Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Adult hippocampal neurogenesis is involved in anxiety-related behaviors


Adult hippocampal neurogenesis is a unique example of structural plasticity, the functional role of which has been a matter of intense debate. New transgenic models have recently shown that neurogenesis participates in hippocampus-mediated learning. Here, we show that transgenic animals, in which adult hippocampal neurogenesis has been specifically impaired, exhibit a striking increase in anxiety-related behaviors. Our results indicate that neurogenesis plays an important role in the regulation of affective states and could be the target of new treatments for anxiety disorders.


It has classically been assumed that the generation of new neurons is a process limited to brain development. Recent studies have, in fact, proven this assumption to be incorrect. Indeed, neurogenesis continues to occur in discrete regions of the adult mammalian brain, including that of humans.1, 2 The functional role of these newborn neurons in the adult dentate gyrus (DG) of the hippocampus has been the subject of extensive research and debate. On the basis of several correlative pieces of evidence, it was initially hypothesized that adult-born hippocampal neurons are involved in spatial memory assessed in this water maze (for review see Refs.3,4). However, it is only recently that two newly developed transgenic models, which allow ablating newborn neurons in the adult brain, have finally shown that complex forms of hippocampal-mediated memory, such as spatial relational memory, depend on the production of newborn neurons.5, 6, 7

The hippocampus is not only involved in cognitive functions but is also a key structure of the so-called emotional brain.8, 9 This structure plays a major role in regulating affective states and has been particularly associated with the modulation of anxiety states.10, 11 Anxiety is an excessive pathological form of fear (i.e., a fear state that is either too excessive for the actual threat or i.e. developed for a stimulus that is not threatening to the species). This disorder is the most frequent pathology in clinical psychiatry, representing nearly 30% of all psychiatric disorders.12, 13 The preponderance of this common public health problem shows the need for a greater understanding of its pathophysiological mechanisms. Here, we examined whether adult hippocampal neurogenesis is involved in affective states such as fear and anxiety. The potential role of neurogenesis in affective states is a matter of intense debate. Some reports have shown that behavioral effects of antidepressants are blocked by neurogenesis inhibition14, 15, 16, 17 suggesting that neurogenesis plays a role in depression-like behavior. However, the inhibition of neurogenesis per se does not modify anxiety- or depression-related behaviors.14, 15, 16, 17, 18, 19, 20, 21

To determine the extent to which neurogenesis is involved in affective states, and anxiety in particular, we studied anxiety-related behaviors using one of the transgenic models that allowed showing the influence of neurogenesis on spatial memory.5 In this model, adult hippocampal neurogenesis is selectively impaired by the over-expression of the pro-apoptotic protein Bax in neuronal precursors (for a more detailed description see Ref 5). Using several complementary behavioral tests, we show that a deficit of hippocampal neurogenesis increases anxiety-related behaviors, as measured by the avoidance of potentially threatening situations.22, 23, 24 In contrast, neurogenesis inhibition does not modify behaviors that are related to the affective sphere underlying depression, such as the reinforcing properties of novelty, novelty-suppressed feeding (NSF), or the forced swim test (FST). These results highlight the role of adult-born neurons in anxiety-related behaviors and provide the ground for the development of new therapeutic tools for anxiety disorders.

Materials and methods


Adult male transgenic mice were obtained from breeding Tet-Bax with Nestin-rtTA founder mice, as described earlier.5 Eight-week-old male Nestine-rtTA/Tet-Bax bigenic mice and their control littermates (Nestine-rtTA, Tet-Bax, and wild-type (Wt) were housed individually from the beginning of Dox treatment through the entirety of the experiments. Twenty-six C57BL6/J Wt mice (C57, Charles River, France) were used to control the effects of benzodiazepines on anxiety levels. A 12 h light/dark cycle (lights on from 8am to 8pm) was implemented in the animal house. Temperature (22±1 °C) and humidity (60±5%) were also controlled. All of the experiments were conducted in strict compliance with European Convention and institutional regulations.

In vivo doxycycline (Dox) treatment

Dox was added to the drinking solution (2 mg ml–1, 2.5% sucrose) starting 6 weeks before behavioral testing. Dox-free animals were given 2.5% sucrose in the drinking solution (Vehicle). In the first experiment (influence of neurogenesis on anxiety-related behaviors), 18 mice were used: bigenic-vehicle mice n=10; bigenic-Dox mice (BD) n=8. In the second experiment (influence of benzodiazepine treatment), all mice (n=38) were treated with Dox: control-Dox mice (CD=20)=(Wt, n=5), Tet-Bax (Bax, n=9), and nestin-rtTA (Nes, n=6)); BD=18. In the third experiment (influence of neurogenesis on depression-related behaviors), all mice were treated with Dox: control-Dox=10 (Wt, n=2; Bax, n=4; Nes, n=4); BD=9.

