Browsing by Subject "microtubules"

Microtubules often adopt non-centrosomal arrays in differentiated tissues, where they are important for providing structure to the cell and maintaining polarity. Although the formation and organization of centrosomal arrays has been well-characterized, little is known about how microtubules form non-centrosomal arrays.

In the mouse epidermis, centrosomes in differentiated cells lose their microtubule-anchoring ability through the loss of proteins from the centrosome. Instead, microtubules are organized around the cell cortex. The cell-cell adhesion protein desmoplakin is required for this organization. Our model is that desmoplakin recruits microtubule-anchoring proteins like ninein to the desmosome, where they subsequently recruit and organize microtubules.

To test this model, we confirmed that the microtubule-binding proteins Lis1, Ndel1, and CLIP170 are recruited by desmoplakin to the cell cortex. Furthermore, by creating an epidermis-specific conditional Lis1 knockout mouse, I found that Lis1 is required for cortical microtubule organization. Surprisingly, however, Lis1 is also required for desmosome stability. This work reveals essential desmosome-associated components that control cortical microtubule organization and unexpected roles for centrosomal proteins in epidermal function.

Although Lis1 is required for microtubule organization, it is not sufficient. I created a culture-based system to determine what other factors may be required for cortical organization for microtubules. My work reveals that stabilization of the microtubules is sufficient to induce their cortical organization. Functionally, cortical microtubules are important for increasing the mechanical integrity of cell sheets by engaging adherens junctions. In turn, tight junction activity is increased. Therefore, I propose that cortical microtubules in the epidermis are important in forming a robust barrier by cooperatively strengthening each cell-cell junction.

To determine whether desmosomes play similar roles in simple epithelia as stratified epithelia, I examined intestinal epithelial-specific conditional desmoplakin conditional knockout mice. Unexpectedly, I found that desmoplakin is not required for cell-cell adhesion and tissue integrity in the small intestine. Furthermore, it does not organize intermediate filaments. Desmoplakin is required, however, for proper microvillus architecture.

Overall, my studies highlight novel tissue-specific roles for desmosomes, in particular desmoplakin, in organizing and integrating different cytoskeletal networks. How desmoplakin's function is regulated in each tissue will be a new interesting area of research.

The exon junction complex (EJC) is comprised of three core components: MAGOH, RBM8A, and EIF4A3. The EJC is canonically known to regulate many aspects of RNA metabolism as well as function in mitosis. Previous work on the EJC has primarily focused on functions for the EJC as a complex, and thus independent roles for EJC components are lacking. It was also recently discovered that EIF4A3 is the causative gene in Richieri-Costa-Pereira Syndrome (RCPS), a craniofacial disease primarily characterized by a severely undersized mandible.

We used two systems to examine EIF4A3 function. First, HeLa cells allowed for dissection of EJC complex requirements. We depleted EIF4A3, MAGOH, or RBM8A and saw that MAGOH and RBM8A protein levels are interdependent, while EIF4A3 levels are independent. We next used point mutant constructs that disrupt EJC core formation to assay EJC complex requirements during mitosis. Constructs that disrupt MAGOH-RBM8A from interacting with EIF4A3 were able to rescue prometaphase arrest, suggesting they may regulate mitosis independently. Further, localization studies show that during mitosis MAGOH and RBM8A localize pericentrosomally whereas EIF4A3 is more expanded across microtubules. Biochemistry studies reveal that EIF4A3 is able to bind to microtubules in the absence of other EJC components or RNA. We also found that overexpression of EIF4A3 results in telophase arrest, suggesting that EIF4A3 dosage is important throughout mitosis.

We next used mouse models to examine the developmental requirements of Eif4a3 both ubiquitously and in the neural crest. We show that heterozygous loss of Eif4a3 at early embryonic ages results in disrupted mandibular arch fusion. These defects later manifest as severe craniofacial abnormalities and loss of adult mandibular structures. Examination of the skeletons of these embryos shows premature ossification of the clavicle. Parallel studies in patient-derived iPSCs show that neural crest cells are less able to migrate and when pushed down an osteogenic lineage, they prematurely differentiate into bone. The craniofacial phenotypes seen in Eif4a3 mutant mice are also distinct from other EJC mutants.

From these data we conclude that EIF4A3 has EJC-independent functions in mitosis, microtubule interaction, and neural crest development. Future studies that disentangle EJC-dependent and independent functions will allow for a more thorough understanding of how these proteins work at the molecular level and in human disease.