Introduction

The production of neurons during adulthood has been demonstrated in the mammalian hippocampus and olfactory bulb (OB) (Gage et al. 1998; Alvarez-Buylla and Garcia-Verdugo 2002; Carleton et al. 2003). The OB, the first relay on the sensory pathway, is the target for neuronal progenitors generated in the subventricular zone (SVZ). Cells born in the SVZ proceed toward the OB along a concise pathway, up to 5 mm long in rodents, called the rostral migratory stream (RMS) (Luskin 1993; Lois and Alvarez-Buylla 1994). Within the OB the newborn cells mature into two main types of interneurons, periglomerular cells and granule cells (Betarbet et al. 1996; Doetsch and Scharff 2001) and establish functional circuits with the existing neurons (Carlén et al. 2002). It is well known that precursor cells of the SVZ maintain the ability to divide during their migration within the RMS (Menezes et al. 1995; Feng and Walsh 2001). The cellular composition of the RMS is not homogenous. In rodents it is composed of cells with neuronal and astrocytic identities. Neuronal cells in the RMS include neuronal progenitors, neuroblasts, and biochemically differentiated neurons can be also distinguished (Jankovski and Sotelo 1996; Doetsch et al. 1997).

In recent years many molecules have been found to be expressed in the RMS and to directly affect processes of neurogenesis. Signaling molecules that control proliferation, migration, or differentiation of neural precursors in the adult brain, are presently intensively studied. We have previously identified one of the phenotypically determined cell types in the RMS, cells producing a versatile diffusible signaling molecule of nitric oxide (NO). Our morphological examination has revealed that certain number of the RMS cells contains the neuronal isoform of the NO synthase (nNOS) and produces this messenger (Račeková et al. 2005). We have also shown that the expression of nNOS is evident from the postnatal day 10 (Račeková et al. 2003) and persists throughout the entire life of the animal (Račeková et al. 2005). The recent study of Gutièrrez-Mecinas et al. (2007) has provided further anatomical evidence for the role of this messenger in adult neurogenesis, indicating that migrating neuroblasts of the RMS contain the transduction machinery required for supporting a regulatory action of NO.

It is well established that survival of newborn cells in the OB depends on sensory input (Biebl et al. 2000; Petreanu and Alvarez-Buylla 2002). For example, the dependence of the survival of newly generated neurons on sensory stimulation in the OB has been demonstrated by decreased neurogenesis due to increased cell death after naris closure (Corotto et al. 1994). In another study, using anosmic mice, it was found that sensory input is critical for the survival of young granule cells during maturation, and once synaptically connected, their survival depends on the level of activity that they receive (Petreanu and Alvarez-Buylla 2002). However, there are only few studies examining whether a similar environmental manipulation may influence the cells in the RMS, which are destined to become OB interneurons.

Maternal separation is a commonly used and well-described model of environmental influences on development and subsequent nervous system function and behavior. Several studies have clearly demonstrated that repeated maternal separation during the first 2–3 weeks of life have long-term consequences on endocrine, behavioral, and brain development later in life. For example, maternal separation in rats caused stress hyperactivity (Francis and Meaney 1999) and long-term effects induced by maternal separation appear to depend upon changes in the structure and function of medial prefrontal cortex neurons, which are involved in the regulation of the stress response (Spencer et al. 2005). In the first weeks of life, mother is the only source of information necessary for pups’ survival and this information is mediated mainly by olfaction. Pups, separated from their mother are thus deprived from the strong source of olfactory stimuli.

Although the evidence for continued cell production in the SVZ/RMS is now strong, the range of exogenous factors that influence this process remains largely unknown. The objective of our experiment was to investigate the possibility that odor deprivation induced by maternal separation has an effect on postnatal neurogenesis in the RMS.

Materials and Methods

Animals and Maternal Separation

Wistar albino rats were employed in this study. Experiments were approved by the Institutional Ethical Committee, in accordance with current national legislation.

Beginning from the first postnatal day (P1), pups were separated from the dam for a period of 3 h daily. Dams were first removed and placed in an adjacent cage. Litters were then transferred to a plastic container and placed into an incubator with the temperature set at approximately 34°C, the temperature consistent with nest measurements. After the isolation period, the pups were returned to their home cage, where they were reunited with the dam. The animals were divided into three groups based on the duration of maternal separation and survival periods: The 1st group—P14 rats (2 weeks of maternal separation; n = 12); The 2nd group—P21 rats (3 weeks of maternal separation; n = 12); The 3rd group—P28 rats (3 weeks of maternal separation followed by 1-week survival; n = 12). Control rats (n = 36) were reared under the same conditions except maternal separation, and sacrificed at postnatal days 14, 21, and 28 (n = 12 each survival period).

