Studies on invadolysin: a novel metalloprotease localizing to lipid droplets

Abstract

Invadolysin (INV) is a member of the M8 family of metzincin metalloproteases. The gene was discovered in the Heck laboratory. Based on studies in Drosophila, INV is important for mitotic progression, nuclear envelope protein dynamics, and germ cell migration. INV-like immunoreactivity has shown its association with lipid droplets (LDs), which are intracellular organelles for lipid and protein storage. INV is the first metalloprotease found on LDs. Thus, INV’s role and LD-associated pathways are the puzzles we would like to investigate. The formation of LDs is dependent on the nutritional status of cells and starvation can disrupt the generation of LDs. Based on this concept, I established a starvation / re-feeding system. When nutrition is sufficient, LDs were surrounded by INV, whereas no INV or LDs were found in the majority of starved cells. With a supply of oleic acid (OA), LDs re-appeared and so did INV localized to LDs. In this system, inhibition of protein kinase C (PKC) disrupts INV’s re-localization to LDs. As I found INV to be phosphorylated by PKC in vitro (residues within the N-terminus might be phosphorylated by PKC), I conclude that PKC might regulate INV’s re-localization in the starvation / re-feeding system. 3T3-L1 mouse fibroblasts can differentiate into adipocytes in vitro; this is termed adipogenesis. Since INV is a LD associated protein, the role of INV in adipogenesis is of interest. INV localized on LDs in the early stage of differentiation but disassociated from LDs in mature adipocytes. The levels of INV mRNA and protein were significantly increased upon differentiation to adipocytes. On the other hand, INV decreased when adipocyte differentiation was inhibited by PKC and PI3K inhibitors, suggesting that the increase of INV is required for adipocyte differentiation. I was interested to examine the possible role of INV in InR/PI3K/Akt signalling, and therefore compared wild type with mutant INV (Drosophila INV4Y7). Decreased levels of phospho-Akt and phospho-S6K, and increased mRNA levels of d4E-BP were observed in INV4Y7 mutant larvae, suggesting that INV may be required for InR/PI3K/Akt signalling. In addition, a decreased level of Lsd2 (LD binding protein) was found in INV4Y7 mutants. These correlations between INV and molecules important for signaling suggest that INV might be a mediator of nutritional metabolism. In light of these data, I speculate that INV plays a homeostatic role, possibly by affecting the InR/PI3K/Akt signaling pathway. In conclusion, the localization of INV to LDs is dependent on the activity of PKC. An increase in invadolysin accompanies adipogenesis, in which PKC and PI3K may be mediators. Examining mutant Drosophila, I found INV to be involved in InR/PI3K/Akt signalling. Collectively, I conclude that INV may serve as a regulator in adipogenesis and the InR/PI3K/Akt signaling pathway.