20 September 2010. In the world of disrupted-in-schizophrenia 1 (DISC1) biology, things are getting curiouser and curiouser. According to a study published in the Journal of Neuroscience on August 4, DISC1 protein, a key suspect in schizophrenia, interacts with amyloid precursor protein (APP), the darling of Alzheimer disease research. The study from Dennis Selkoe's lab at Harvard University shows that the two proteins team up to direct migration of newborn neurons in the developing brain.
The new work places DISC1 downstream of APP, and shows that APP is required for DISC1 to localize to the centrosome, a microtubule organizing center which DISC1 must reach for the cell to migrate correctly. "The idea is that APP at the cell surface may act as kind of a scaffolding factor where it binds DISC1, and another protein called Disabled-1," says first author Tracy Young-Pearse, who has just started her own research lab at Harvard.
There have been whiffs of a connection between the molecular players of Alzheimer disease and schizophrenia before, mainly involving β-site amyloid precursor protein cleaving enzyme 1 (BACE1). In addition to helping turn APP into amyloid-β protein—the main component of the amyloid plaques indicative of Alzheimer's—BACE1 also cleaves neuregulin-1 (NRG1; see SZGene entry), a top suspect in schizophrenia. One study described schizophrenia-like behaviors in an Alzheimer's mouse model missing BACE1 (see SRF related news story), and another found that BACE1-induced changes in NRG1 levels could regulate an isoform of DISC1 (see SRF related news story). Other intriguing tidbits include a link between γ-secretase, another APP processing enzyme, and NRG1-Erb4 signaling (see SRF related news story); binding between DISC1 and APLP, a protein similar to APP (Millar et al., 2003); and an association between a single nucleotide polymorphism in DISC1 and late-onset Alzheimer disease (Beecham et al., 2009).
By exploring how APP normally works, rather than on what goes wrong with it in Alzheimer disease, Young-Pearse first discovered a role for APP in cell migration while a postdoc in Selkoe's lab (Young-Pearse et al., 2007). Despite massive amounts of research into how APP is cleaved and processed to form the amyloid-β protein, not many people had bothered to study what APP normally does, she says.
She found that knocking down APP levels in cortical precursor cells resulted in neurons that did not migrate correctly: instead of progressing up into the cortical plate where they belong, they got stuck in the lower intermediate zone.
While finishing up this initial study, a talk by Akira Sawa alerted Young-Pearse to a potential connection between DISC1 and APP: his slides showing migration defects when DISC1 is knocked down (see SRF related news story) looked remarkably like her own APP knockdown data, she says.
Despite the connection she turns up between APP and DISC1 in the new study, Young-Pearse is careful to draw a distinction between APP's role in migration, and its involvement in the manufacture of amyloid-β protein. "We don’t think this role of APP in migration is necessarily linked to Alzheimer disease," she says. "I'm more interested in what APP and migration could be doing in schizophrenia."
DISC1 to the rescue
Young-Pearse and colleagues began by replicating Sawa's DISC1 results. In utero electroporation of shRNA that inhibits DISC1 into cortical precursor cells of rat embryos led to a clear migration defect, with new neurons unable to make it into the cortical plate. This defect could be rescued by overexpressing full-length DISC1, and further experiments established the critical DISC1 regions: the C-terminal half of the protein, containing both the self-association binding site and the NDEL1 binding site for normal migration. Notably, a truncated version of DISC1 corresponding to the human translocation breakpoint led to the abnormal migration pattern.
Because of the similar migration defects in DISC1 and APP knockdown experiments, the researchers next asked if one protein could compensate for the other. Co-electroporating the DISC1 shRNA along with APP did not rescue the migration defect. But strikingly, adding DISC1 could rescue APP knockdown. While APP knockdown limits migration to the cortical plate to only about 10 percent of electroporated cells, with extra DISC1 on board, nearly 40 percent of electroporated cells made it. The C-terminal portion of DISC1 containing the self-association site and the NDEL1 site could also rescue APP knockdown, getting over 60 percent of electroporated cells into the cortical plate; the other, N-terminal half could not. Because DISC1 can spur normal migration without APP around, Young-Pearse and colleagues conclude that it must be acting downstream of APP.
This functional interaction stems from a physical one, as shown by a series of co-immunoprecipitation experiments. DISC1 directly binds to APP, and this interaction takes place between the cytoplasmic side of APP and the N-terminal half of DISC1. APP spans the membrane, with a large portion sticking out of the cell, and its smaller DISC-binding portion inside of the cell. Though DISC1 localizes to different regions within a cell, it did co-immunoprecipitate with APP in isolated membrane fractions of the cell—as expected if DISC1 transiently interacts with APP.
Centered on the centrosome
To begin to understand how the APP-DISC1 interaction mucks up migration, the researchers tracked what happened to DISC1 inside the cell when APP was knocked down. DISC1 levels were not changed, but where it ended up within the cell did. Normally DISC1 staining forms a tight knot within a neuron, and it colocalizes with other centrosome markers. When the researchers knocked APP down, however, this seemed to set DISC1 adrift, leaving DISC1 staining diffusely spread throughout the cell.
The researchers offer a working model for their results that casts APP as a kind of meeting place for DISC1 and its binding partners that are critical for migration—lissencephaly protein 1 (LIS1) and nuclear distribution factor E homolog like-1 (NDEL1) (see SRF related news story; SRF news story). The DISC1-LIS1-NDEL1 complex then leaves APP and travels to the centrosome, enabling migration. With APP missing, DISC1 drifts aimlessly throughout the cell, unable to connect with its partners. Adding extra DISC1 rescues this situation because flooding the cell with DISC1 increases the chances that it will bump into its binding partners. This idea explains the odd combination of results showing that the N-terminal portion of DISC1 binds to APP, but the C-terminal portion of DISC1 can rescue APP knockdown.
So is the schizophrenia research community ready to welcome APP into its fold of molecules? Though it's unclear whether there are APP gene variants associated with schizophrenia (Forsell et al., 1995), this APP-DISC1 connection helps to elucidate DISC1's role in migration. Understanding this is arguably one of the best avenues available for getting at the perturbations during brain development that predispose to schizophrenia.—Michele Solis.
Young-Pearse TL, Suth S, Luth ES, Sawa A, Selkoe DJ. Biochemical and functional interaction of disrupted-in-schizophrenia 1 and amyloid precursor protein regulates neuronal migration during mammalian cortical development. J Neurosci. 2010 Aug 4; 30:10431-10440. Abstract