of some growth factors and suppresses others, although the downstream kinase cascade and transcription factors are activated. (e) In the striatum, nicotine activates both the growth factor and the death receptor pathways, although the expression of the downstream transcription factors of the growth factors is slightly inhibited. Open arrows indicate directions of expression change in response to nicotine and the final effects of nicotine on cell survival.


In humans, maternal smoking leads to numerous physiological defects and neurobehavioural disorders in offspring that become manifest during adolescence (Dwyer et al. 2008; Franke et al. 2007; Shacka & Robinson, 1998), a period of brain reorganization marked by both cell growth and cell pruning in the cortical and limbic circuitry (Spear, 2000). Although gestational nicotine exposure changes the vulnerability of specific brain regions and causes neural damage that may last into adolescence and affect brain development (Abreu-Villaca et al. 2004), the molecular mechanisms underlying the effects of gestational nicotine exposure on adolescent brain development have not been explored fully. In this study, we showed that pathways related to cell survival and death may play important roles.

Nicotine exposure has both pro-apoptotic and anti-apoptotic effects, depending on cell type and nicotine concentration (Zeidler et al. 2007). Nicotine exerts its effects on cell death or survival by binding to nAChRs and modulating biochemical pathways related to growth factors [BDNF, fibroblast growth factor, neural growth factor (NGF)] (Zeidler et al. 2007), TNF family members (Kamer et al. 2006; Park et al. 2007; Wang et al. 2004), and kinases (ERK, PI3K/AKT, PKC) (Toborek et al. 2007; Zeidler et al. 2007). We here extended these findings, showing that pathways associated with multiple growth factors, death receptors, and kinases are regulated by nicotine during fetal development, with effects that persist far beyond the period of exposure.

Growth factor proteins stimulate cell growth, proliferation, or differentiation and activate multiple downstream kinase cascades (e.g. PI3K and MAPK signalling) by binding to surface receptors (Miloso et al. 2008). In our study, pathways related to multiple growth factors (e.g. BDNF, insulin, PDGF and integrin) were found to be modulated. All these pathways were reported to promote neuron survival (Nguyen et al. 2009; Peng et al. 2008; Zeidler et al. 2007). For example, BDNF is necessary for survival and phenotypic maintenance of mature neurons (Zuccato & Cattaneo, 2009), whereas PDGF promotes survival and proliferation of immature neurons (Peng et al. 2008). Regulation of these pathways by gestational nicotine suggests that nicotine acts on the brain at a different development stage, resulting in long-term effects on central neurons. Furthermore, BDNF and PDGF are involved in several neuronal disorders, e.g. Alzheimer's disease, Huntington's disease, and human immunodeficiency virus type 1-associated dementia (Zuccato & Cattaneo, 2009), suggesting protective effects of nicotine against neurodegenerative disorders. We also observed regulation of death receptors and related pathways in different brain regions. In contrast to growth factors, death receptors transmit apoptotic signals by activating caspase cascades initiated by their ligands (Adam-Klages et al. 2005). Moreover, suppressing growth factor pathways by death receptor pathways is an important way to induce neuron death (Venters et al. 1999, 2000; Ye et al. 2003). Modulation of death receptor pathways is widely reported in neuron development, differentiation, and some disorders (Bessis et al. 2005; Lorz & Mehmet, 2009). Gestational nicotine exposure regulates both growth factor and death receptor pathways in the developing brain, suggesting the interaction of these two opposite pathways affects differentiation and proliferation of neurons. Kinase cascades are widely activated and can be involved in both growth factor and death receptor pathways and can also be activated directly by nicotine via nAChR-mediated calcium signalling (Carlisle et al. 2007; Tang et al. 1998). Multiple kinase cascade pathways were identified in our study, which belong to two subcategories, MAPK and PI3K. Three groups of MAPKs, p38, JNK, and ERK, are modulated by gestational nicotine in all the brain regions we examined, suggesting broad effects of nicotine on MAPK pathways. In addition, kinase cascades not only mediate downstream effects of cell survival pathways (Miloso et al. 2008), but also regulate the expression of growth factors and death receptors themselves (Kanda & Watanabe, 2007). Thus, regulation of kinase pathways by nicotine plays important roles in both upstream and downstream control of growth factor and death receptor pathways.