Tissue Processing

Both control and maternally separated rats were deeply anesthetized with xylazine/ketamine and intracardially perfused with a solution of 4% paraformaldehyde in 0.1 M phosphate buffer (PB). The brain was dissected out, post-fixed in the same fixative and cryoprotected with 30% sucrose in PB overnight. Sagittal serial sections (40 μm) were cut on a cryostat and processed either for immunohistochemical labeling or histochemical staining (n = 6 for each method of processing).

Bromodeoxyuridine Administration and Detection

To label proliferating cells, six animals from each experimental group received a single i.p. injection of 50 mg/kg of bromodeoxyuridine (BrdU; Sigma; dissolved in 0.9% NaCl with 0.007 M NaOH), under light Halothane anesthesia. Two hours after BrdU administration, the rats were perfused as described above.

To reveal the BrdU-immunoreactive cells, the sections were washed in 0.1 M PBS, treated with 2 N HCl at 60°C for 30 min to fragment DNA, and subsequently neutralized in 0.1 M borate buffer, pH 8.4. To suppress endogenous peroxidase activity, the sections were incubated for 30 min in 20% methanol containing 0.3% H2O2. After blocking in normal serum the sections were exposed to 1:500 diluted BrdU primary monoclonal antibody (AbD Serotec, OBT0030CX, Oxford, UK), overnight at room temperature. The sections were then incubated with the secondary antibody (biotinylated goat anti-rat IgG, dilution 1:200), followed by standard ABC peroxidase solution (Vector, CA). DAB was used as a sensitive chromogen for visualization of BrdU-immunoreactive cells.

Fluoro-Jade C Histochemistry

In order to label dying cells in RMS, sections from one hemisphere from six animals in each experimental group were processed for Fluoro-Jade C (FJ-C; Histo-Chem Inc.) histochemistry. Sections were mounted on 2% gelatin-coated slides and air dried at 50°C for 30 min. Subsequently the slides were immersed in absolute alcohol for 3 min, for 1 min in 70% alcohol, for 1 min in distilled water, and then incubated in 0.06% potassium permanganate for 12 min. Then the slides were transferred for 1 h to FJ-C solution (1 μg/ml in 0.1% acetic acid). The proper dilution of staining solution was accomplished by first making a 0.01% stock solution of the dye in distilled water and then adding 1 ml of the stock solution to 99 ml of 0.1% acetic acid vehicle. The working solution was used within 2 h of preparation. The slides were then rinsed through three changes of distilled water for 1 min per change. The air-dried slides were cleared in xylene and cover slipped with DPX.

NADPH-d Histochemistry

NADPH-d reaction was performed on sections taken from one hemisphere from six animals in each experimental group. Free-floating sections were incubated in 0.1 M PB, pH 7.4 containing 0.4 mg/ml Nitroblue tetrazolium, 0.3% Triton X-100, 5 mg/ml malic acid, 4 mg/ml magnesium chloride and 0.8 mg/ml NADPH (ICN Biomedicals Inc.) at 37°C for 1 h (Scherer-Singler et al. 1983). The sections were then rinsed in 0.1 M PB, mounted on gelatin slides, air-dried overnight, cleared with xylene, and cover slipped with Entellan.

Quantitative Analysis

We have examined the number of BrdU, FJ-C, and NADPH-d-positive cells in the RMS of experimental as well as control rats. Only sections in which the entire RMS was visible were analyzed (six to eight sections for each animal). The caudal limit of the RMS was defined in sagittal sections as the point at which the lumen of the lateral ventricle opens up at the interface of the corpus callosum and striatum. The rostral limit was defined as the point at which the OB begins.

Sections processed for BrdU-immunohistochemistry were examined by light microscope (Olympus BX-51) equipped with camera system DP50 (Olympus). To evaluate these sections, images were captured. BrdU-positive cells were counted separately in three anatomical parts of the RMS, i.e., in the vertical arm, the horizontal arm, and in the elbow (Fig. 1) with aid of Disector program (version 2.0) for stereometric unbiased estimation of particle number (Tomori et al. 2001). The counting was performed in three-dimensional space browsing through serial sections. We applied the point-counting method needed for estimation of the particle density. The outcomes were expressed as the average number of BrdU-positive cells per mm3.

