Alongside the well-established endocrinological and somatic changes associated with puberty (Sizonenko,1989; Rogol et al.,2002), a burgeoning line of research has revealed that the pubertal brain also undergoes significant transformations, which may mediate some of the cognitive and emotional changes observed during this stage of development (reviewed in; Juraska and Markham,2004; Crews et al.,2007; Giedd et al.,2009; Casey et al.,2010; Giedd and Rapoport,2010; Spear,2010). For instance, studies in both human and nonhuman animals have shown significant pubertal-related cortical thinning and neuron pruning in select cortical regions (Jernigan et al.,1991; Pfefferbaum et al.,1994; Giedd et al.,1999; Courchesce et al.,2000; Nunez et al.,2002; Giedd,2004; Gogtay et al.,2004; Markham et al.,2007; Sowell et al.,2007; Knickmeyer et al.,2010), whereas the volume of limbic areas, such as the amygdala and hippocampus, and myelinated fiber tracts have been reported to increase (Giedd et al.,1996; Giedd et al.,1997; Giedd et al.,1999; De Bellis et al.,2001; Romeo and Sisk,2001; Isgor et al.,2004; Juraska and Markham,2004; Suzuki et al.,2005; Cooke,2010; Knickmeyer et al.,2010; Lebel and Beaulieu,2011). Although it is still unclear what factor(s) mediate these morphological changes during puberty, a few studies have provided evidence that gonadal hormones may be involved (Nunez et al.,2002; Cooke et al.,2007; Perrin et al.,2008; Raznahan et al.,2010).
In addition to these structural alterations, changes at a more cellular level have been reported in the peripubertal brain as well. One such notable change is the pubertal-related decrease in both cellular proliferation and neurogenesis in the dentate gyrus of the hippocampal formation. In particular, prepubertal males have levels of cellular proliferation and neurogenesis that are twice as high as those found in adults (Heine et al.,2004; Kim et al.,2004; McDonald and Wojtowicz,2005; Crews et al.,2007; He and Crews,2007; Cowen et al.,2008; Hodes et al.,2009). Although hippocampal neurogenesis in adulthood has been purported to play a role in learning and memory (Shors et al.,2001; Shors et al., in press) and mood disorders (Samuels and Hen,2011), the neurobehavioral implications of the change in neurogenesis during puberty remain unknown.
Similar to some of the gross morphological alterations noted above, it is possible that the change in gonadal hormones associated with puberty may mediate the pubertal-related decrease in cell proliferation and neurogenesis in the dentate gyrus. Gonadal hormones in adulthood have been shown to modulate hippocampal neurogenesis (Tanapat et al.,1999; Tanapat et al.,2005; Galea et al.,2006; Spritzer and Galea,2007; Barker and Galea,2008; Galea,2008), suggesting that hormones may also have an impact on this process during puberty. Moreover, exposure to gonadal hormones during puberty does appear to alter cellular proliferation, ultimately contributing to the sex difference in the volume of specific brain nuclei, such as the sexually dimorphic nucleus of the hypothalamus and medial amygdala (Ahmed et al.,2008). However, the relationship between pubertal changes in gonadal hormones and hippocampal neurogenesis has not been fully explored.
In an effort to gain a deeper understanding of pubertal brain development and changes in neurogenesis in the dentate gyrus, two experiments were conducted. In the first experiment using 5-bromo-2′-deoxyuridine (BrdU) and doublecortin (DCX) immunohistochemistry in male rats, we examined the temporal nature of the pubertal decline in cellular proliferation and neurogenesis in the dentate gyrus and measured the testosterone levels of prepubertal, midpubertal, and adult rats. In the second experiment, we tested the hypothesis that the pubertal rise in gonadal hormones was responsible for the observed pubertal decrease in hippocampal cell proliferation and neurogenesis. We report here that even though pubertal decreases in BrdU and DCX cell numbers are coincident with an increase in testosterone levels, this decline in hippocampal neurogenesis is independent of the pubertal rise in gonadal hormones.