Measurements of anxiety-related behaviors

In the first experiment, the influence of depleting neurogenesis on anxiety-related behaviors was examined. At 14 weeks of age, 6 weeks after the beginning of Dox treatment, the mice were tested in four behavioral tasks (each separated by a 1-week interval): the elevated plus maze (EPM) in mildly anxiogenic conditions, the light/dark emergence test, the novel object test, and the predator exposure test.

The EPM was conducted in an apparatus composed of transparent plexiglas with two open (45 × 5 cm) and two enclosed (45 × 5 × 17 cm) arms that extended from a common central squared platform (5 × 5 cm). The floor of the maze was covered with black makrolon and was elevated 116 cm above the floor. A small raised lip (0.5 cm) around the edges of the open arms prevented animals from slipping off. The test session began with the mouse individually placed on the center square facing an open arm. Animals were allowed to freely explore the maze for 5 min under mildly anxiogenic conditions (dim light of 90 lux, using a halogen lamp in an experimental room with dark-colored posters pinned to the walls) or under anxiogenic conditions (bright light of 350 lux, using a classic lamp in an experimental room with bare white walls). A camera connected to a computer was used to track the mouse path during the entire session (VideoTrack, Viewpoint). Automatic path analysis measured the time spent in, the total number of entries into, and the distance traveled in the open and closed arms, the center square, and the entire apparatus.

The light/dark emergence test was conducted using an open field (50 × 50 cm, white PVC) containing a cylinder (10 cm deep, 6.5 cm in diameter, dark gray PVC) located length-wise along one wall, with the open end 10 cm from the corner. The mice were placed into the cylinder and tested for 15 min under bright light conditions (350 lux). Two trained observers who were blind to genotype scored the initial latency to emerge from the cylinder, defined as placement of all four paws into the open field, as well as the total number of exits from the cylinder and the total time spent inside the cylinder.

The novel object test was conducted in the same open field that was used for the light/dark emergence test. Although the mice were familiarized with the apparatus, they were allowed to freely explore the open field for 30 min in the absence of the object (‘empty’ condition). After this habituation phase, the mice were temporarily placed back into their home cage while a novel object (an earthenware cup measuring 8 cm in height and 7 cm in diameter) was placed upside-down in the center of the open field. Then animals were placed back into the open-field, now containing the cup (‘object’ phase), and tested for an additional 30 min. The time spent exploring the center of the open field (target quadrant) in the presence and absence of the cup was measured.

The predator exposure test was studied using a modified version of the amphetamine-activated rat exposure test.25 The test was conducted under dim light conditions (100 lux) in two contiguous open fields (50 × 50 cm each, white PVC) separated by a wire mesh. The first open field contained an escape cylinder (10 cm deep, 6.5 cm in diameter, gray PVC) located along one wall, with the open end 10 cm from one of the corners. During the habituation phase, the two open fields were separated by an opaque plastic wall. A mouse was placed in the center of the open field containing the cylinder and was allowed to freely explore the apparatus and the cylinder for 10 min. The mouse was then removed and placed back into its home cage while a second experimenter substituted the plastic separation wall with the wire mesh wall and placed the predator in the second open field. The predators used were adult male Sprague–Dawley rats (350–400 g) that were injected with D-amphetamine (2.5 mg kg–1 i.p.) 30 min before being placed in the open field. Each rat was used once and for only one mouse. The mouse was then placed back into the center zone of the first open-field and its behavior was analyzed for 5 min. Two trained observers who were blind to genotype scored the initial latency to enter into the cylinder, defined as placement of all four paws into the cylinder, as well as the total number of entries and the total time spent inside the cylinder. During exposure to the predator, three mice climbed the wire mesh and jumped-off of the open field and, thus, were not included in the data analysis. One mouse was excluded from the predator test as the rat showed excessive fear behavior during the exploration of the apparatus and, thus, did not represent a threat for the mouse.

Pharmacological treatment

In the second experiment, we examined the influence of benzodiazepine treatment on anxiety-related behaviors. Mice were injected with the widely used benzodiazepine chlordiazepoxide ((CDP), 7.5 mg kg–1, ip, Sigma-RBI, USA) or Vehicle solution (1% Cremophor EL, Sigma, USA) 15 min before exposure to the EPM (control-Dox-Veh=10, bigenic-Dox-Veh=9, control-Dox-CDP=10, bigenic-Dox-CDP=9). In this experiment, all animals received Dox treatment to avoid potential differences in the pharmacokinetics of CDP that might have occurred between sucrose-treated and Dox-treated animals. The control-Dox group included the following three genotypes: Wt, (n=5), Tet-Bax (Bax, n=9), and nestin-rtTA (Nes, n=6) mice. The dose of 7.5 mg kg–1 of CDP was chosen on the basis of a preliminary experiment (see supplementary Figure S1) carried out in C57BL6/J mice (C57). This experiment showed that 7.5 mg kg–1 of CDP decreased avoidance of the open arms when the EPM was carried out under anxious conditions (C57-veh=7, C57-CDP=7) (brightly lit environment) but had no effects under mildly anxious conditions (C57-veh=5, C57-CDP=7).