Fig. 1
figure 1

Sagittal section of the rat forebrain demonstrating BrdU-labeled cells in the rostral migratory stream (RMS). Vertical lines separate the individual parts of the RMS: I.—the vertical arm, II.—the elbow, III.—the horizontal arm. CC—corpus callosum; LV—lateral ventricle; SVZ—subventricular zone; OB—olfactory bulb. Scale bar = 1 mm

Full size image

Sections processed for FJ-C histochemistry were examined by the Olympus Reflected Fluorescence system U-RFL-T, the Olympus BX51 and the digital camera DP50. The amount of FJ-C and NADPH-d-positive cells was assessed semi-quantitatively by counting all the positive cells along the RMS manually with supporting Image Tool software. The number of NADPH-d and FJ-C positive cells was expressed as mean number of cells per section.

Statistical analyses were performed using the one-way ANOVA test, Tukey–Kramer and Student t-tests. The values were expressed as means ± S.E.M.

Results

Proliferating Cells in the RMS

The results of BrdU-immunohistochemistry showed that maternal separation for 3 h daily induced obvious qualitative changes in the RMS (Fig. 2a–f). In P14 control rats (Fig. 2a) a very low density of BrdU-positive nuclei has been observed in comparison with the examined later postnatal ages (Fig. 2c and e). At maternally separated animals of the same age, i.e., at P14, low density of dividing cells was even more striking (Fig. 2b). At P21 maternally separated rats, rare occurrence of BrdU-positive cells was most apparent in the RMS vertical arm (Fig. 2d). The density of dividing cells at P28 maternally separated rats lightly increased but it was still reduced when compared to the control animals.

Fig. 2
figure 2

Immunohistochemical staining for BrdU in the RMS of P14, P21, P28 control (left column) and maternally separated (right column) rats. The BrdU-positive cells are visible by brown stained nuclei. Note the low density of labeled cells both at P14 control (a) and experimental rats (b). Marked reduction of BrdU-positive cells in the RMS vertical arm of P21 maternally separated rat (d) in comparison with control animal of the same age (c). Different density of proliferating cells of P28 control (e) and maternally separated rats (f). OB-olfactory bulb. Scale bar = 100 μm

Full size image

In order to obtain detailed picture on the extent of proliferation, the number of BrdU-positive cells was counted in the RMS individual anatomical parts—the vertical and horizontal arms joined by the elbow. In our previous study, we have shown that the number of proliferating cells in the RMS gradually increases in the postnatal stage P14–P28 (Martončíková et al. 2006). In maternally separated rats, this pattern of increasing proliferation activity remained similar to that in the corresponding control groups; however, the number of proliferating cells in their RMS was decreased at all related intervals of investigation (Fig. 3a–c). Concerning the RMS single parts, this decrease was statistically significant in its caudal parts—the vertical arm and the elbow (Fig. 3a and b). These findings are inversely proportional to the control rats of the same ages as in these parts of the migratory stream the number of proliferating cells is the highest under physiological conditions throughout the life. Maternal separation induced decrease in proliferation was found also in the RMS horizontal arm, but this did not reach statistical significance in any experimental group (Fig. 3c).

Fig. 3
figure 3

Graphs show the number of BrdU-positive cells in the RMS vertical arm (a), elbow (b), and horizontal arm (c). Maternal separation decreased the number of proliferating cells at all intervals of investigation. Asterisk (*) indicates significant difference, p < 0.05

Full size image

Dying Cells in RMS

For detection of dying cells we have used a fluorochrome, FJ-C, which has been confirmed to detect dying neurons regardless of the cause of cell death (Schmued 2003; Mitrušková et al. 2005). Although FJ-C stained cells were present in the RMS both of control and experimental animals, the maternally separated rats showed higher density of labeled cells (Fig. 4a and b). Evaluation of the number of FJ-C-positive cells revealed that 2 weeks of maternal separation did not cause significant changes in dying cells number (Fig. 5) However, after 3 weeks lasting maternal separation (P21 rats) and at P28 animals, which survived 1 week after 3 weeks of maternal separation, the number of these cells was markedly increased in comparison with control rats (Fig. 5).

