cAMP response element-binding protein deficiency allows for increased neurogenesis and a rapid onset of antidepressant response

Abstract

cAMP response element-binding protein (CREB) has been implicated in the molecular and cellular mechanisms of chronic antidepressant (AD) treatment, although its role in the behavioral response is unclear. CREB-deficient (CREB(alpha delta) mutant) mice demonstrate an antidepressant phenotype in the tail suspension test (TST) and forced-swim test. Here, we show that, at baseline, CREB(alpha delta) mutant mice exhibited increased hippocampal cell proliferation and neurogenesis compared with wild-type (WT) controls, effects similar to those observed in WT mice after chronic desipramine (DMI) administration. Neurogenesis was not further augmented by chronic DMI treatment in CREB(alpha delta) mutant mice. Serotonin depletion decreased neurogenesis in CREB(alpha delta) mutant mice to WT levels, which correlated with a reversal of the antidepressant phenotype in the TST. This effect was specific for the reversal of the antidepressant phenotype in these mice, because serotonin depletion did not alter a baseline anxiety-like behavior in CREB(alpha delta) mutant mice. The response to chronic AD treatment in the novelty-induced hypophagia (NIH) test may rely on neurogenesis. Therefore, we used this paradigm to evaluate chronic AD treatment in CREB(alpha delta) mutant mice to determine whether the increased neurogenesis in these mice alters their response in the NIH paradigm. Whereas both WT and CREB(alpha delta) mutant mice responded to chronic AD treatment in the NIH paradigm, only CREB(alpha delta) mutant mice responded to acute AD treatment. However, in the elevated zero maze, DMI did not reverse anxiety behavior in mutant mice. Together, these data show that increased hippocampal neurogenesis allows for an antidepressant phenotype as well as a rapid onset of behavioral responses to AD treatment.

Figure 1.

A–C, CREB deficiency increases the number of BrdU-positive cells at 4 d (A) and 6 weeks (B) in the dentate gyrus, with the majority of BrdU-positive cells differentiating into neurons (C). Saline-treated CREB^αΔ mutant mice demonstrated significantly increased number of BrdU-positive cells compared with WT controls (*p < 0.05; n = 6–8). This increase was maintained at the 6 week time point, at which the majority of the BrdU-positive cells matured into neurons, indicating a net increase in neurogenesis in the CREB^αΔ mutant mice. Chronic DMI treatment significantly increased BrdU labeling in the dentate gyrus of WT mice at both time points. The majority of the surviving cells matured into neurons, indicating a DMI-induced increase in neurogenesis in WT mice. In contrast to the DMI-induced increase in neurogenesis seen in WT mice, the CREB^αΔ mutant mice show no additional changes in BrdU labeling or neurogenesis. Saline-treated CREB^αΔ mutant mice demonstrated comparable numbers of BrdU-positive cells and neurogenesis with that of DMI-treated WT controls (*p < 0.05; n = 6–9). In all groups, the majority of surviving BrdU-positive cells expressed the neuronal marker, NeuN. DMI treatment had no effect on the phenotypic distribution of surviving BrdU-positive cells in WT or CREB^αΔ mutant mice. Error bars indicate SEM.

Figure 2.

Serotonin depletion reduced proliferation in CREB^αΔ mutant mice to WT levels. CREB^αΔ mutant mice show a significant increase in cell proliferation compared with WT mice (p = 0.0494). Treatment with PCPA (250 mg/kg) twice a day for 3 d reverses this increase in cell proliferation in CREB^αΔ mutant mice to WT levels (p = 0.0340), whereas PCPA treatment had no effect on cell proliferation in WT mice (N = 5–6). Error bars indicate SEM. *p < 0.05; ^†p < 0.05.

Figure 3.

Serotonin depletion reversed the antidepressant phenotype of CREB^αΔ mutant mice in the TST. Mice were treated with PCPA twice a day for 3 d. Immobility in the TST was measured 18 h after the last injection. The TST revealed a significant effect of genotype and treatment, with the PCPA-treated CREB^αΔ mutant mice returning to WT levels (n = 4–8). Error bars indicate SEM. *p < 0.05; ^†p < 0.05.

Figure 4.

Serotonin depletion does not have an effect on anxiety in the elevated zero maze. Mice were treated with PCPA and tested in the elevated zero maze 18 h after the last dose. CREB^αΔ mutant mice spent significantly less time in the open arm of the maze, reflecting a proanxiety phenotype. Treatment with PCPA did not alter anxiety levels in either WT or CREB^αΔ mutant mice (n = 8–9). Error bars indicate SEM. *p < 0.05.

Figure 5.

Chronic DMI treatment decreased the latency to consume peanut butter chips in the novel environment in both WT and CREB^αΔ mutant mice. Mean latencies to consume the highly palatable food in the home and novel environment are shown. Mice were treated with DMI (12.5 mg/kg) for 25 d before testing. There was a decreased latency in DMI relative to saline-treated animals (*p < 0.004). There was an overall increased latency in mutant relative to WT animals (⁺p < 0.04) (n = 9–10). Error bars indicate SEM.

Figure 6.

CREB^αΔ mutant mice demonstrated a significant decrease in latency in response to acute DMI treatment, whereas there was no effect of treatment in the WT mice. Mean latencies to consume the highly palatable food in the home and novel environment are shown. Mice (n = 9–10 per group) received a dose of DMI (12.5 mg/kg) in the morning and evening of each of the 3 d of testing. A total of three doses of DMI was received before exposure to the novel environment. There was a decreased latency in DMI relative to saline-treated CREB^αΔ mutant mice (*p < 0.01). There was also an increased latency in mutant relative to WT animals (⁺p < 0.04) (n = 7). Error bars indicate SEM.

Figure 7.

A, CREB^αΔ mutant mice showed significantly reduced time spent in the open arm relative to WT (⁺p < 0.0001) in the elevated zero maze. This reduction was not affected by acute treatment with DMI. Mice (n = 9–10 per group) received a total of three doses of DMI (12.5 mg/kg) before testing, with the same schedule used in the NIH. B, In regard to ethologically relevant parameters, there is a significant effect of genotype on stretch-attend-posture, and head dips (*p < 0.05), but not on rearing. C, CREB^αΔ mutant mice demonstrated an increased latency to enter the open arm (⁺p < 0.05). Error bars indicate SEM.