Lithium-induced neuroprotection is associated with epigenetic modification of specific BDNF gene promoter and altered expression of apoptotic-regulatory proteins

Abstract

Bipolar disorder (BD), one of the most debilitating mental disorders, is associated with increased morbidity and mortality. Lithium is the first line of treatment option for BD and is often used for maintenance therapy. Recently, the neuroprotective action of lithium has gained tremendous attention, given that BD is associated with structural and functional abnormalities of the brain. However, the precise molecular mechanism by which lithium exerts its neuroprotective action is not clearly understood. In hippocampal neurons, the effects of lithium (1 and 2 mM) on neuronal viability against glutamate-induced cytotoxicity, dendritic length and number, and expression and methylation of BDNF promoter exons and expression of apoptotic regulatory genes were studied. In rat hippocampal neurons, lithium not only increased dendritic length and number, but also neuronal viability against glutamate-induced cytotoxicity. While lithium increased the expression of BDNF as well as genes associated with neuroprotection such as Bcl2 and Bcl-XL, it decreased the expression of pro-apoptotic genes Bax, Bad, and caspases 3. Interestingly, lithium activated transcription of specific exon IV to induce BDNF gene expression. This was accompanied by hypomethylation of BDNF exon IV promoter. This study delineates mechanisms by which lithium mediates its effects in protecting neurons.

Keywords: BDNF; apoptotic regulatory genes; bipolar; hippocampal culture; lithium; methylation.

Figure 1

Pretreatment of lithium prevents neuronal death by glutamate. Lithium (1 or 2 mM) was added to the medium 36 h prior to glutamate treatment. Glutamate (100 μM) was added 12 h prior to examining neuronal viability. The data are mean ± SD from 3 independent experiments. ^*P < 0.001 compared with control group and ^**p < 0.001 compared with glutamate treated group.

Figure 2

Neuronal morphology after glutamate (100 μM) and glutamate + lithium (1 or 2 mM) treatment. Treatment of glutamate and lithium was the same as mentioned in Figure 1.

Figure 3

Dendritic length after glutamate (100 μM) and glutamate + lithium (1 or 2 mM) treatment. Ten photographs per well were obtained from three wells for each group. Thirty well-stained neurons that made connections to no more than two cells were selected for measurements of their primary dendritic length and dendritic number. ^*P < 0.001 compared with control group and ^**p < 0.001 compared with glutamate treated group.

Figure 4

Dendritic number after glutamate (100 μM) and glutamate + lithium (1 or 2 mM) treatment. The data are mean ± SD from 3 independent experiments. ^*P < 0.001 compared with control group and ^**p < 0.001 compared with glutamate treated group.

Figure 5

(A) BDNF mRNA expression after 48 h of lithium (1 or 2 mM) treatment. ^*P < 0.001 compared with control group. (B) Western blots showing immunolabeling of BDNF and β-actin in hippocampal neurons after 1 or 2 mM lithium treatment. (C) Bar diagram showing effect of lithium on protein expression of BDNF. The results are presented as percent of control. The data are mean ± SD from 3 independent experiments. ^*P < 0.001 compared with saline treated group.

Figure 6

Effect of lithium (1 or 2 mM) on expression of individual BDNF exons after 48 h after lithium (1 or 2 mM) treatment. The data are mean ± SD from 3 independent experiments. ^*P < 0.001 compared with control group.

Figure 7

DNA methylation of BDNF exons after 48 h of lithium (1 or 2 mM) treatment. The data are mean ± SD from 3 independent experiments. ^*P < 0.001 compared with control group.

Figure 8

Effects of 48 h of lithium (1 or 2 mM) on the expression of anti-apoptotic (A) and pro-apoptotic (B) genes. The data are mean ± SD from 3 independent experiments. ^*P < 0.001 compared with control group.