A polymorphism associated with depressive disorders differentially regulates brain derived neurotrophic factor promoter IV activity

Abstract

Background: Changes in brain derived neurotrophic factor (BDNF) expression have been associated with mood disorders and cognitive dysfunction. Transgenic models that overexpress or underexpress BDNF demonstrate similar deficits in cognition and mood. We explored the hypothesis that BDNF expression is controlled by balancing the activity of BDNF promoter IV (BP4) with a negative regulatory region containing a polymorphism associated with cognitive dysfunction and mood disorders.

Methods: We used comparative genomics, transgenic mouse production, and magnetofection of primary neurons with luciferase reporters and signal transduction agonist treatments to identify novel polymorphic cis-regulatory regions that control BP4 activity.

Results: We show that BP4 is active in the hippocampus, the cortex, and the amygdala and responds strongly to stimuli such as potassium chloride, lithium chloride, and protein kinase C agonists. We also identified a highly conserved sequence 21 kilobase 5' of BP4 that we called BE5.2, which contains rs12273363, a polymorphism associated with decreased BDNF expression, mood disorders, and cognitive decline. BE5.2 modulated the ability of BP4 to respond to different stimuli. Intriguingly, the rarer disease associated allele, BE5.2(C), acted as a significantly stronger repressor of BP4 activity than the more common BE5.2(T) allele.

Conclusions: This study shows that the C allele of rs12273363, which is associated with mood disorder, modulates BP4 activity in an allele-specific manner following cell depolarization or the combined activity of protein kinase A and protein kinase C pathways. The relevance of these findings to the role of BDNF misexpression in mood disorders and cognitive decline is discussed.

Figure 1

Brain derived neurotrophic factor promoter 4 can support LacZ expression in cortex, hippocampus, and amygdala. (A–C) Photomicrographs of 60-μm sections of brain derived neurotrophic factor promoter 4-LacZ neonatal (postnatal days 3–7) transgenic mouse brains following staining by X-Gal showing LacZ expression (blue cells) in the amygdala (A), hippocampus (B), and prefrontal cortex (C). Scale bars represent 500 μm (A), 1000 μm (B), and 880 μm (C), respectively. BLA, basolateral amygdala; CA, cornu ammonis; CeA, central amygdala; Hippo, hippocampus; PC, piriform cortex; PFC, prefrontal cortex.

Figure 2

Diagrams representing luciferase constructs used for the investigation and comparative genomic analysis to identify BE5.2 and allelic frequency of rs12273363 in different human populations. (A) Diagrammatic representation of the different luciferase constructs used to investigate the effects of different alleles of BE5.2 and on human brain derived neurotrophic factor promoter 4 (BP4) activity (not to scale). (B) Stacked pairwise alignment from the evolutionary conserved region genome browser between genomic sequences of different animal species (in descending order; chicken, possum, mouse, dog, and rhesus macaque) against the human genomic sequence. Brain derived neurotrophic factor promoter 4 and BE5.2 are indicated by a black box and the direction of transcription from BP4 is indicated by a bent black arrow. The single nucleotide polymorphism rs12273363 is found in BE5.2. Yellow, pink, green, blue, and red peaks highlight areas of conservation in untranslated regions, introns, repetitive regions, coding regions, and intergenic DNA, respectively (from evolutionary conserved region genome browser). The position of rs12273363 is indicated below BE5.2. (C) Allelic frequency of rs12273363 in different human populations from the National Center for Biotechnology Information single nucleotide polymorphism database. BDNF, brain derived neurotrophic factor; CEU, Central Europeans; Chrom., chromosome; Cnt., count; HCB, Hans Chinese in Bejing; JPT, Japanese in Tokyo; YRI, Yoruba in Ibadan.

Figure 3

Brain derived neurotrophic factor promoter 4 (BP4) and brain derived neurotrophic factor activity in primary neuronal cultures following potassium chloride (KCl) depolarization. Primary hippocampal (A, D) cortical (B, E), and amygdala (C, F) cultures were co-transfected with BP4 and renilla plasmid. Cells were either untreated (A–C) or treated (D–F) with 60 mmol/L KCl for 24 hours before harvesting for analysis by dual luciferase assay. Relative light units were calculated by normalizing luciferase signal to renilla signal. n ≥ 3, **p < .01.

Figure 4

Brain derived neurotrophic factor promoter 4 (BP4) and BP4-BE5.2(T) response to potassium chloride (KCl) depolarization in primary neuronal cultures. Primary hippocampal (A, D, E), cortical (B, E, H), and amygdala (C, F, I) cultures were co-transfected with BP4 or BP4-BE5.2(T) and renilla plasmid. Cells were either untreated (A–C) or treated (D–I) with 60 mmol/L KCl for 24 hours before harvesting for analysis by dual luciferase assay. Relative light units were calculated by normalizing luciferase signal to renilla signal. n ≥ 3, **p < .01, *p < .05, not significant; ^#two-way analysis of variance p > .05 but t test p < .01.

Figure 5

Comparison of BE5.2(T) and BE5.2(C) interaction with brain derived neurotrophic factor promoter 4 (BP4) in potassium chloride (KCl) depolarized hippocampal, cortical, and amygdala cultures. Primary hippocampal (A, D), cortical (B, E), and amygdala (C, F) cultures were co-transfected with either BP4-BE5.2(T) or BP4-BE5.2(C) with renilla plasmid. Cells were untreated (A–C) or treated (D–I) with 60 mmol/L KCl for 24 hours before harvest for dual luciferase assay. Relative light units were calculated by normalizing luciferase signal to renilla signal. Fold induction was calculated by dividing relative light units of treated culture by relative light units of vehicle-treated cultures. n ≥ 3, **p < .01, *p < .05. n.s., not significant.

Figure 6

Interaction of BE5.2 (C and T) with brain derived neurotrophic factor promoter 4 (BP4) in response to lithium chloride (LiCl), phorbol-12-myristate-13-acetate (PMA), and forskolin in primary cortical cultures. Primary cortical cultures were co-transfected with either BP4 (A, C, D) or BP4-BE5.2(T), BP4-BE5.2(C), or ΔBP4 (B, D, E) with renilla plasmid. Cells were treated with 1 mmol/L LiCl for 72 hours (A, B), 150 nmol/L PMA for 24 hours (C, D), or forskolin for 24 hours (E, F) before harvest for dual luciferase assay. Relative light units were calculated by normalizing luciferase signal to renilla signal. n ≥ 3, **p < .01. DMSO, dimethyl sulfoxide; n.s., not significant.

Figure 7

Combined treatment of phorbol-12-myristate-13-acetate (PMA) and forskolin increased luciferase expression by brain derived neurotrophic factor promoter 4 (BP4) BE5.2 (T) and BP4 BE5.2 (C) in cortical cultures. Primary cortical cultures were co-transfected with either (A) BP4 or (B, C) BP4 BE5.2(T) and BP4 BE5.2(C). Cells were treated with 150 nmol/L PMA and 25 μmol/L forskolin or dimethyl sulfoxide (DMSO) as vehicle treatment for 24 hours before harvest for dual luciferase assay. Relative light units were calculated by normalizing luciferase signal to renilla signal. Fold induction was calculated by dividing relative light units of treated culture by relative light units of vehicle-treated cultures. n ≥ 3, **p < .01, *p < .05, and ^#two-way analysis of variance p > .05 but t test p < .05.