Fig. 4
figure 4

Representative photomicrographs of FJ-C positive cells in the RMS of P21 rats. Low number of positive cells (arrows) in the RMS of control animal (a). Increased number of FJ-C positive cells (arrows) in the RMS after 3 weeks of maternal separation (b). Scale bar = 50 μm

Full size image
Fig. 5
figure 5

Quantification of FJ-C positive cells in the RMS of control and maternally separated rats shows significant increase of labeled cells at the end of 3 weeks of maternal separation (P21 rats) and 1 week later (P28 rats). Asterisk (*) indicates significant difference, p < 0.05

Full size image

Cell Differentiation

The results of quantitative analysis of NADPH-d-positive cells showed that the number of these cells in the RMS of maternally separated rats was increased in comparison with corresponding control animals in all experimental groups (Fig. 6). This increase was statistically significant at P14 and P21 rats, i.e., immediately after the end of 2 or 3 weeks of maternal separation, respectively. In the RMS of rats surviving 1 week after 3 weeks of maternal separation, the number of nitrergic cells was still increased but this increase was not significant.

Fig. 6
figure 6

Quantification of NADPH-d positive cells in the RMS. The differences between control and maternally separated rats are significant (*p < 0.05) at the end of 2 weeks as well as 3 weeks of maternal separation (P14 and P21 rats)

Full size image

In addition to quantitative changes, maternally separated animals displayed an accelerated maturation of NO producing cells. NADPH-d-positive cells in P14 and P21 control rats had bipolar spindle-shaped cell body and at these ages only short fibers were visible (Fig. 7a). However, nitrergic cells at P14 and P21 maternally separated rats showed morphological characteristics that are typical for NADPH-d-positive cells of adult rats (Fig. 7b). In the RMS of these animals besides bipolar cells, triangular and multipolar cells were also present like in adult RMS. Typically, these cells had well-developed abundant varicose processes, which are normally detected only in adult animals.

Fig. 7
figure 7

Comparison of representative photomicrographs of NADPH-d staining in the RMS of P21 control (a) and maternally separated rats (b). The labeled cells in control animal have bipolar spindle-shaped labeled cell body with short fibers (a). The NADPH-d positive cells, in the RMS of maternally separated rat of the same age, are rather multipolar with well-developed varicose processes (b). The RMS is demarcated by dotted lines. Scale bar = 50 μm

Full size image

Discussion

Continuing neurogenesis in adult central nervous system is a phenomenon that might be of specific importance for neuroplasticity and regeneration in CNS. The mechanisms involved in the regulation of adult neurogenesis are therefore intensively studied. It has been confirmed that survival of newly generated neurons in the OB is dependent on their being incorporated into an active circuit. For example, exposure to specific odors increases the number of adult-born granule cells within the OB (Rochefort et al. 2002). Conversely, olfactory deprivation enhances OB granule cell apoptosis (Najbauer et al. 2002) and olfactory discrimination is decreased by genetic manipulations aimed at reducing the number of new cells integrated into the adult OB (Enwere et al. 2004).

Although mitotically active cells are present in the RMS of the adult rodent (Altman 1969; Lois and Alvarez-Buylla 1993), little is known about the factors that influence the ongoing neurogenesis specifically in this region. The present study demonstrates that early maternal separation strongly influences the process of neurogenesis in the RMS. Our quantitative analysis showed that the animals after MS had significantly reduced number of proliferating cells and increased cell dying in the neurogenic pathway. Previous studies investigated the effects of stressful experiences on adult neurogenis mainly in the hippocampus (Gould et al. 1998; Tanapat et al. 2001). In the similar maternal separation paradigm, Mirescu et al. (2004) investigated the effect of maternal separation on neurogenesis in the hippocampus and have reported that rat pups, subjected to daily maternal separation between P1 and P14, showed decreased proliferation of granule cell precursors in adulthood. The authors suggested that the observed downregulation of neurogenesis in the hippocampus is mediated through a corticosteroid-dependent mechanism. Although many evidence confirm the possibility that glucocorticoids are involved in the suppression of cell proliferation, this influence is not detectable by merely examining total peripheral levels of glucocorticoids (Mirescu and Gould 2006). Maternally separated rats by Mirescu et al. (2004) showed normal basal levels of plasma corticosterone. The authors explained the reduction in cell proliferation associated with maternal separation as a result of hypersensitivity to normal levels of corticosterone.