Measurements of depression-related behaviors

In the third experiment, the influence of depleting neurogenesis on depression-related behaviors was examined. At 16 weeks of age, 8 weeks after the beginning of Dox treatment, the mice were tested in two tasks classically used to evaluate the behavioral effects of antidepressants: the NSF and the FST (each task was separated by a 1-week interval).

The NSF was carried out during a 5 min period, as described earlier.16 Before testing, all food was removed from the home cage for 24 h; water was provided ad libitum. During the NSF, a single small food pellet (regular chow) was placed on a circular piece of white filter paper positioned in the center of an open field (50 × 50 cm) that was filled with approximately 2 cm of animal bedding. Each mouse was removed from its home cage and placed in a corner of the open field. The latency to the first bite of the food pellet was recorded (defined as the mouse sitting on its haunches and biting the pellet with the use of its forepaws).

The FST was carried out by individually placing mice into a glass cylinder (height 25 cm; diameter 18 cm) filled with 22 °C water to a depth of 15 cm. Behavior was recorded for 6 min with a camera positioned to view the side of the cylinder. The latency to float and the duration of immobility were scored off-line by an experimenter who was unaware of the experimental groups. A mouse was judged to be immobile when it remained floating in an upright position, making only the movements necessary to keep its head above the water.

Thymidine analog injection

Newly born cells were labeled by the incorporation of 5-iodo-2′-deoxyuridine (IdU, one ip injection of 57.6 mg kg–1 per day for 4 days).26

Immunohistochemistry and stereological analysis

Mice were perfused transcardially with a phosphate-buffered solution of 4% paraformaldehyde. After a 1-week post-fixation period, 40 μm sections were cut on a vibratome. Free-floating sections were processed in a standard immunohistochemical procedure in order to visualize doublecortin (Dcx; 1:1000; Santa Cruz, CA, USA), phosphorylated histone 3 (HH3, 1:2000, Upstate), activated caspase 3 (1:500, Cell Signaling Technology, MA, USA), and IdU (1/1000, BD PharMingen #347580). The number of immunoreactive (IR) cells throughout the entire granule and subgranular layers of the DG was estimated using the optical fractionator method.5, 26

Cell density measurements

The X-IR cell densities were estimated by the number of X-IR cells divided by the area of the structure, measured with Stereo Investigator software (Microbrightfield). Results are expressed as the mean number of X-IR cells mm–2.

Analysis of dendritic morphology

The overall dendritic tree in Dcx-IR neurons was measured as described earlier.27 Briefly, for each group, the more mature cells expressing Dcx were selected on the basis of the following criteria: (i) neurons exhibited vertically orientated dendrites that extended into the dentate molecular layer and (ii) dendrites of selected neurons had minimal overlap with the dendrites of adjacent cells to unambiguously trace the dendritic tree. All measurements were made with a × 100 objective, using a semi-automatic neuron-tracing system (Neurolucida; Microbrightfield, Colchester, VT, USA). In each age group, 48 DCX-IR neurons (provided from four animals) were traced in their entirety, and area of cell body, number of dendritic nodes and ends, and total dendritic length were calculated. To measure the extent of dendritic growth away from the soma and the branching of dendrites at different distances from the soma, the concentric circle analysis of Sholl28 was carried out using the NeuroExplorer component of the neurolucida program.

Measurement of EYFPBax intracellular aggregates

EYFPBax clusters were visualized using a Zeiss LSM 510 META confocal microscope5 and individually counted in the DG, the amygdala, and the hypothalamus.29 Results are expressed as cell densities.

Statistical analysis

Data were submitted to log transformation when a failure to follow normal distribution was observed. Differences between groups were analyzed using Student's t-test or an ANOVA, followed by a post-hoc comparison using the Duncan test when appropriate. Relationships between the various behavioral measures recorded in the EPM were analyzed using a factor analysis with a principal components solution and orthogonal rotation (varimax) of the factor matrix.24 This method ensures that the extracted factors are independent of one another. Factor pattern matrices were identified using a combination of the Kaiser criterion (factors must have eigenvalues≥1) and the Cattell Scree test (on a simple line plot, the point at which the smooth decreases in eigenvalues levels-off to the right). The factor loading of each behavioral item indicates the degree to which that item correlates with the factor; thus, a loading of ±1.0 indicates a perfect (positive/negative) correlation, whereas a loading of <0.4 would suggest that the item is weakly linked to the factor.


