Autophagy: Starved β‐cells seem different from starved body

Macroautophagy (hereafter referred as “autophagy”) is a process where cellular components are sequestered into a vesicle (autophagosome) and are then degraded by lysosomal enzymes through formation of a fused vesicle (autolysosome). “Constitutive” autophagy was recently shown to have a role in maintaining mass and function of pancreatic islet β‐cells1, 2. Additionally, dysregulated autophagy in β‐cells was observed to contribute to the development of type 2 diabetes3. However, the role of “inductive” autophagy has not been clearly elucidated in the function of β‐cells. Autophagy is induced by various physiological and pathological conditions. A representative condition is starvation. During nutrient deprivation, cells induce autophagy, supplying catabolites to use for themselves, and maintaining energy homeostasis. However, β‐cells should consider not only their own energy homeostasis, but also that of the whole body, because it is the responsibility of insulin‐secreting β‐cells. If β‐cells induce autophagy on starvation, resultant catabolism of nutrients could trigger insulin secretion, which should not occur for the balance of the whole‐body energy status and for maintaining blood glucose levels. Recently, Goginashvili et al.4 dissected the acute response of β‐cells to nutrient deprivation in terms of autophagy. They showed that deprivation of serum and either glucose or amino acids suppressed formation of autophagosomes in INS1 cells (a rat insulinoma cell line). Such a phenomenon appeared approximately 20 min after fasting, and was maintained for at least 6 h. Four‐hour fasting of mice also inhibited in vivo formation of autophagosomes in β‐cells. Such a response to starvation was explained by starvation‐induced nascent granule degradation (SINGD). On starvation, secretory granules were found co‐localized with lysosomes, and granule‐containing lysosomes (GCLs) increased in INS1 cells on electron microscopy. Further starvation for 6 h decreased proinsulin levels, a marker for nascent secretory granules. Co‐localization of (pro)insulin and lysosomes was also observed in β‐cells of fasted mice, but that of (pro)insulin and autophagosomes was not observed. Therefore, starvation acutely induced lysosomal degradation of (pro)insulin in the nascent secretory granules, although autophagy was depressed. Then, could there be a relationship between the two phenomena? Lysosome‐derived amino acids had been reported to induce translocation of the mechanistic target of rapamycin complex 1 (mTORC1) to lysosomal membranes, and mTORC1 activation. The authors showed that starved INS1 cells activated the mechanistic target of rapamycin (mTOR) through co‐localization between mTOR and GCLs near the Golgi complex. mTORC1 is known to suppress autophagy through Unc‐51‐like kinase 1 (ULK1) phosphorylation at S757 during sufficient nutrition. However, in the starved INS1 cells, ULK1 phosphorylation was kept high, and treatment of rapamycin attenuated it causing autophagy. Therefore, SINGD could suppress autophagy in a mTOR‐dependent manner. Then, what would happen if autophagy was induced during starvation? Beclin1‐triggered autophagy in murine islets and human islets enhanced insulin secretion even in the low‐glucose condition, as far as 70% of that in the high‐glucose condition. Therefore, SINGD‐mediated suppression of autophagy seemed important to keep insulin secretion low during fasting. If the primary target of starvation in β‐cells is secretory granules, regulators of the secretory granule would be involved in the process. As protein kinase D (PKD) controls insulin granule biogenesis at the Golgi, the authors searched for the role of PKD in SINGD. They found that starvation downregulated PKD1 activity, and PKD1 inhibition reduced the proinsulin amount along with unchanged proinsulin synthesis and increased GCLs in INS1 cells, which suggested enhanced degradation of insulin granules. PKD1‐depleted INS1 cells also increased mTOR expression, co‐localization with lysosomes and ULK1 phosphorylation, which would subsequently inhibit autophagy. In the meantime, PKD1‐activated p38δ−/− mouse β cells were resistant to fasting with regard to the formation of GCLs in vivo and ex vivo. In contrast, autophagic compartments were increased in fasted p38δ−/− mouse β‐cells, suggesting high PKD1 activity prevented SINGD and SINGD‐dependent suppression of autophagy. Considering that β‐cells are the first defender against hypoglycemia (Figure 1), the aforementioned findings are highly appropriate. Suppression of insulin secretion is the first action of the body to maintain euglycemia during nutrient deficiency. Because exocytosis of insulin granules is the last step of the series in the insulin secretory process, and the nascent granules are preferentially secreted, degradation of nascent insulin granules shortly after starvation would be most effective in the defense against hypoglycemia. In addition, such SINGD suppressed autophagy, and it prevented autophagy‐induced insulin secretion. β‐Cells appear to use a distinct mechanism to overcome a shortage of nutrients, but it might be different in the case of prolonged starvation. After the completion of the mission as a primary defender against hypoglycemia and passing it to following defenders, such as α‐cells and sympathoadrenal signals, β‐cells might induce autophagy just like other cells to maintain cellular homeostasis and survival. The depletion of secretory granules and suspension of SINGD would contribute to this conversion (Figure 1). Indeed, there are observations that overnight fasting was insufficient for in vivo induction of autophagy in mice β‐cells, but 24 h‐fasting was sufficient2, 5. Figure 1 A model summarizing autophagic response in β‐cells and its implication according to starvation period. During starvation and subsequent hypoglycemia, the first defense mechanism of the body is to decrease insulin secretion. Goginashvili ... In conclusion, according to the study by Goginashvili et al.4, fasted β‐cells acutely induced inactivation of PKD and lysosomal degradation of nascent secretory granules, and subsequently activated mTOR through co‐localization with lysosomes. mTOR activation suppressed so‐called “inductive” autophagy in the starved β‐cells. As compulsory induction of autophagy in the β‐cells enhanced insulin secretion ex vivo despite low‐glucose condition, this strategy of starved β‐cells seems appropriate to cope with hypoglycemia in a nutrient‐deficient environment. The role of “inductive” autophagy on not only the suppression of insulin secretion but also its stimulation would be further elucidated.