A Physical Mechanism for Apical Protein Sorting

Abstract

Despite decades of research, the apical sorting of epithelial membrane proteins remains incompletely understood. We noted that apical cytoplasmic domains are smaller than those of basolateral proteins; however, the reason for this discrepancy is unknown. We used a synthetic biology approach to investigate whether a size barrier at the Golgi apparatus might hinder apical sorting of proteins with large cytoplasmic tails. We focused on Crb3, Ace2, and Muc1 as representative apical proteins with short cytoplasmic tails. By incorporating a streptavidin-binding peptide, these proteins can be trapped in the endoplasmic reticulum (ER) until addition of biotin, which triggers synchronous release to the Golgi and subsequent transport to the apical cortex. Strikingly, increasing the size of their cytoplasmic domains caused partial mislocalization to the basolateral cortex and significantly delayed Golgi departure. Moreover, N-glycosylation of “large” Crb3 was delayed, and “small” Crb3 segregated into spatially distinct Golgi regions. Biologically, Crb3 forms a complex through its cytoplasmic tail with the Pals1 protein, which could also delay departure, but although associated at the ER and Golgi, Pals1 disassociated prior to Crb3 departure. Notably, a non-dissociable mutant Pals1 hampered exit of Crb3. We conclude that, unexpectedly, a size filter at the Golgi facilitates apical sorting of proteins with small cytoplasmic domains and that timely release of Pals1, to reduce cytoplasmic domain size, is essential for the normal kinetics of apical protein sorting. Additionally, I laid out a synthetic biology framework for performing a pooled genome-wide CRISPR/Cas9 knock-out screen to identify genes which are required sorting and delivery of apical proteins to the appropriate cell surface.