Male Sprague-Dawley rats were obtained from our breeding colony at Barnard College. All animals were weaned at 21 days of age, housed two per cage in clear polycarbonate cages (45 × 25 × 20 cm3) with wood chip bedding, and maintained on a 12 h light–dark schedule (lights on at 0900 h). Animals had ad libitum access to food and water, and the animal room was maintained at 21 ± 2°C. All procedures were carried out in accordance with the guidelines established by the National Institutes of Health Guide for the Care and Use of Laboratory Animals and approved by the Institutional Animal Care and Use Committee (IACUC) of Columbia University.
Two experiments were conducted. Experiment 1 examined the effects of pubertal development on cell proliferation and neurogenesis in the dentate gyrus of the hippocampal formation. Experiment 2 investigated the role of the pubertal rise of gonadal hormones on pubertal-related changes in these parameters.
In Experiment 1, male rats at 28 (prepubertal), 43 (midpubertal), or 88 (adult) days of age received two intraperitoneal (i.p.) injections of BrdU (200 mg/kg in a 0.9% saline vehicle; Sigma, St. Louis, MO) separated by 24 h and were killed 24 h after the second injection [i.e., at 30, 45, or 90 days of age; n = 6 per age; Fig. 1(A)]. For tissue collection, animals were weighed, given an overdose of sodium pentobarbital (150 mg/kg), and following a cardiac puncture, were perfused with 0.9% heparinized saline followed by 4% paraformaldehyde in 0.1 M phosphate buffer (PB). Brains were postfixed in 4% paraformaldehyde for 24 h then stored in 20% sucrose at 4°C. Thirty-micron coronal sections were made on a cryostat and stored in cryoprotectant at −20°C until processed for BrdU and DCX immunohistochemistry (see below) and Nissl staining.
In Experiment 2, four groups of rats were assessed: (i) intact prepubertal (28 days of age) males, (ii) intact late-pubertal (58 days of age) males, (iii) late-pubertal (58 days of age) males that had been sham-gonadectomized (SHAM) at 30 days of age, and (iv) late-pubertal (58 days of age) males that had been gonadectomized (GDX) at 30 days of age. During surgery, SHAM and GDX animals were anesthetized with ketamine (100 mg/kg; i.p.) and xylazine (10 mg/kg, i.p.). Similar to Experiment 1, these intact, SHAM, and GDX animals at 28 or 58 days of age received two i.p. injections of BrdU (200 mg/kg in a 0.9% saline vehicle) separated by 24 h and killed 24 h after the second injection [i.e., at 30, or 60 days of age; n = 6 per group; Fig. 1(B)]. Perfusions, tissue collections, and sectioning were conducted as described above.
BrdU Immunohistochemistry and Cell Counting
In both experiments, every 8th section throughout the dorsal hippocampal formation (plate 55–70; Paxinos and Watson,2005) was processed for BrdU immunohistochemistry. To reduce endogenous peroxidase activity before staining, tissue sections were treated with 0.6% H2O2 in 0.1 M PB for 20 min. Sections were then rinsed three times in 0.1 M PB for 5 min each and incubated for 2 h at 65°C in 50% formamide and 1× sodium chloride and sodium citrate buffer (SSC, Ambion, Foster City, CA). Sections were then washed two times in 2× SSC preheated to 65°C for 2 min each and two times in 0.1 M borate buffer (pH 8.5) for 5 min each. Following these washes, sections were incubated in blocking buffer (2% normal goat serum and 0.1% Triton X-100 in 0.1 M phosphate buffered saline; PBS) for 1 h at room temperature (RT) and then at 4°C in blocking buffer containing the primary antibody (BrdU;1:500; mouse; Roche Diagnostic GmbH, Mannheim, Germany). Forty-eight hours later, sections were incubated in biotinylated goat anti-mouse IgG (1:500; Vector, Burlingame, CA) and then avidin-biotin horseradish peroxidase complex (1:250; Vectastain ABC Kit, Vector) in 0.1 M PBS for 1 h each at RT, with three 5 min washes in PBS between each of the incubations. Horseradish peroxidase was visualized with 3,3′diaminobenzidine (DAB) in a 3 M sodium acetate buffer containing 0.05% H2O2. Sections were washed, mounted on to Fisher Brand Plus slides (Fisher Scientific, Pittsburgh, PA), dried, counter-stained with cresyl violet, dehydrated in increasing concentrations of alcohol (70, 95, and 100%), cleared in xylenes, and coverslipped with DPX Mountant (Sigma).
