The present study identified five new β-catenin target genes in thalamic neurons, in addition to previously described Cacna1g. Three of them, Kcna6, Calb2, and Gabra3, were validated by ChIP in vivo and a loss-of-function experiment in cultured neurons, confirming that they might be directly regulated by β-catenin. Two other genes, Cacna2d2 and Kcnh8, also displayed β-catenin-dependence in the latter experiment, although the binding of β-catenin to their regulatory elements was not found. Based on these data, we propose that β-catenin is a regulator of the electrophysiological properties of thalamic neurons in the adult brain.
Numerous genes that we selected in silico as potentially regulated by β-catenin belong to expected functional categories: transcription regulation, cell proliferation, morphogenesis, motility, adhesion, differentiation, and programmed cell death. Similar clusters were observed by others in the genes bound by TCF7L2 in a human colorectal cancer cell line . These results support the well-established role of Wnt/β-catenin in development. Interestingly, the genes involved in neuronal differentiation exhibited the highest enrichment scores in our list. This was consistent with a study that identified β-catenin-LEF1/TCF targets based on a ChIP assay in NIH3T3 cells. Thirty percent of the target genes were implicated in developmental processes, and more than half of the targets from this group were involved in neuronal development . Indeed, Wnt signaling has been particularly implicated in central nervous system development, from early brain patterning to embryonic and adult neurogenesis [30–33, 45–50].
Although our in silico analysis corroborated the involvement of LEF1/TCFs in the regulation of well-known groups of genes, it also identified a group that has not been previously proposed to be a β-catenin-LEF1/TCF target. These were the genes of proteins involved in signal transmission in neurons, including voltage-gated ion channels, neurotransmitter receptors, synaptic vesicle proteins, and synaptic structural proteins. Moreover, we provided experimental evidence of the authentic regulation of some of these genes by β-catenin. The above gene clusters have not yet been identified, probably because the screenings for β-catenin target genes were performed on established cell lines or cancer cells [40, 44, 51–54]. Additionally, studies of hippocampal neurons, in which β-catenin nuclear translocation was observed after NMDA stimulation, did not attempt to identify specific neuronal targets [35–37].
While examining the β-catenin-chromatin association and acetylation of histone H3, we did not observe any relationship between these two phenomena in the analyzed regions with the LEF1/TCF motif. This suggests that the interaction between the β-catenin-LEF1/TCF complex and DNA might not require the open conformation of chromatin. These results may also suggest that the β-catenin complex does not always increase histone acetylation, although it potentially has such an ability [55, 56]. This is consistent with a recent study performed on embryonic stem cells, in which knockdown of Tcf7 and Tcf7l2 did not affect the active chromatin conformation of their targets . We also noticed an interesting pattern of LEF1/TCF motif occurrence in the examined genes. The motifs were usually clustered downstream of the first exon and not in the promoter regions. This suggests that the transcription of these genes may be regulated by LEF1/TCF factors by gene looping, which has been demonstrated for the COX2 and MMP13 genes, in which LEF1/TCF binding sites were located in the 3′ untranslated region .
The neuronal genes with conserved LEF1/TCF motifs that were highly expressed in the thalamus, the regulation of which by β-catenin was confirmed experimentally, encode proteins involved in neuronal excitability. Cav3.1 (encoded by Cacna1g), Cavα2δ2 (Cacna2d2), Kv1.6 (Kcna6), Kv12.1/ELK1 (Kcnh8), and GABAA receptor 3 (Gabra3) are all voltage- or ligand-gated ion channels [59–61]. As such, they underlie the cell membrane conductance of Ca2+, K+, and Cl- (in the case of the GABA receptor) ions and directly propagate, inhibit or modify electric signals [62–64]. Calretinin, in turn, is an intracellular Ca2+-binding protein [65, 66] with diverse functions, including the modulation of intrinsic neuronal excitability . We propose that β-catenin contributes to the proper excitability of thalamic neurons by regulating the expression of the above genes. However, more research is required to determine the real impact of β-catenin and LEF1/TCF factors on the expression of the identified genes and electrophysiology of the thalamus.
The other classes of putative neuronal targets of LEF1/TCF (i.e., the genes that encode structural synaptic proteins, mainly with the PDZ domain, and synaptic vesicle proteins) did not show high expression in the thalamus. However, they still might be regulated by β-catenin and LEF1/TCF factors in some subtypes of neurons or under specific physiological conditions because the regulation of gene expression by β-catenin is very much context-dependent [8, 44, 68, 69]. Particularly interesting would be the exploration of this possibility in future research because a membranous fraction of β-catenin interacts with PDZ proteins in synapses and is implicated in synaptic vesicle localization [70–73]. The role of nuclear β-catenin in the regulation of PDZ and synaptic vesicle protein expression might complement the function of membranous β-catenin in neurons.
We do not yet know whether variations in the nuclear level of β-catenin affect the expression of genes that encode VGCCs and neurotransmitter receptors and shape neuronal excitability in vivo. If so, then we could speculate that the inappropriate activity of β-catenin might affect proper signal transmission in thalamocortical circuits. Thalamocortical desynchronization underlies absence epilepsy , and many anticonvulsant drugs target voltage-gated channels (e.g., T-type Ca2+ channels ). Specifically, the T-type voltage-gated channel Cav3.1 has been proposed to be implicated in absence seizures [75, 76], in addition to the Cavα2δ2 regulatory subunit of voltage-gated channels  and GABAA receptor 3 . Schizophrenia has also been associated with thalamic dysfunction [79–84]. Moreover, some variants of Tcf7l2 have been recently shown to be a risk factor in schizophrenia [23, 85], and a group of synaptic genes involved in excitability has been found to be associated with the risk of schizophrenia . Interestingly, Gabra3-deficient mice display impairments in sensorimotor gating, which is a feature of this disorder . These results suggest a possible role for β-catenin-dependent gene expression in thalamic pathologies, but further in vivo studies are required to elucidate this issue.