Doxycycline treatment in bigenic rtTA-Bax mice reduces hippocampal neurogenesis

The genetic model that we have used is based on double transgenic mice (bigenic mice), in which neural precursors can be selectively killed through the over-expression of the pro-apoptotic protein Bax. The inducible over-expression of Bax in neural precursors was obtained using the reverse tetracycline-controlled transactivator (rtTA)-regulated system that is activated by the oral administration of an exogenous tetracycline analog, doxycycline (Dox). As shown earlier,5 treatment with Dox induced a profound decrease in cell proliferation and in neurogenesis in the DG of bigenic mice. Indeed, the number of proliferating cells measured by staining for HH3, (Figure 1a) was reduced in BD. Furthermore, the number of newborn neurons expressing Dcx, a reliable marker of adult neurogenesis,27 was also largely decreased in BD (Figure 1b). A morphometric evaluation of the dendritic arbor of Dcx-IR neurons showed that the number of nodes and endings, and dendritic length of Big-Dox and Big-Veh neurons were similar (Supplementary Table S1). Moreover, the Sholl analysis confirmed that the dendritic complexity of Big-Dox neurons was similar to that of control Big-Veh neurons (Figure 1c).

Figure 1

Decrease in neurogenesis in the adult dentate gyrus (DG) of bigenic animals treated with Dox. (a) Number of HH3-IR cells (t16=1.8, P=0.04). Insert: illustration of a HH3-IR cells. Bar scale=10 μm. (b) Number of Dcx-IR (t15=2.86, P=0.01). Illustration of Dcx-IR immature neurons. Bar scale=50 μm. (c) Illustration of Dcx-IR immature neurons at a higher magnification. Bar scale=10 μm. Sholl analysis of Big-Veh and Big-Dox neurons (group effect: F1,6=0.07, P=0.79; group x level interaction: F25,150=0.45, P=0.99). *P<0.05; **P<0.01 in comparison to the control group. White bar=bigenic-vehicle mice. Black bar=bigenic-Dox mice.

Cell genesis in the subventricular zone (SVZ), another neurogenic area, was not significantly modified after Dox treatment (Supplementary Figure S2a), which confirms results from our earlier report.5 Given the important role of the amygdala and hypothalamus in fear and anxiety, we also analyzed the expression of the EYFPBax fusion protein and of the activated form of caspase 3 in these brain areas. No increase in EYFPBax expression or in caspase 3 was found in the amygdala or in the hypothalamus (Supplementary Figure S3, Table S2), indicating that the transgene is not expressed in these zones.

Finally, in a subsequent experiment, we examined whether compensatory mechanisms could modify the long-term survival of newly born cells that were unaffected by Bax over-expression. To this end, bigenic-Veh mice and BD were injected with IdU, an analogue of thymidine, and were killed 4 months later. In this condition, the number of IdU-labeled cells also decreased in the DG of bigenic mice treated with Dox (Supplementary Figure S2b).

Inhibition of hippocampal neurogenesis increases anxiety-related behaviors

Anxiety-related behaviors in rodents are primarily studied by measuring avoidance responses to potentially threatening situations, such as open and brightly lit environments, which expose mice to predators. Anxiety-like behaviors were first tested in bigenic mice using the EPM. As expected, control animals (bigenic-vehicle mice) spent less time in the open arms (constituting the threatening areas) than in the enclosed arms (Figure 2a). The avoidance of the threatening area was largely increased in animals in which neurogenesis was impaired (BD); these animals almost completely avoided the open arms (Figure 2a–d). This effect did not depend on a non-specific modification of locomotor activity (Supplementary Table S3). As shown by a principal component analysis, behavior in the EPM is explained by two factors: a locomotor activity factor and an anxiety factor (Supplementary Figure S4). Inhibition of neurogenesis had no effects on the behavioral parameters loading on the locomotor activity factor (see also Supplementary Table S3). In contrast, neurogenesis inhibition selectively modified all of the behavioral parameters loading on the anxiety factor (Supplementary Figure S4).

Figure 2

Depletion of adult hippocampal neurogenesis increases anxiety-related behaviors in the elevated plus maze (EPM) and in the light/dark emergence test. (a) Illustrations of the 3D spatial distribution of the exploratory path in bigenic-Dox mice, in which neurogenesis has been inhibited, and in control bigenic-vehicle mice. During the EPM test, bigenic-Dox mice (in comparison to control bigenic-vehicle mice): (b) spent less time in the open arms (t16=5.554, P=0.00004); (c) covered less distance in the open arms (t16=2.219, P=0.041); (d) visited the extremities of the open arms fewer times (t16=2.162, P=0.046). In the light/dark emergence test, bigenic-Dox mice (in comparison to control bigenic-vehicle mice): (e) showed a higher initial latency to emerge from the protective cylinder (t16=−2.93, P=0.009); (f) spent more time in the protective cylinder (t16=−2.16, P<0.05); (g) had a higher mean time per entry into the cylinder (t16=−3.67, P<0.01). *P<0.05; **P<0.01; ***P<0.001 in comparison to the bigenic vehicle controls. White bar=bigenic-vehicle mice. Black bar=bigenic-Dox mice.

In the light/dark emergence test, the latency to exit a reassuring cylinder (emergence) is considered as an index of anxiety-like behavior in mice. Animals with inhibited neurogenesis (BD) showed significantly longer latencies to emerge from the protective cylinder (Figure 2e) than did controls (bigenic-vehicle mice). In addition, the total time spent in the cylinder (Figure 2f) and the mean time per entry (Figure 2g) were higher in BD than in bigenic-vehicle mice.

