The Effects of Stress on Hippocampal Neurogenesis and Behavior in the Absence of Lipocalin-2

Abstract

Lipocalin-2 (LCN2) is an acute phase protein able to bind iron when complexed with bacterial siderophores. The recent identification of a mammalian siderophore also suggested a physiological role for LCN2 in the regulation of iron levels and redox state. In the central nervous system, the deletion of LCN2 induces deficits in neural stem cells proliferation and commitment, with an impact on the hippocampal-dependent contextual fear discriminative task. Additionally, stress is a well-known regulator of cell genesis and is known to decrease adult hippocampal cell proliferation and neurogenesis. Although voluntary running, another well-known regulator of neurogenesis, is sufficient to rescue the defective hippocampal neurogenesis and behavior in LCN2-null mice by promoting stem cells' cell cycle progression and maturation, the relevance of LCN2-regulated hippocampal neurogenesis in response to stress has never been explored. Here, we show a lack of response by LCN2-null mice to the effects of chronic stress exposure at the cellular and behavioral levels. Together, these findings implicate LCN2 as a relevant mediator of neuronal plasticity and brain function in the adult mammalian brain.

Keywords: anxiety; corticosterone; hippocampal neurogenesis; lipocalin-2; memory.

Figure 1

The effects of chronic CORT treatment on biological parameters. (a) Representative scheme of the experimental procedure of CORT injections performed. The thymidine analog 5′-bromo-2′-deoxyuridine (BrdU) was injected for 5 days at the beginning of the protocol and, for an additional 28 days, Wt and LCN2-null mice were daily injected subcutaneously with corticosterone (CORT group) or sesame oil (Vehicle group). (b) Body weight gain in Wt and LCN2-null mice after chronic CORT injection (n = 6–10 mice per group) were monitored weekly during the protocol of CORT treatment. (c,d) CORT treatment significantly decreased adrenal and thymus weight, regardless of the genotype. Data are presented as mean ± SEM and were analyzed by two-way ANOVA with Bonferroni’s multiple comparison test. ^δ Denotes differences between vehicle and CORT Wt, and ^# between vehicle and CORT LCN2-null mice. ^δδδ p ≤ 0.001, ^{####,δδδδ} p ≤ 0.0001.

Figure 2

Chronic CORT administration reduces cell proliferation and neural stem cells self-renewal and survival. (a) CORT exposure robustly decreased the number of Ki67⁺ proliferative cells, specifically in Wt animals, whereas it did not significantly impact cell survival (BrdU⁺ cells) (n = 4–6 per group). (b) Representative images of Ki67 and BrdU immunostaining (indicated by white arrows) in the DG of Wt and LCN2-null mice injected with vehicle and CORT. (c) Quantification of the number of radial glia-like type-1 stem cells, as GFAP⁺ BrdU⁺ cells in the SGZ, revealed that CORT treatment induced a significant decrease in both Wt and LCN2-null mice (n = 4 mice per group). The effect of CORT on type-2 stem cells was only evident in Wt animals (n = 5 mice per group). (d) Representative images of GFAP⁺/BrdU⁺ and Sox2⁺/Ki67⁺ immunostaining (indicated by white arrows) in the DG of Wt and LCN2-null control mice and those injected with CORT. Scale bars, 100 μm. Data are presented as mean ± SEM and were analyzed by two-way ANOVA with Bonferroni’s multiple comparison test. * Denotes differences between vehicle Wt and LCN2-null mice; ^δ between vehicle and CORT Wt; and ^# between vehicle and CORT LCN2-null mice. * p ≤ 0.05, ^δδ p ≤ 0.01, ^δδδ p ≤ 0.001,****^{, ####} p ≤ 0.0001.

Figure 3

Animal behavior and hippocampal neurogenesis are differentially affected by CORT in Wt and LCN2-null animals. (a) Anxiety assessment in the EPM test, after CORT treatment, showed that Wt mice, but not LCN2-null animals, spent less time in the open arms, with no effect among groups on general motor activity (n = 6–10 mice per group). (b) In the NSF paradigm, Wt animals increased their latency to feed after the chronic treatment with CORT. No differences were observed between groups in appetite drive (n = 6–10 mice per group). (c) Contextual retrieval in context A by Wt mice was impaired after CORT treatment (n = 6–10 mice per group) but not by LCN2-null mice. (d) Discrimination index was affected by the chronic CORT treatment, but only in Wt animals (n = 6–10 mice per group). (e) Quantitative analysis of the number of newborn mature neurons in DG, as Calb⁺ BrdU⁺ cells, revealed that CORT treatment induced a significant decrease in Wt mice (n = 5 mice per group), with no effect on LCN2-null DG. (f) Representative images of calbindin and BrdU immunostaining (indicated by white arrows) in DG of Wt and LCN2-null vehicle mice and those injected with CORT. Scale bar, 100 μm. Data are presented as mean ± SEM and were analyzed by two-way ANOVA with Bonferroni’s multiple comparison test. * Denotes differences between vehicle Wt and LCN2-null mice, and ^δ between vehicle and CORT Wt. *^,δ p ≤ 0.05, ^δδ p ≤ 0.01.