BrdU-positive cells were counted using a Zeiss Axiovert 200 microscope. Each dentate gyrus was centered under a 10× objective and then the magnification was increased to 40×. Similar to a previous study (Leuner et al.,2009) our data are presented as the estimated number of BrdU-positive cells in the dentate gyrus. Briefly, the total number of BrdU-positive cells was counted in the granule cell layer and subgranular zone (excluding the hilus) of seven bilateral dentate gyri. The sum of each of these bilateral counts was multiplied by eight (the number of intervening sections) and then added to provide the estimate. Additionally, because we measured the cross-sectional area of each dentate gyrus (see below), we also report our data as mean number of BrdU cells per unit area (100 mm2).
Cross-Sectional Area Measurements
To assess possible effects of age and gonadal hormone status on the cross-sectional area of the dentate gyrus, adjacent sections to those processed for BrdU were mounted on Fisher Brand Plus slides, dried, stained with 1.5% cresyl violet, cleared in xylenes, and coverslipped with DPX Mountant. Cross-sectional area was measured by capturing an image of the dentate gyrus under a 2.5× objective with an Olympus DP71 camera and then subjected to analysis using ImageJ (Wayne Rasband, National Institutes of Health, Bethesda, MD). For each animal, the complete cross-sectional area of each dentate gyrus was measured from each section and then averaged. Therefore, the data are presented as average cross-sectional area (μm2) per section.
DCX Immunohistochemistry and Cell Counting
For both experiments, six anatomically matched sections from each animal's dorsal hippocampal formation (plate 58–63; Paxinos and Watson,2005) were processed for DCX immunohistochemistry. Specifically, free-floating sections were washed in 0.1 M PB and incubated for 10 min in 0.05% H2O2 in 0.1 M PBS. Sections were then washed in 0.1 M PB with 0.1% Triton X-100 (PBT), blocked in 2% normal goat serum (NGS) in PBT for 1 h, and incubated in anti-DCX (1:10,000; guinea pig; Millipore, Billerica, MA) in 2% NGS in PBT for 24 h at 4°C. Sections were then incubated in a goat anti-guinea pig secondary antibody (1:200; Vector) in PBT for 1 h at RT and then in avidin-biotin horseradish peroxidase complex (1:250) in PBT for 1 h at RT. Horseradish peroxidase was visualized with DAB in a 3 M sodium acetate buffer containing 0.05% H2O2. Sections were washed, mounted on to Fisher Brand Plus slides, dried, dehydrated in increasing concentrations of alcohol (70, 95, and 100%), cleared in xylenes, and coverslipped with DPX Mountant.
The areal densities (cells per unit area) of DCX-positive cell numbers were quantified in both the upper and lower blades of the dentate gyrus. Cells were counted on a light microscope under a 40× objective (Nikon, Eclipse E400) using an ocular grid measuring 6250 μm2 [please see Fig. 5(F) for approximate grid size and placement]. Four bilateral counts were made for each brain section and averaged. The data are presented as mean number of DCX-positive cells per mm2.
A testosterone radioimmunoassay was conducted using a Coat-A-Count 125I RIA kit (Siemens Medical Solutions Diagnostics; Malvern, PA) and performed as indicated by the supplier. All samples were run in duplicate, and the lower limit of detectability was 0.1 ng/mL and intraassay coefficient of variation was 5.8%.
All data are presented as the mean ± S.E.M. One-way ANOVAs were used for statistical analyses, and significant effects were further analyzed with Tukey's Honestly Significant Difference tests. Differences were considered significant when p < 0.05. All statistical analyses were performed using GraphPad Prism software (version 5.04).
As expected, there were significant effects of age on body weight and plasma testosterone levels (F (2,16) = 764.1 and 16.54, respectively, p < 0.05), such that body weights (Table 1) and testosterone concentrations (Table 2) were increased significantly at each of the ages tested. These data confirm that male rats at 30, 45, and 90 days of age are indeed at disparate stages before, during, and after pubertal development.