To verify the non-specific effects of genetic and pharmacological manipulations, the complete range of mice strains (bigenic mice, Wt mice, and the two parental strains: Tet-Bax and nestin-rtTA mice) treated or not treated with Dox was also tested. It was found that none of the control groups treated either with vehicle or Dox differed from bigenic mice treated with vehicle (Supplementary Figures S5).

Anxiety-related behavior results from a distorted assessment of risk-related information

The results of these experiments suggest that inhibition of hippocampal neurogenesis increases anxiety-related behaviors. However, the time spent in novel open areas results from the computation of two opposite motivational forces. The first force, driving avoidance, is the fear of potential threat such as predators, which can easily stalk a prey in an open space. The second force, driving exploration, is novelty seeking. Indeed, novel environments also constitute an opportunity to discover new foraging sources. Consequently, a decrease in the time spent in the open arms of the EPM or an increase in the latency to exit the protective cylinder in the light/dark emergence test could result from either an increase in the fear of potential threats or a decrease in novelty exploration.

To address this issue, we evaluated animals in which neurogenesis was impaired in both the defensive reaction to predator exposure25, 30 and the exploration of a novel object in a non-threatening environment.22 The latency to hide in a protective cylinder, which is considered an index of fear of the predator, was much shorter (Figure 3a) in animals in which neurogenesis was impaired (bigenic-Dox) than in control mice (bigenic-vehicle). Bigenic-Dox mice also spent more time in the protective cylinder than did controls (Figure 3b). In contrast, inhibiting neurogenesis did not influence the exploration of a novel object in a non-threatening environment (Figure 3c). Both bigenic-vehicle and BD significantly explored the novel object, as shown by the increase in the time spent in the target quadrant when the object was present. However, the two groups did not differ on this behavior (Figure 3c, see also Supplementary Figure S5).

Figure 3

Depletion of adult hippocampal neurogenesis increases predator avoidance but does not modify novelty exploration. When placed in the center of an open-field in the presence of a predator, bigenic-Dox mice, in which neurogenesis was inhibited (in comparison to control bigenic-vehicle mice): (a) retreated more quickly into the protective cylinder (t15=2.12, P=0.05); (b) spent more time inside the cylinder (t15=−2.86, P=0.011). (c) Bigenic-vehicle and bigenic-Dox mice increased exploration of the target zone of the open-field when it contained a novel object (F1,16=41.63, P=0.000008) and showed a comparable level of novelty exploration (F1,16=0.13, P>0.7). *P<0.05 in comparison to the control group; ooP<0.01, oooP<0.001, test phase (object) compared with habituation phase (empty). White bar=bigenic-vehicle mice. Black bar=bigenic-Dox mice.

Neurogenesis does not mediate the anxiolytic effect of benzodiazepine

To further attest the implication of neurogenesis in anxiety-related behaviors, we analyzed the effects of the prototypic anxiolytic, the benzodiazepine CDP. CDP decreased the anxiety-related behaviors of animals in which neurogenesis was impaired. Thus, CDP increased the exploration of the open arms in BD, such that the level of exploration was similar to that of control animals (Figure 4). CDP had similar effects on control animals only when their level of anxiety was increased by carrying out the test in more anxiogenic conditions (Supplementary Figure S1).

Figure 4

The increase in anxiety-related behaviors induced by neurogenesis inhibition is reversed by benzodiazepine treatment. The injection of the benzodiazepine chlordiazepoxide ((CDP), 7.5 mg kg–1, i.p.) before the elevated plus maze (EPM) completely reversed the increased avoidance of the open arm of bigenic-Dox mice (BD), while it did not modify the behavior of control-Dox mice (CD) (F1,34=5.211, P<0.05; BD different from all other groups at *P<0.05). White bar=vehicle injection, Stripped bar=CDP injection.

Inhibition of hippocampal neurogenesis does not modify depression-related behaviors

In a supplementary experiment, we analyzed whether a deficit in neurogenesis also modified behaviors related to other dimensions of the affective sphere. In particular, we studied the effects of neurogenesis inhibition on behavioral tests that are generally used in the context of studies on depression, such as the NSF and the FST. In the NSF, the latency to feed in CD and BD was similar (Figure 5a). In the FST, the latency to immobility (Figure 5b) and the total time spent floating (Figure 5c) were similar in CD and BD mice. These data indicate, in contrast to what was observed for anxiety-related behaviors, that inhibition of hippocampal neurogenesis does not modify depression-related behaviors.

Figure 5

Depletion of adult hippocampal neurogenesis does not modify depression-related behaviors such as the novelty-suppressed feeding (NSF) test and the forced swim test (FST). (a) During the NSF test, the latency to feed in control-Dox mice (CD) and bigenic-Dox mice (BD) was similar (t17=−0.008, P=0.99). In the FST, the latency to immobility (b, t17=−0.17, P=0.87) and the time of floating (c, t17=−1.188, P=0.1) in control-Dox mice (CD) and bigenic-Dox mice (BD) were similar. White bar=CD. Black bar=BD.