At the same time, the modulation patterns of growth factors, death receptors, and kinases differ among brain regions, suggesting that nicotine exerts different pharmacological effects on the three functional groups in different parts of the brain, probably by virtue of region-specific expression patterns and compositions of nAChRs. In the adolescent NAc, expression of growth factors, kinases, and transcription factors were all up-regulated by gestational nicotine treatment, suggesting that prenatal nicotine exposure enhances cell survival pathways in this region (Fig. 2). This increase in activity may indicate interference with normal synaptic pruning, which could alter the function of the corticostriatal circuitry in which the NAc participates. Indeed, in the same animal model, gestational nicotine alters locomotor and reward behaviours regulated by the NAc in adolescence (Franke et al. 2008). The striatum, a structurally related brain region that similarly undergoes synaptic pruning in adolescence, shows up-regulation of expression of the growth factors, as the expression of death receptors and caspases is also up-regulated (Fig. 2), suggesting a more complex effect of gestational nicotine exposure. The differences in gene expression in the cell survival/death pathways in the striatum and NAc may highlight the regionally selective nature of gestational nicotine impact on the activity of these pathways. The PFC, another actively maturing adolescent brain structure, also appears to have greater activity of cell survival pathways as a result of gestational nicotine treatment; however, the mechanism is distinct from that of the NAc. The PFC shows no changes in the expression of growth factors but rather exhibits decreases in the expression of death receptors and caspases (Fig. 2). An increase in the activity of cell survival pathways could have implications for both synaptic pruning efficiency and overproliferation of growing limbic afferents. Although gestational nicotine's effects on the activity of cell survival/death pathways are more complex in the amygdala and PVN, clearly, long-term, region-dependent changes in brain development last into the adolescent period.

Cell death is an important feature of early brain development, and neuronal morphology and survival can be altered by nicotine exposure (Dwyer et al. 2009). Cell survival and programmed death are also integral to adolescent brain development. During adolescence, communications between the cortex and limbic system increase as amygdalocortical (Cunningham et al. 2002) and corticoaccumbens connections proliferate (Brenhouse et al. 2008). However, at the same time synaptic pruning is occurring in the PFC (Andersen et al. 2000), amygdala (Zehr et al. 2006), CPu, and NAc (Tarazi et al. 1999), which could result from more programmed cell death or retraction of synaptic connections.

The present work shows that fetal nicotine exposure induces long-lasting changes in cell death and survival cascades in regions developing during adolescence. This is consistent with previous findings that many neurochemical effects of gestational nicotine exposure continue or re-emerge during adolescence (Slotkin, 2008). The findings suggest that adolescents are vulnerable to gestational nicotine exposure, which may be related to their active brain development. However, it is not clear whether the alterations in the cell death/survival-related pathways are continuous across postnatal development or occur specifically during adolescence.

Adolescents are vulnerable to the initiation of drug abuse, and many psychiatric disorders associated with maternal smoking begin or change symptomatology during adolescence (Spear, 2000). The brain regions examined in this study are important for motor and emotional control and motivated behaviours (Arnsten, 1997; Herman et al. 1996; Herrero et al. 2002; Ikemoto, 2007; Squire et al. 1993). Therefore, the modulation of cell death/survival-related pathways that are important to adolescent brain development implies alterations in these pathways in the deficits seen at this age in the offspring of maternal smokers.

