19 July 2010. The protein encoded by DISC1, the disrupted in schizophrenia 1 gene, partners with the Wnt signaling protein Dixdc1 in mice according to a study published on July 15 in Neuron. This finding adds a new player to the roster of proteins known to associate with DISC1, a major suspect in schizophrenia and other psychiatric disorders. Depending on the complex formed, this interaction can promote either neural proliferation or migration during brain development.
The study, from Li-Huei Tsai's lab at Massachusetts Institute of Technology, Cambridge, builds on earlier work connecting DISC1 to neural proliferation (see Mao et al., 2009 and SRF related news story) through Wnt signaling. The Wnt pathway controls cell fate and can trigger cancer when inappropriately activated. DISC1 directly inhibits glycogen synthase kinase 3β (GSK3β), which keeps levels of the transcription activator β-catenin high, spurring the birth of neurons.
The new study finds that Dixdc1's role in neural proliferation is so similar to that of DISC1 that each protein can stand in for the other. This means that DISC1 and Dixdc1 (short for "DIX domain-containing-1") may offer redundant ways of ensuring proper neural proliferation, and suggests that schizophrenia and other psychiatric diseases may disrupt the Wnt pathway. Consistent with this, lithium, a drug commonly used to treat mood disorders and to augment antipsychotic medication for schizophrenia, indirectly inhibits GSK3β (Beaulieu et al., 2008 and see SRF related news story).
For neural migration, however, the results were different. Dixdc1 could not correct migration defects induced by inhibiting DISC1. Instead, the researchers found evidence for a three-part complex among DISC1, Dixdc1, and another schizophrenia suspect, the protein called nuclear distribution factor E homolog like-1 (Ndel1), which regulated neuron migration without relying on the Wnt pathway. Overall, the study reveals a complex dance between DISC1 and Dixdc1 in regulating neuron birth and migration. It reinforces the idea that perturbations of brain development could lead to the disorganized brain circuitry suspected in schizophrenia and other psychiatric disorders.
A neuron is born
Interested in how DISC1 is regulated during brain development, first author Karun Singh and colleagues began by looking for proteins that bind DISC1 early on when new neurons are proliferating. Co-immunoprecipitation of embryonic mouse brain tissue taken on embryonic day E14 showed that DISC1 bound to Dixdc1. Other experiments showed that Dixdc1 was highly expressed throughout development, and was found in both neural progenitor cells and in neurons.
Having tied Dixdc1 to the right place at the right time, the researchers turned to manipulating the protein’s levels in vivo to probe its role in brain development. The results mirrored their previous DISC1 findings in many ways. Dixdc1 knockdown, in which a Dixdc1-inhibiting shRNA was introduced into the brains of embryonic mice via in utero electroporation, disrupted the normal distribution of neurons in the brain by prematurely halting the proliferation of neural progenitors. This result looked similar to the 2009 DISC1 knockdown experiments by Tsai and colleagues.
Singh and colleagues also found that, like DISC1, Dixdc1 worked through the Wnt pathway. Knockdown of Dixdc1 reduced the pathway’s activity in vitro, as reported by a construct sensitive to β-catenin levels in the cells. Overexpression of Dixdc1 boosted this activity, and the sizes of these effects were similar to those induced by DISC1 manipulations. Further experiments showed synergy between DISC1 and Dixdc1: for example, downregulating both lowered Wnt pathway activity more than downregulating either protein alone. These effects seemed to require an interaction between the two proteins, and other experiments placed Dixdc1, like DISC1, upstream of GSK3β and β-catenin.
Dixdc1 and DISC1 acted so similarly that the researchers tested whether one could compensate for the other. Strikingly, the low levels of Wnt activity induced by knockdown of DISC1 could be fully rescued by overexpression of Dixdc1 and vice versa.
The mechanics of this partnership held in vivo, with Dixdc1 activating the Wnt pathway in neural progenitors and compensating for a loss of DISC1. This translated into real results for developing neurons in the brain: knocking down Dixdc1 and DISC1 together led to fewer proliferating cells than inhibiting either protein alone, and overexpression of one protein could fully rescue neural proliferation that had been compromised by knockdown of the other.
A migration trio
The researchers then asked whether Dixdc1 would continue to mimic DISC1 in the realm of neural migration, which begins after neural proliferation. When they downregulated Dixdc1 in mouse brains at a time when new neurons migrate to their ultimate locations in the brain, Singh and colleagues found a striking migration defect four days later: cells were stuck in the intermediate zone of the developing cortex, rather than making their way to the cortical plate. Further experiments indicated that this was an effect on migration, rather than changes in cell fate.
But here's where a different kind of DISC1-Dixdc1 partnership emerges. Adding extra DISC1 could not compensate for these migration deficits, nor could extra Dixdc1 rescue the migration problems induced by DISC1 knockdown. The Wnt pathway did not seem to be involved either, because throwing it into overdrive with degradation-resistant β-catenin did not reverse the migration deficits induced by either DISC1 or Dixdc1 knockdown.
To identify the pathway involved, the researchers turned to Ndel1, which binds DISC1 to regulate neuron migration and has itself been associated with schizophrenia risk (Kamiya et al., 2005). They found that Dixdc1 also bound Ndel1, and that Dixdc1 can form a three-part complex with DISC1 and Ndel1. Interestingly, Ndel1 bound to the same Dixdc1 region as DISC1 does, suggesting that Dixdc1 may facilitate the interaction between DISC1 and Ndel1.
The researchers found that formation of the Dixdc1-DISC1-Ndel1 complex depended on phosphorylation of Dixdc1. Cyclin-dependent kinase 5 (Cdk5), also involved in migration, phosphorylated Dixdc1 in the same region where Ndel1 and DISC1 bind. Using a mutant Dixdc1 construct that could not be phosphorylated at this site, the researchers found reduced Ndel1, but not DISC1, binding to Dixdc1. Introducing this mutant Dixdc1 into embryonic mouse brains disrupted migration, indicating that phosphorylation of Dixdc1 at this site was crucial. Likewise, inhibiting the formation of the Dixdc1-DISC1-Ndel1 complex by using a peptide fragment that interfered with DISC1 and Ndel1 binding to Dixdc1 also stalled migration in vivo.
An intriguing idea
The similarity of Dixdc1 to DISC1, particularly in neuron proliferation, might help explain why some members of the Scottish DISC1 family who carry the DISC1 translocation do not have a psychiatric diagnosis (Blackwood et al., 2001). Singh and colleagues speculate that Dixdc1 may be compensating for a disrupted DISC1 in these people and that, therefore, certain single nucleotide polymorphisms (SNPs) in the Dixdc1 gene may well segregate with disease. This provides a concrete candidate for those murky compensatory factors that are often invoked to explain different outcomes in people who harbor the same mutation.
Together, these findings emphasize how multiple molecules team up to govern brain development. Teasing out how each molecule works, whether it acts alone or in partnership with others, is essential to understanding how problems with any one molecule may wreak havoc with brain development in ways that lead to psychiatric disease.—Michele Solis.
Singh KK, Ge X, Mao Y, Drane L, Meletis K, Samuels BA, Tsai LH. Dixdc1 is a critical regulator of DISC1 and embryonic cortical development. Neuron. 2010 July 15; 67: 33-48. Abstract