Altogether, these results indicate that a decrease in adult-born neurons increases anxiety-related behaviors and may be involved in the pathophysiology of anxiety. Animals in which neurogenesis was impaired showed an increased avoidance of novel and potentially threatening environments. This enhanced avoidance was not because of a decreased sensitivity to the positive reinforcing effects of novel stimuli as inhibition of neurogenesis did not modify novelty exploration in non-threatening situations. In contrast, decrease in neurogenesis likely increased the negative impact of the potential threat that is associated with novel unprotected environments. Thus, impairment of neurogenesis strongly increased the response to an actual threat, such as the presence of a predator. However, in the case of a learned fear evaluated during the contextual or the cue conditioning, no behavioral deficits were observed.5 This finding indicates that altering neurogenesis selectively targets anxiety and not fear. This seeming specificity of neurogenesis impairment on the affective sphere related to anxiety is further supported by the lack of effects of neurogenesis inhibition on depression-like behaviors, such as NSF and FST.

Two recent reports have suggested a link between neurogenesis and anxiety-related behavior.31, 32 In both studies, a partial (around 20%) or a profound decrease in neurogenesis was associated with an increase in some anxiety-related behaviors. However, in both studies, the genetic manipulation that resulted in a decrease in neurogenesis also modified other biological factors, such as the TrkB receptor or the expression of Activin, which could modify brain functioning and behavior. Consequently, our results extend these earlier studies by providing the first solid evidence of a causal link between neurogenesis and anxiety-related behaviors.

Our results provide a further link between neurogenesis and the pathophysiology of anxiety by showing that reference treatments of anxiety, such as benzodiazepine anxiolytics, can reverse the enhanced fear response of animals in which neurogenesis has been impaired. These results are in agreement with those found in another pathophysiological model of anxiety involving the hippocampus. In this model, heterozygous mice with a deletion of the γ2 subunit of GABAA receptor in the DG33 display increased anxiety-like behavior that is reversed by benzodiazepine. These observations suggest that neurogenesis, and the likely the hippocampus more generally, is involved in the genesis of anxiety-related behaviors but does not mediate the anxiolytic effect of benzodiazepine.

Earlier attempts to show modifications of anxiety-related behaviors after manipulation of neurogenesis in the adult have failed to provide conclusive results.18, 19, 20, 21 However, in these studies, manipulation of neurogenesis was carried out using less specific tools, such as methylazoxymethanol acetate (MAM) administration or irradiation. This discrepancy is similar to that observed earlier with spatial relational memory. Thus, MAM administration or irradiation also failed to show an effect of neurogenesis on spatial memory measured in the water maze. This lack of effect could be because of a lesser specificity of MAM treatment and irradiation, which could induce confounding behavioral perturbations. In addition, differences in behavioral protocols could also account for differences between studies (for a detailed discussion see Refs 5).

It has been reported earlier that depletion of adult neurogenesis using transgenic models leads to an impairment in spatial memory in the water maze.5, 7 These findings do not conflict with the present results or challenge the functional implication of adult-born neurons in the relational processing of incoming information. Indeed, the hippocampus, beyond its classical role in memory, forms the central part of the ‘emotional brain’. These two different components have been related to regional dissociations between the dorsal and the ventral hippocampus.10 Indeed, the dorsal hippocampus is preferentially involved in memory-related functions, whereas the ventral hippocampus has a preferential role in processing emotional information such as anxiety-related fear. Neurogenesis in bigenic-Dox mice is reduced throughout the rostro-caudal axis of the DG.5 Thus, it can be hypothesized that a failure in ventral hippocampal neurogenesis is involved in the appearance of anxiety whereas depletion of adult-born neurons in the dorsal section sustains memory deficits.

It is also possible that the impairment of cognitive functions induced by neurogenesis inhibition contributes to the increased anxiety-related behaviors observed in BD. Thus, it has been proposed that anxiety results from a cognitive dysfunction that does not provide an appropriate filter for environmental information; thereby, producing an inappropriate ‘emotional’ response.34, 35 A reduction in neurogenesis could then alter the ability to efficiently integrate novel incoming information, amplifying the weight of negative associations and/or leading to an over-reaction to aversive cues that are interpreted as threatening. Following this hypothesis, the anxiety-like phenotype observed here would be the consequence of a distorted assessment of risk-related information.

In conclusion, using an inducible transgenic strategy that allows specific ablation of newborn neurons in the adult DG of the hippocampus, we show that a deficit of hippocampal neurogenesis leads to an increase in anxiety-related behaviors. These results reinforce the fascinating hypothesis that adult neurogenesis represents a unique example of structural plasticity that plays a pivotal role in both the physiology and the pathophysiology of the hippocampus. The extension of this knowledge to affective states provides a new avenue of investigation for developing new treatments of anxiety.