Table 1. Mean (±SEM) Body Weight (g) of Rats in Experiments 1 and 2. Values That Share the Same Letter Are Not Significantly Different from Each Other
The estimated number of BrdU-positive cells in the dentate gyrus was significantly affected by age (F (2,15) = 17.68, p < 0.05) such that 30-day-old animals had a greater number of cells than animals at either 45 or 90 days of age [Fig. 2(A)]. Although there was a decrease in cell numbers between the 45- and 90-day-old animals, this difference was not statistically significant. Average cross-sectional area of the dentate was also significantly affected by age (F (2,15) = 7.57, p < 0.05). Specifically, 30- and 45-day-old animals had significantly smaller cross-sectional areas compared to 90-day-old animals [Fig. 2(B)]. As these morphological data indicate the area of the dentate is changing during the ages we examined, we further investigated whether the age-related differences in the estimated number of BrdU-positive cells could be accounted for by the changes in cross-sectional area. However, even when expressing the number of BrdU-positive cells per unit area (i.e.,100 mm2), we found that 30-day-old animals still had a significantly greater number of BrdU cells compared to either the 45- or 90-day-old animals [F (2,15) = 29.76, p < 0.05; Fig. 2(C)].
Using DCX immunohistochemistry, we were able to show these age-dependent changes in cellular proliferation were paralleled by changes in neurogenesis, such that 30-day-old animals had a significantly greater number of DCX-positive cells per mm2 compared to the 45- and 90-day-old animals [F (2,16) = 12.45, p < 0.05; Fig. (3)].
One-way ANOVA revealed significant differences between the body weights of these intact and surgically manipulated animals (F (3,22) = 265.3, p < 0.05). Specifically, similar to the data in Experiment 1, we found that 30-day-old males weighed significantly less than the 60-day-old males, independent of their gonadal status (Table 1). However, we also found that 60-day-old males that had been GDX at 30 days of age weighed significantly less than the 60-day-old males that were intact or sham castrated (Table 1). There was also a difference in plasma testosterone (F (3,22) = 16.81, p < 0.05) in that 30-day-old intact and 60-day-old GDX males had significantly lower testosterone levels than the 60-day-old intact and SHAM animals (Table 2).
Although we again found a significant pubertal-related decline in estimated BrdU-positive cell numbers (F (3,20) = 6.80, p < 0.05), such that gonadally intact 30-day-old animals had a significantly greater number of cells compared to gonadally intact 60-day-old males, we did not find any significant differences between intact, SHAM, or GDX 60-day-old males [Figs. 4(A) and 5(A–C)]. Furthermore, because there were no significant differences in cross-sectional area per section of the dentate gyrus in any of the groups [p = 0.12, Fig. 4(B)], we observed the same mode of results as those from the cells per section analysis when the data are expressed as BrdU-positive cells per 100 mm2 [F (3,20) = 8.25, p < 0.05, Fig. 4(C)].
Similar to Experiment 1, our DCX data parallel the BrdU data such that the 30-day-old animals had a significantly greater number of DCX-positive cells per mm2 compared to all of the groups at 60 days of age, regardless of their gonadal status [F (3,19) = 14.98, p < 0.05; Figs. 6 and 5(D–F)].
These results indicate that the significant decline in cellular proliferation and neurogenesis in the dentate gyrus during puberty are independent of the changes in gonadal hormones observed during this stage of development. Thus, although some modifications of the pubertal brain are clearly dependent on the exposure to gonadal hormones, such as the increase in medial amygdala volume (Cooke et al.,2007; Ahmed et al.,2008), cell pruning in the visual cortex (Nunez et al.,2002), and decreases in androgen receptor expression in the hypothalamus (Romeo et al.,2000), our data indicate that certain neurodevelopmental processes can occur during puberty in the absence of gonadal hormones. Similar to our current findings, Andersen et al. reported that the pubertal-related decrease in dopamine receptor density in the striatum of male rats is also independent of the pubertal rise in gonadal hormones (Andersen et al.,2002). Therefore, it appears that factors in addition to gonadal hormones shape the pubertal brain, but the specific intracellular and/or extracellular signals that mediate the substantial change in neurogenesis during puberty remain unknown.
