The spatiotemporal control of ER membrane fragmentation during reticulophagy

ABSTRACT

Reticulophagy is an evolutionarily conserved mechanism essential to maintain the endoplasmic reticulum (ER) homeostasis. A series of studies identified a panel of reticulophagy receptors. However, it remains unclear how these receptors sense upstream signals for spatiotemporal control of reticulophagy and how ER is fragmented into small pieces for sequestration into phagophores. Recently, we and others showed that the oligomerization of RETREG1/FAM134B (reticulophagy regulator 1), an reticulophagy receptor, triggers the scission of ER membrane to facilitate reticulophagy. Furthermore, we demonstrated that upstream signals are transduced by sequential phosphorylation and acetylation of RETREG1, which stimulate its oligomerization, ER fragmentation and reticulophagy. Our work provides further mechanistic insights into how reticulophagy receptor conveys cellular signals to fine-tune of ER homeostasis.

Abbreviations: ER, endoplasmic reticulum; MAP1LC3, microtubule-associated protein light chain 3; RETREG1, reticulophagy regulator 1; RHD, reticulon-homology domain

ER is the largest intracellular organelle in eukaryotic cells and it is composed of continuous membranous network. ER is highly dynamic, and the half-life of ER membrane proteins and lipids is approximately 3–5 days. When cells are stimulated by ER stress, the unfolded protein response (UPR) will be activated, so that misfolded proteins will accumulate in the ER lumen to reduce the cytotoxicity caused by environmental damage. Subsequently, physiological ER homeostasis needs to be reestablished, and cells will start the ubiquitin-proteasome system, so that misfolded proteins will be degraded by the proteasome. The process is named “endoplasmic reticulum-related protein degradation” (ERAD). However, the turnover of the ER membrane is regulated by a selective autophagy process termed as reticulophagy/ER-phagy, which plays an essential role in maintaining ER homeostasis. Not surprisingly, dysregulation of reticulophagy is associated with various human diseases.

However, the entire ER network is too bulky to be engulfed by autophagosomes, which are approximately 0.5–2 µm in diameter. Thus, the breakup of ER into “bite-size” fragments becomes a prerequisite for their subsequent sequestration by phagophores. Work from the Dikic and Nakatogawa laboraotries demonstrated that reticulophagy receptors can fragment the ER subdomains in which they are located. and subsequently mediate their recognition by MAP1LC3 (microtubule-associated protein 1 light chain 3) protein family members present in the inner leaflet of forming autophagosomes for engulfment. At least two reticulophagy receptors, RETREG1 (reticulophagy regulator 1) and RTN3 (reticulon 3), have been proposed to have such dual functions because they both contain reticulon-homology domains (RHDs), which possesses ER membrane-bending properties, and MAP1LC3-interacting regions (LIRs), which equip them with “eat-me” signal. However, the mechanisms through which RETREG1-mediated reticulophagy is regulated in mammalian cells remain poor understood.

RETREG1 has been the first mammalian reticulophagy receptor to be identified and it is highly conserved in eukaryotes. Dysfunction of RETREG1 leads to hereditary sensory and autonomic neuropathy type 2, a neurobiological disease. The RHDs of RETREG1 are indispensable for ER membrane fragmentation and reticulophagy. In a proof-of-concept study, we discovered that RETREG1 forms homo-oligomers through its RHDs, driving ER membrane scission during reticulophagy [1]. Protein mass-spectrometry analysis revealed that RETREG1 has three phosphorylation sites, at Ser149, Ser151 and Ser153, and these residues are located in the flexible cytoplasmic loops that bridges the the second and third RHD (Figure 1). These phosphorylations promote RETREG1 oligomerization, thereby enhancing ER scission and reticulophagy. Further biochemical characterization of these modifications uncovered that the CAMK2B (calcium/calmodulin-dependent protein kinase II, beta) phosphorylates RETREG1 at Ser151 to accelerate reticulophagy in response to ER stress caused for example by an imbalance of the ER calcium concentration.

Figure 1.

Schematic representation of predicted structure of RETREG1/FAM134B. TM, transmembrane helices; RHD, reticulon-homology domain; LIR, MAP1LC3-interacting region.

Interestingly, protein mass-spectrometry also revealed the acetylation of Lys160. We further investigated the upstream signal mediating this post-translational modification by generating a polyclonal antibody specifically recognizing Lys160 acetylation (Ace-K160), which allowed us to show that acetylation occurs prior to phosphorylation at S151. Mimicking acetylation or phosphorylation mutants of RETREG1 detect RETREG1 self-interaction, ER fragmentation and reticulophagy. Lys160 acetylation elicits strong and transient reticulophagy when compared with phosphorylation, which induces mild and persistent reticulophagy. Mechanistically, ER stress leads to the recruitment of the CREBBP/CBP (CREB binding protein) acetyltransferase to the ER, which catalyzes acetylation of RETREG1 Lys160 and subsequent recruitment of CAMK2B for RETREG1 phosphorylation at S151 which were confirmed in Qiming Sun’s previous study. SIRT7 (sirtuin 7) deacetylates Lys160 shortly after the reticulophagy catalyzed by CBP, which tempers reticulophagy flux to avoid excessive ER degradation. Together, our work describes how reticulophagy receptor integrates divergent signals for spatiotemporal control of ER fragmentation to facilitate reticulophagy.

It is possible that other reticulophagy receptors may cooperate with either RETREG1 or RTN3 for ER fragmentation to intiate reticulophagy. Given the fact that mitochondrial and peroxisomal fission machineries are required for mitophagy and pexophagy, respectively, we think the cargo downsizing strategy may also apply in other selective type of autophagy such as lipophagy and chlorophagy.

Funding Statement

This study is supported by The National Natural Science Foundation under Grant 32025012, 92254307, 31970695, 31771525 to Q.S, and by Ministry of Science and Technology of the People’s Republic of China under Grant 2021YFC2700901 to Q. S.

Disclosure statement

No potential conflict of interest was reported by the authors.