  1. 1

    Abrous DN, Koehl M, Le Moal M . Adult neurogenesis: from precursors to network and physiology. Physiol Rev 2005; 85: 523–569.

    CAS  Article  Google Scholar 

  2. 2

    Eriksson PS, Perfilieva E, Björk-Eriksson T, Alborn AM, Nordborg C, Peterson DA et al. Neurogenesis in the adult human hippocampus. Nat Med 1998; 4: 1313–1317.

    CAS  Article  Google Scholar 

  3. 3

    Abrous DN, Wojtowicz JM . Neurogenesis and hippocampal memory system. Cold Spring Harbor Laboratory Press: New York, 2008, 445–461.

  4. 4

    Leuner B, Gould E, Shors TJ . Is there a link between adult neurogenesis and learning? Hippocampus 2006; 16: 216–224.

    Article  Google Scholar 

  5. 5

    Dupret D, Revest JM, Koehl M, Ichas F, De Giorgi F, Costet P et al. Spatial relational memory requires hippocampal adult neurogenesis. PlosOne 2008; 3: e1959.

    Article  Google Scholar 

  6. 6

    Imayoshi I, Sakamoto M, Ohtsuka T, Takao K, Miyakawa T, Yamaguchi M et al. Roles of continuous neurogenesis in the structural and functional integrity of the adult forebrain. Nat Neurosci 2008; 11: 1153–1161.

    CAS  Article  Google Scholar 

  7. 7

    Zhang CL, Zou Y, He W, Gage FH, Evans RM . A role for adult TLX-positive neural stem cells in learning and behaviour. Nature 2008; 451: 1004–1007

    CAS  Article  Google Scholar 

  8. 8

    MacLEAN PD . Psychosomatic disease and the visceral brain; recent developments bearing on the Papez theory of emotion. Psychosom Med 1949; 11: 338–353.

    CAS  Article  Google Scholar 

  9. 9

    Papez JW . A proposed mechanism for emotion. Arch Neurol Psychiatry 1937; 79: 237–244.

    Google Scholar 

  10. 10

    Bannerman DM, Matthews P, Deacon RM, Rawlins JN . Medial septal lesions mimic effects of both selective dorsal and ventral hippocampal lesions. Behav Neurosci 2004; 118: 1033–1041.

    CAS  Article  Google Scholar 

  11. 11

    Gray JA . The Neuropsychology of Anxiety. Clarendon Press: Oxford, 1982.

    Google Scholar 

  12. 12

    Kessler RC, Birnbaum H, Demler O, Falloon IR, Gagnon E, Guyer M et al. The prevalence and correlates of nonaffective psychosis in the National Comorbidity Survey Replication (NCS-R). Biol Psychiatry 2005; 58: 668–676.

    Article  Google Scholar 

  13. 13

    Mineka S, Watson D, Clark LA . Comorbidity of anxiety and unipolar mood disorders. Annu Rev Psychol 1998; 49: 377–412.

    CAS  Article  Google Scholar 

  14. 14

    Li Y, Luikart BW, Birnbaum S, Chen J, Kwon CH, Kernie SG et al. TrkB regulates hippocampal neurogenesis and governs sensitivity to antidepressive treatment. Neuron 2008; 59: 399–412.

    CAS  Article  Google Scholar 

  15. 15

    Pollak DD, Monje FJ, Zuckerman L, Denny CA, Drew MR, Kandel ER . An animal model of a behavioral intervention for depression. Neuron 2008; 60: 149–161.

    CAS  Article  Google Scholar 

  16. 16

    Santarelli L, Saxe M, Gross C, Surget A, Battaglia F, Dulawa S et al. Requirement of hippocampal neurogenesis for the behavioral effects of antidepressants. Science 2003; 301: 805–809.

    CAS  Article  Google Scholar 

  17. 17

    Surget A, Saxe M, Leman S, Ibarguen-Vargas Y, Chalon S, Griebel G et al. Drug-dependent requirement of hippocampal neurogenesis in a model of depression and of antidepressant reversal. Biol Psychiatry 2008; 64: 293–301.

    CAS  Article  Google Scholar 

  18. 18

    Madsen TM, Kristjansen PE, Bolwig TG, Wortwein G . Arrested neuronal proliferation and impaired hippocampal function following fractionated brain irradiation in the adult rat. Neuroscience 2003; 119: 635–642.

    CAS  Article  Google Scholar 

  19. 19

    Raber J, Rola R, LeFevour A, Morhardt D, Curley J, Mizumatsu S et al. Radiation-induced cognitive impairments are associated with changes in indicators of hippocampal neurogenesis. Radiat Res 2004; 162: 39–47.

    CAS  Article  Google Scholar 

  20. 20

    Saxe MD, Battaglia F, Wang JW, Malleret G, David DJ, Monckton JE et al. Ablation of hippocampal neurogenesis impairs contextual fear conditioning and synaptic plasticity in the dentate gyrus. Proc Natl Acad Sci USA 2006; 103: 17501–17506.