Given that gonadal hormones can modulate hippocampal neurogenesis in adulthood (Galea et al.,2006; Galea,2008) that androgen receptors are present in the dentate gyrus (Tabori et al.,2005; Hajszan et al.,2007; Feng et al.,2010), and testosterone levels and neurogenesis during puberty are temporally related, it was surprising that we found no effect of gonadectomy on either cell proliferation or neurogenesis. However, an earlier experiment did indicate that although testosterone could increase the survival of newly born neurons in the dentate gyrus of adult male rats, testosterone has little effect on cellular proliferation (Spritzer and Galea,2007). Similarly, the stress-induced suppression of hippocampal cellular proliferation in male rats following social defeat could not be reversed by supplementing the socially defeated adults with exogenous testosterone (Buwalda et al.,2010). Thus, along with our current data, it appears that testosterone does not influence hippocampal cellular proliferation in males during adulthood (Spritzer and Galea,2007; Buwalda et al.,2010) or during pubertal development (Experiment 2). Our present data also extend the previous observation that the presence or absence of the pubertal rise in gonadal hormones does not significantly affect the survival of newly born cells in the dentate gyrus of either male or female rats (Ahmed et al.,2008).
Because DCX is a marker of immature neurons (Rao and Shetty,2004; von Bohlen und Halbach,2007) and our BrdU and DCX data parallel one another, these results suggest that the changes in cellular proliferation during puberty are in part reflective of changes in the birth of new neurons. Therefore, it will be important to further investigate the impact of both pubertal development and gonadal hormones on postproliferative steps of neurogenesis, such as migration and synaptic integration. It would also be interesting to assess the proportional contribution of glial and/or vascular cells to these pubertal-related changes in respect to the number of proliferating cells and whether these types of cells are differentially affected by the presence or absence of gonadal hormones during this developmental transition.
In addition to the pubertal decrease in BrdU- and DCX-positive cell numbers, we also observed a slight, but significant, increase in the cross-sectional area of the dentate gyrus between 45- and 90-day-old intact males (Experiment 1). These results are agreement with a previous study that reported that various subfields of the hippocampal formation, including the dentate gyrus, show continued growth between 56 and 77 days of age (Isgor et al.,2004). Although it is paradoxical that the size of a brain region increases while the cells within that same region exhibit a decrease in their proliferative potential, these data indicate that the pubertal development of the dentate gyrus goes beyond changes in just neurogenesis. Future studies will need to address the mechanisms that mediate this structural change in the dentate gyrus and whether these mechanisms require the presence of gonadal hormones.
It is interesting to note that a recent study showed that chronic stress, and subsequent exposure to stress-related hormones, leads to suppressed neurogenesis in adults, but increases in adolescent animals (Toth et al.,2008). These data suggest that particular neurochemicals may have disparate effects on the brain before and after pubertal maturation. Along these lines, brain-derived neurotrophic factor (BDNF) has been demonstrated to promote neurogenesis in the hippocampal formation of rats both in vitro and in vivo (Scharfman et al.,2005; Li et al.,2009). However, a previous study in rats noted that hippocampal levels of BDNF protein increase during puberty (Kozisek et al.,2008). Hence, the inverse relationship between neurogenesis and BDNF during this stage of development suggests that factors that regulate neurogenesis in adulthood may not operate in the same manner as they do during puberty.
In conclusion, we did not find support for our hypothesis that the pubertal rise in gonadal hormones was responsible for the observed decrease in hippocampal cell proliferation and neurogenesis during puberty. It appears, therefore, that factors other than gonadal hormones shape the pubertal change in hippocampal neurogenesis, but what these factors are remain unknown. However, given the role of neurogenesis in such processes as learning and memory, emotionality and responsiveness to antidepressants in adulthood (Shors et al.,2001; Santarelli et al.,2003; Samuels and Hen,2011), it will be imperative to continue to study factors that alter neurogenesis during puberty and the ramifications of these changes on a developing organism's neurobehavioral function.
The authors thank Page Buchanan for excellent animal care.