1) Molecular basis of the actomyosin-driven membrane remodeling during regulated secretion at the plasma membrane. Secretory epithelia in exocrine glands such as the salivary glands (SGs) represent an ideal model system to study the remodeling of membranes during regulated exocytosis. The major secretory units of the SGs are the acini that are formed by polarized cells in which the apical plasma membrane (APM) forms small canaliculi where proteins and water are released. Proteins destined to secretion are packed in secretory granules at the trans-Golgi network (TGN) and transported to the cell periphery where they fuse with the APM upon stimulation of GPCRs.
Our aim i s to elucidate the molecular machinery regulating the fusion and integration of the secretory granules with the APM and the maintenance of APM homeostasis. To this end, we developed an experimental system in live rodents aimed at imaging and tracking individual secretory granules, and visualizing the dynamics of the APM. We established that upon stimulation of the beta-adrenergic receptor, the granules fuse with the APM, releasing their content into the lumen of the canaliculi. In addition, we found that a complex composed of F-actin and two isoforms of non-muscle myosin II (NMIIA and NMIIB) is recruited onto the granules after the fusion step. We showed that the actomyosin contractile activity regulates the integration of the granular membranes into the APM and the completion of exocytosis. In the last year we focused on elucidating two aspects of the regulation of the actomyosin complex during regulated exocytosis, and specifically: the machinery regulating F-actin assembly onto the granules, and the mechanisms of recruitment and regulation of NMII. F-actin is assembled around the secretory granules after their fusion with the APM, and plays three distinct roles: first, stabilizes the granular membranes, second, prevents compound exocytosis, and finally, facilitates their gradual collapse. We have hypothesized that the F-actin scaffold is progressively assembled around the granules in two steps: the first requires the formation of linear filaments and provides the stabilization of the membranes, and the second requires the formation of branched filaments, and regulates the collapse of the granules. Consistent with this hypothesis, we found that mDia1 and mDia2, two components of the machinery initiating the formation of linear filaments, are recruited onto the secretory granules right after fusion. Impairment of the activity of mDia1/mDia2 by either pharmacological approaches or using selected conditional knock-out mice resulted in the expansion of the granules after fusion with the APM. In addition, we found that the actin branching factors Arp2/3, cortactin, WASP, and Wave2 are recruited after the F-actin linear filaments are assembled onto the granules. Impairing the activity of the Arp2/3 complex by a specific pharmacological inhibitor, significantly delays the integration of the granules into the APM. We showed that this is a universal mechanism that is conserved across species, as similar results were obtained in the SGs of fruit flies. We have also shown that NMIIA and NMIIB are recruited onto the secretory granules after their fusion with the APM, and by using a pharmacological approach we have determined that their contractile activity drives the gradual integration of the granules into the APM. This contrasts with other cellular processes where actomyosin-based contractions employ only one isoform of NMII. Notably, by using conditional knock-out mice we determined that NMIIB is required to control the initial steps of the integration of the granular membrane, by stabilizing the F-actin scaffold and providing a continuous contractile activity that pushes the membranes towards the APM. On the other hand, NMIIA is required at later stages of the process to control the expansion of the fusion pore. Since both NMIIA and NMIIB are recruited after the formation of the F-actin scaffold, we had assumed that this process would be mediated by their well-characterized actin-binding site. Unexpectedly, we found that both NMII isoforms are recruited in an actin-independent fashion, and that the main role of F-actin is to facilitate the proper assembly of the myosin filaments. We determined that this step is regulated by the phosphorylation of the regulatory chain of NMII that is catalyzed by the myosin light chain kinase (MLCK), which is also recruited onto the granules. 2) Bioenergetics of membrane remodeling The remodeling of the membranes of the secretory granules and the APM is an energetically unfavorable process. A single exocytic event consumes significant energy to bring together and fuse a single secretory granule and the plasma membrane. We found that 150-200 secretory granules undergo exocytosis after beta-adrenergic stimulation of an acinar cell and that each granule is retrieved by 50-75 endocytic vesicles. Therefore, a central question is: how is cell metabolism regulated during remodeling? Our hypothesis is that mitochondrial function is increased during beta-adrenergic stimulation and is temporally and spatially tightly coupled to the exo-endocytic events. We have developed a method to follow the dynamics of the mitochondria and their metabolic activity in the SGs of live rodents. Since there are no reliable tools to quantitatively image the levels of cellular ATP in vivo, we used two-photon microscopy to determine NADH levels (the main substrate of the electron transport chain, ETC) and mitochondrial potential. We discovered that mitochondrial metabolism undergoes spontaneous oscillations in SGs under basal conditions (period: 10-15 sec). This finding contrasts with what reported in vitro where transient metabolic oscillations were observed only after agonist stimulation. Moreover, these oscillations are regulated by reactive oxygen species but are insensitive to manipulation of intracellular Ca2+, indicating a completely novel regulation of this process. Most notably, we found that mitochondrial oscillations are highly coordinated throughout the SG epithelium via the activity of gap junctions, which may transport a not yet identified small molecule that regulates the synchronization of the oscillations. Our work has also shown that in the acinar cells of the SGs vivo mitochondria are organized in two distinct populations: one localized beneath the basolateral membrane, the other, localized in the cytoplasm and extending toward the APM. We found that beta-adrenergic stimulation: 1) transiently halts NADH oscillations; 2) induces an initial increase in mitochondrial potential followed by a sustained depolarization, a behavior consistent with an increased production of ATP; and 3) increases the motility of the mitochondria at the site of exocytosis. These finding are consistent with a rearrangement of the mitochondrial dynamics and bioenergetics in order to sustain regulated exocytosis. Finally, blocking the mitochondrial ATP-synthase inhibits the assembly of the actomyosin complex (a major energy requiring step), whereas blocking the activity of complex I does not affect exocytosis, suggesting that a different pathway is used to regulate mitochondrial metabolism during stimulated secretion.
Showing the most recent 10 out of 12 publications
