G-protein-coupled receptors regulate autophagy by ZBTB16-mediated ubiquitination and proteasomal degradation of Atg14L

1) While it appears to be clear that the Akt-GSK3b-ZBTB16-ATG14L pathway play a role in starvation-induced autophagy, how much this pathway contributes to the overall autophagy regulation is not very clear. In particular, Akt can also regulate autophagy through inhibition of mTORC1, another important regulator of autophagy, and the major signaling pathway is this unclear. Can inhibition of mTORC1 (e.g. by torin 1) induce autophagy in cells expressing ZBTB16S184A/T282A? Do the levels of ATG14L and ZBTB16 change only by serum starvation, but not by amino acid starvation, which is another potent autophagy-inducible condition? If so, upregulation of ATG14L by ZBTB16 suppression is not critically important for starvation-induced autophagy.

As suggested by the reviewers we have done the assays below, and found that inhibition of mTORC1 by Torin1 could still induce autophagy in HeLa cells expressing ZBTB16 S184A/T282A (Author response image 1A), and furthermore, amino acid starvation inducing autophagy did not change the levels of ATG14L or ZBTB16 (Author response image 1B). Author response image 1A and B has been incorporated as Figure3–figure supplement 2F and Figure3–figure supplement 1G, respectively. Thus, we conclude that a subset of GPCRs can suppress autophagy by ZBTB16-mediated ubiquitination and proteasomal degradation independent of mTORC1 or nutritional status.

Author response image 1

Download asset Open asset

2) It is not shown in this manuscript whether overexpression of ATG14L is sufficient for the increase in autophagic flux. This is a critical point that should be addressed experimentally. Furthermore, it was already suggested that overexpression of Atg14L can induce autophagy even in nutrient-rich conditions (Matsunaga et al. 2010 J. Cell Biol., Fan et al. 2011 PNAS). The authors should cite and discuss these previous findings.

We thank this reviewer for pointing out this key fact that overexpression of Atg14L is sufficient to induce autophagy under normal nutritional condition. We have done a flux experiment and found that overexpression of ATG14L is indeed sufficient for the increase in autophagic flux (Author response image 2).

We have specifically cited these two references in this sentence: “Overexpression of Atg14L has been shown to induce autophagy under normal nutritional conditions (Fan et al., 2011; Matsunaga et al., 2010).”

Author response image 2

Download asset Open asset

3) Although the authors show that serum starvation for as short as 30 min induces ZBTB16 reduction and Atg14L accumulation (Figure 3A), they determined the induction of autophagy by serum starvation at 24 h (Figure 3D). To be consistent, the authors should monitor the LC3 conversion at earlier time points. If autophagy is not induced at earlier time points even though Atg14L accumulation occurs quickly, the authors should clearly mention the fact. The current manuscript may give an impression that Atg14L accumulation quickly induces autophagy and thus may be misleading.

As suggested by the reviewers, we monitored LC3II level in a time course experiment, and found that LC3II level was increased after serum withdrawal at early time points as well (Author response image 3). This figure now replaced the original Figure 3A.

Author response image 3

Download asset Open asset

Minor comments:

1) The authors should be careful by not over-interpreting the results. They should include the discussion about the possibility that GPCR-GSK3beta-ZBTB16-Atg14L axis, though important, may represent an over-simplified pathway of autophagy regulation, as GPCR-GSK3beta regulates a variety of pathways and ZBTB16 has many other substrates beyond Atg14L. The observed change in the Atg14L levels alone may not be sufficient to drive alteration of autophagy in particular in the context of AMD3100.

We have added one sentence at the end of first paragraph of the Discussion: “Our data, however, do not rule out that GPCRs might regulate additional pathways in control of autophagy, nor do we exclude the possibility that ZBTB16 has substrates, in additional to Atg14L, in regulating autophagy”.

2) It would be helpful for readers if the authors could include a summary model showing how serum starvation or inhibition of GPCRs induces autophagy through stabilization of ATG14L because many factors and negative regulations are involved in the authors' model.

Our model is now shown in Figure 6H.

3) As ZBTB16 knockout mice are viable, it would be informative to provide or discuss any phenotypes of these mice that can be explained by a defect in autophagy.

Male ZTBT16-/- mice are sterile whereas female ZTBT16-/- mice are fertile, suggesting the spermatogenesis is defective without ZBTB16. Autophagy is known to be involved in the process of spermatogenesis. E.g. autophagy has been reported to be involved in paternal mitochondrial destruction during spermatogenesis of Drosophila (Politi et al., 2014). Deletion of Atg7 blocks spermatogenesis at late stage (Wang et al., 2014). Thus, it is possible that the male sterility of ZBTB16 is due to over-activation of autophagy. However, since the male sterility of ZBTB16-/- mice has not been characterized, it is difficult to suggest anything even in the Discussion at this stage.