Similar to our findings, altered cell proliferation induced by olfactory cues has been demonstrated in the RMS of prairie voles (Smith et al. 2001). In female prairie voles, male exposure induced an increase in cell proliferation within the migratory pathway, possibly via an estrogen-mediated process. In the RMS of female prairie vole, enhanced cell proliferation has been also shown following social isolation (Fowler et al. 2001). Interestingly, this increase was significantly higher in comparison to male exposure. The authors explain these differences by the fact that prairie voles are typical glucocorticoid-resistant animal. In both experiments, changes in proliferation are supposed to be influenced by hormonal regulation. Glucocorticoids are known to negatively influence cell proliferation in the adult hippocampus (Tanapat et al. 2001; Yu et al. 2004). However, a role of these hormones in the regulation of RMS neurogenesis has not been confirmed and thus their contribution to the altered cell proliferation in the migratory stream is questionable.

In this study, quantitative analysis of proliferating cells number showed that the decrease of proliferation was not uniform in individual anatomical parts of the RMS. The most prominent decrease was observed in the caudal parts, i.e., in the vertical arm and elbow. It is known that the density of dividing cells decreases following a caudorostral gradient, with a maximum around the lateral ventricle and minimum in the subependymal layer of the OB (Altman 1969). We suppose that regional differences in response of the RMS anatomical parts to maternal separation is due to the higher vulnerability of those parts of the RMS, where the proliferation is highest under physiological conditions.

Changes in the number of BrdU-positive cells may have resulted not only from altered cell proliferation but also from their survival. Along the route of migration a considerable number of SVZ-generated cells undergo cell death in the rat (Brunjes and Armstrong 1996; Morshead et al. 1998). Here, we showed increased number of dying cells at the end of 3 weeks lasting maternal separation and 1 week later. Interestingly, the number of dying cells was not changed by 2 weeks of maternal separation. Considering the deepest decrease of BrdU-positive cells amount at this time point, the role of cell death on altered cell proliferation appears to be subtle.

Besides changes in cell proliferation and cell dying, our results indicate that stressful experiences caused by maternal separation during early life can alter the pattern of NO producing cells within the RMS. The observed increase of the NADPH-d-positive cells number and their early maturation raises a question whether NO could contribute to the altered proliferation or cell dying.

Recently NO has been shown to affect adult neurogenesis in different ways. The role of this gaseous messenger in adult neurogenesis has been demonstrated as antiproliferative effect described both in the SVZ and dentate gyrus of the hippocampus in physiological conditions. For example, Packer et al. (2003) found that the number of newly generated cells in the neurogenic regions of nNOS knockout mouse strain is dramatically increased. Moreover it has been demonstrated that the systematic or pharmacological inhibition of NO synthesis causes an increase of neuronal precursor proliferation in the SVZ (Matarredona et al. 2004; Moreno-López et al. 2004; Romero-Grimaldi et al. 2006; Torroglosa et al. 2007). The increase of the NADPH-d-positive cells number and coincident decrease of proliferating cells number in the RMS of maternally separated rats seems to support the antiproliferative effect of NO.

Another reported action of NO is the regulation of cell migration in the CNS. Although the role of NO in the migration of cells of the RMS is not fully understood, some facts indicate that NO produced from the blood vessels can influence cell migration within this neurogenic region. The investigations of Chen et al. (2005) using endothelial NOS (eNOS) knockout mice showed that the absence of eNOS impairs the migration of cells from SVZ explants. Moreover, migrating neuroblasts containing the machinery to respond to NO signaling are clustered together in small groups adjacent to blood vessels and can therefore be exposed to the NO produced by the eNOS from these vessels (Gutièrrez-Mecinas et al. 2007). We can speculate that the enhanced NADPH-d-positivity observed in blood vessels within the RMS of maternally separated rats (data not shown) could also influence the migration of the RMS precursor cells. Further studies are needed to identify the mechanisms by which environmental factors regulate the dynamic changes of cellular proliferation and migration in the RMS.

In conclusion, our findings have revealed that manipulation of olfactory environment by maternal separation exerts strong acute effect on neurogenesis in the rat RMS. Determining the mechanisms involved in regulation of neurogenesis by exogenous factors would help to better understand the functional relevance of postnatal neurogenesis.