In addition to their effects on cell survival, growth factors may play roles in signal transduction in the brain. One example is BDNF, recently linked to dopamine receptor signalling (Williams & Undieh, 2009), which may be important in responses to drugs of abuse (Bahi et al. 2008). Gestational nicotine alters BDNF expression in three of the five brain regions studied (Table 2), with significant up-regulation in the dorsal and ventral striatum. Others have shown that increased BDNF in the NAc both enhances locomotor sensitization to cocaine and alters cocaine reward (Bahi et al. 2008). These results are consistent with behavioural studies in gestational nicotine-treated adolescent animals, which show behavioural sensitization to cocaine whereas those treated gestationally with saline do not exhibit this effect (Franke et al. 2007). Further testing would be required to determine whether increases in BDNF in the NAc mediate this enhancement of the behavioural response to cocaine. These studies have strong parallels to human research, in which maternal smoking increases the risk of substance abuse during adolescence and adulthood (Weissman et al. 1999). BDNF has also been suggested to be involved in the corticostriatal circuitry in humans, with higher plasma concentrations linked to attention deficit hyperactivity disorder (ADHD) (Shim et al. 2008). ADHD is one of several neurobehavioural disorders whose incidence is increased by maternal smoking (Weissman et al. 1999), and thus nicotine-induced increases in adolescent BDNF observed in this work may be a novel mechanism underlying these clinical findings.

However, because of the potential limitation of the experimental design and protocols used, some of the results should be viewed with caution. First, we focused on adolescent female animals. Previous studies have demonstrated the existence of sexual dimorphism in both humans and animals in response to prenatal exposure to tobacco components. For example, a female offspring whose mother smoked during pregnancy is more vulnerable to drug abuse than a male offspring (Kandel et al. 1994; Oncken et al. 2004; Weissman et al. 1999). Moreover, female offspring show greater reduction in visual attention performance (Jacobsen et al. 2007). Similar to humans, female animals exposed to nicotine prenatally differ greatly from males in both behaviours and brain development (Klein et al. 2003; Lichtensteiger & Schlumpf, 1985; Xu et al. 2001). Because of this sexual dimorphism, as well as the number of regions being examined, for simplicity, only female offspring were used in the present study. The results and conclusions therefore may not be extrapolated directly to male offspring. Second, the potential influences of nicotine withdrawal on brain development cannot be excluded. In our study, the pregnant animals received a moderate dose of nicotine until GD 18–20, and the newborn pups were cross-fostered to drug-naive surrogate dams to avoid the effect of withdrawal on maternal behaviours. Although this model resembles the clinical condition of pregnancy smoking in humans and has been a standard in the literature, and provided highly convergent biochemical and behavioural data to the effects of maternal smoking on adolescent human offspring (Dwyer et al. 2008, 2009), we cannot fully separate the effects of gestational nicotine from those of gestational nicotine combined with withdrawal. Third, our results demonstrated significant expression changes of genes encoding growth factors, death receptors, caspases, apoptosis-related kinases, and transcriptional factors in response to prenatal nicotine exposure. Consistently, previous studies have shown that nicotine exposure alters growth factors such as BDNF and IGF at both the mRNA and protein levels (Andresen et al. 2009; Cheeta et al. 2000; Fox et al. 2009; Gruslin et al. 2009; Yeom et al. 2005). Moreover, nicotine exposure alters apoptotic markers, including caspase-3, in the piglet brain (Machaalani et al. 2005). Although the consistent changes of gene expression in the pathways examined may be sufficient to indicate the regulation of the related biochemical processes such as apoptosis, given that only mRNA was evaluated in this study, it remains to be seen whether the same or similar expression patterns can be observed at the protein level.

In summary, by analysing gene expression patterns, we have gained systematic insight into the pathways modulated by gestational nicotine exposure in the brains of female adolescent offspring. Identification of multiple growth factor, death receptor, and kinase cascade pathways in different brain regions indicates nicotine has different effects on cell death/survival in critical brain circuitry that is maturing during adolescence. These findings suggest that gestational exposure to nicotine alone, such as that found in nicotine replacement therapy, induces long-lasting changes in cell survival and death pathways that may fundamentally alter proper brain development. These data provide clues to novel mechanisms of the nicotine-induced physiological and behavioural disorders observed during adolescence in both rodent models and the clinical literature.


This project was in part supported by NIH grant DA-13783 to Ming D. Li and DA-10612 to Frances Leslie. The authors thank Dr David L. Bronson for his excellent editing of this manuscript. The authors also thank Celina Mojica and Susan McQuown for their technical assistance.


Supplementary material accompanies this paper on the Journal's website.

Statement of Interest



View Abstract