A key to developing new, more specific therapies, is to understand the molecular mechanisms that drive the disease characteristics. This can be exceedingly difficult because a particular disease may result from changes at any point in a complex signaling pathway. Diseases affecting brain function compound this challenge because there is no convenient way to access and study the affected human tissue. These challenges are well exemplified by malformations of the cerebral cortex (MCC). The cerebral cortex plays a key role in memory, perception, awareness, thought, language, and consciousness. Incorrect development of this tissue results in epilepsy, intellectual disability, and cognitive disorders. Defects to the cerebral cortex present similar symptoms in patients, but the molecular causes are varied and still largely unknown. Genetic or heritable cases of MCC allowed researchers to screen for mutations being passed through families and were the first indications of genetic causes of many human MCCs. New research performed in the labs of Dr. Valera Vasioukhin and Dr. Jonathan Cooper demonstrated that the correct development of the cerebral cortex requires robust cell-cell adhesion between embryonic neural stem cells. The findings published in Developmental Cell focused on the protein Lethal giant larvae (Llgl1), which is well known to regulate cell polarity, but here was also shown to directly aid the formation of cell-cell adhesion junctions.
Image provided by Dr. Valera Vasioukhin
As Llgl1’s name implies, complete deletion of the gene is lethal in genetic models like fruit flies and mice so that no offspring survives after birth. The brains of embryos lacking Llgl1 have significant defects, including abnormal brain structures, which gave the indication that among other activities Llgl1 may be a player in some MCC. For this study researchers specifically deleted this gene in developing embryonic neural stem cells of mice using the Nestin promoter which drives Cre recombinase expression in the developing brain. This approach yielded mice that were prone to seizures, and while the Llgl1 conditional knock out mice developed brains of normal size the brain structures were significantly different from their normal siblings with intact Llgl1. In fact, the histology of Llgl1 deficient brains was similar to a specific type of human MCC called Periventricular Heterotopia (PH). In these mice and in patients with PH a groups of neurons are abnormally deposited between the lateral ventricles and the white matter, forming an extra layer of gray matter that doesn’t exist in wild type mice. These experiments reveal that the loss of cell-cell adhesion structures between embryonic neural stem cells in Llgl1 mutant mice is responsible for mislocalization of these cells and subsequent misplacement of neurons generated by stem cells.
A thin layer of neural stem cells normally maintains its structure by forming apical ‘tight’ junctions that link the cytoskeletons of each cell to one another through the plasma membrane and forming a physical barrier. Time-lapse imaging of embryonic brain slices revealed that disruption of the apical junctions was common in conditional knockout mice. The loss of this barrier allowed neurons to be internalized into the ventricular space and caused the excessive malformations observed in young mice.
This work was a useful description of how loss of Lgl1 causes brain defects similar to PH; however, Dr. Vasioukhin and colleagues went further to understand how the loss of cell polarity protein LLGL1 was altering cell-cell junctions. N-cadherin, β-catenin, and α-catenin are essential components of apical junction complexes and thus likely players in the observed phenomena. While the synthesis and degradation of these proteins was unaffected by Llgl1 deletion, N-cadherin levels at the cell surface were elevated in Llgl1 mutants, due to decreased rates of internalization. For tight junctions to form correctly, components must be continually internalized from non-appropriate locations and concentrated at the apical side of the cell. This finding was the first evidence that the cell polarity machinery and apical junctions are connected in these cells. Researchers further demonstrated a direct interaction between Llgl1 and N-cadherin using co-immunoprecipitation experiments. Co-immunoprecipitation experiments using a series of protein fragments also revealed a ~30 amino acid region in N-cadherin was responsible for binding a ~50 amino acid region of Llgl1. Finally this interaction was demonstrated in mice. Embryos expressing an shRNA targeting Llgl1 presented similar brain defects as the conditional knock out, and this could be rescued by expressing a non-endogenous copy of Llgl1; however, if the N-cadherin interacting region was deleted from Llgl1 it failed to rescue the defects. Moreover, if only the N-cadherin interacting domain of Llgl1 was expressed at high levels in embryos PH-like structures formed – as if Llgl1 had been deleted. This was likely because the fragment prevented endogenous, fully functional Llgl1 from binding N-cadherin.
Researchers are far from fully understanding this exciting link between cell-cell adhesion and cell polarity. For example, there were indications that Llgl1 loss would cause some sort of defect in the cerebral cortex, but why it resembles PH is unclear. As Dr. Vasioukhin explained, “The most frequently mutated gene in human Periventricular Heterotopia is Filamin A, not Llgl1. Filamin A has been implicated in cell-cell adhesion formation. Presently, it is not known whether LLGL1 is mutant in some of the patients with PH and future research will address this important question.”. As with most good science, the answers from this study raise many new questions.
Jossin Y, Lee M, Klezovitch O, Kon E, Cossard A, Lien W-H, Fernandez TE, Cooper JA, Vasioukhin V. 2017. Llgl1 Connects Cell Polarity with Cell-Cell Adhesion in Embryonic Neural Stem Cells. Dev Cell. 41, 481-495.
Funding for this research was provided by the National Institutes of Health and Fred Hutchinson Cancer Research Center bridge funds.