    CAS  Article  Google Scholar 

  21. 21

    Shors TJ, Townsend DA, Zhao M, Kozorovitskiy Y, Gould E . Neurogenesis may relate to some but not all types of hippocampal-dependent learning. Hippocampus 2002; 12: 578–584.

    Article  Google Scholar 

  22. 22

    Dulawa SC, Grandy DK, Low MJ, Paulus MP, Geyer MA . Dopamine D4 receptor-knock-out mice exhibit reduced exploration of novel stimuli. J Neurosci 1999; 19: 9550–9556.

    CAS  Article  Google Scholar 

  23. 23

    Lister RG . Ethologically-based animal models of anxiety disorders. Pharmacol Ther 1990; 46: 321–340.

    CAS  Article  Google Scholar 

  24. 24

    Rodgers RJ, Johnson NJ . Factor analysis of spatiotemporal and ethological measures in the murine elevated plus-maze test of anxiety. Pharmacol Biochem Behav 1995; 52: 297–303.

    CAS  Article  Google Scholar 

  25. 25

    Yang M, Augustsson H, Markham CM, Hubbard DT, Webster D, Wall PM et al. The rat exposure test: a model of mouse defensive behaviors. Physiol Behav 2004; 81: 465–473.

    CAS  Article  Google Scholar 

  26. 26

    Dupret D, Fabre A, Döbrössy MD, Panatier A, Rodríguez JJ, Lamarque S et al. Spatial learning depends on both the addition and removal of new hippocampal neurons. PLOS Biology 2007; 5: 1683–1694.

    CAS  Article  Google Scholar 

  27. 27

    Rao MS, Shetty AK . Efficacy of doublecortin as a marker to analyse the absolute number and dendritic growth of newly generated neurons in the adult dentate gyrus. Eur J Neurosci 2004; 19: 234–246.

    Article  Google Scholar 

  28. 28

    Sholl DA . Dendritic organization in the neurons of the visual and motor cortices of the cat. J Anat 1953; 87: 387–406.

    CAS  PubMed  PubMed Central  Google Scholar 

  29. 29

    Kokoeva MV, Yin H, Flier JS . Neurogenesis in the hypothalamus of adult mice: potential role in energy balance. Science 2005; 310: 679–683.

    CAS  Article  Google Scholar 

  30. 30

    Belzung C, El Hage W, Moindrot N, Griebel G . Behavioral and neurochemical changes following predatory stress in mice. Neuropharmacology 2001; 41: 400–408.

    CAS  Article  Google Scholar 

  31. 31

    Ageta H, Murayama A, Migishima R, Kida S, Tsuchida K, Yokoyama M et al. Activin in the brain modulates anxiety-related behavior and adult neurogenesis. PLoS ONE 2008; 3: e1869.

    Article  Google Scholar 

  32. 32

    Bergami M, Rimondini R, Santi S, Blum R, Götz M, Canossa M . Deletion of TrkB in adult progenitors alters newborn neuron integration into hippocampal circuits and increases anxiety-like behavior. Proc Natl Acad Sci USA 2008; 105: 15570–15575.

    CAS  Article  Google Scholar 

  33. 33

    Crestani F, Lorez M, Baer K, Essrich C, Benke D, Laurent JP et al. Decreased GABAA-receptor clustering results in enhanced anxiety and a bias for threat cues. Nat Neurosci 1999; 2: 833–839.

    CAS  Article  Google Scholar 

  34. 34

    McNaughton N . Cognitive dysfunction resulting from hippocampal hyperactivity-a possible cause of anxiety disorder? Pharmacol Biochem Behav 1997; 56: 603–611.

    CAS  Article  Google Scholar 

  35. 35

    Tsetsenis T, Ma XH, Lo IL, Beck SG, Gross C . Suppression of conditioning to ambiguous cues by pharmacogenetic inhibition of the dentate gyrus. Nat Neurosci 2007; 10: 896–902.

    CAS  Article  Google Scholar 

Download references


The authors are grateful to Dr F Chaouloff for his helpful comments. The technical help of Mr C Dupuy, Miss S Lamarque, Miss D Gonzales, and Mrs V Roullot–Lacarrière is greatly acknowledged. This work was financially supported by INSERM and Agence Nationale pour la Recherche (ANR to DNA). CFR was funded by the ‘Ministère délégué à l’Enseignement Supérieur et à la Recherche' and NG by ANR.

Author information



Corresponding authors

Correspondence to P-V Piazza or D N Abrous.

Additional information

Supplementary Information accompanies the paper on the Molecular Psychiatry website (

Supplementary information

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Revest, JM., Dupret, D., Koehl, M. et al. Adult hippocampal neurogenesis is involved in anxiety-related behaviors. Mol Psychiatry 14, 959–967 (2009).

Download citation


  • anxiety
  • neurogenesis
  • hippocampus
  • behavior
  • transgenesis
  • tet-on system
  • pro-apoptotic protein Bax

Further reading


Quick links