Biomolecular condensation of ERC1 recruits ATG8 and NBR1 to drive autophagosome formation for plant heat tolerance

Ka Kit Chung, Kai Ching Law, Ziwei Zhao, Juncai Ma, Xiao-Tong Zhan, Cheuk Him Chiang, Kwan Ho Leung, Ruben Shrestha, Yixin Wu, Chaorui Li, Ka Ming Lee, Lei Feng, Xibao Li, Kam Bo Wong, Shou-Ling Xu, Caiji Gao, and Xiaohong Zhuang 

PNAS; November 10, 2025; 122 (46) e2425689122; https://doi.org/10.1073/pnas.2425689122

Significance

Biomolecular condensation is emerging as a crucial process for cellular stress responses. Upon stress conditions, macroautophagy serves as a fundamental metabolic pathway by forming a double-membrane autophagosome. However, the initiation mechanisms via biomolecular condensation in plant autophagy remain largely unexplored. Here, we have identified an uncharacterized but conserved plant ERC protein family as a ATG8-interacting scaffold in plant autophagy. We report an ER-associated condensate mediated by ERC1 for autophagosome initiation in plant cells. Further analysis of the ERC1 interactome and functional analysis reveal an undescribed role of ERC1 in NBR1 turnover and heat stress recovery. This work unveils a biomolecular condensation–driven mechanism for plant autophagosome formation and sheds light on plant adaptation to heat stress.

Abstract

Macroautophagy (hereafter autophagy) is essential for cells to respond to nutrient deficiency by delivering cytosolic contents to vacuoles for degradation via the formation of a multilayer organelle named an autophagosome. A set of autophagy-related (ATG) regulators are recruited to the phagophore assembly site for phagophore initiation, including its expansion and closure, and subsequent delivery into the vacuole. However, it remains elusive how the phagophore assembly is regulated under different stress conditions. Here, we described an uncharacterized Arabidopsis (Arabidopsis thaliana) ERC (ELKS/Rab6-interacting/CAST) protein family as an interacting partner of ATG8. ERC1 proteins translocate to the phagophore membrane and develop into ring-like autophagosomes upon autophagic induction. Notably, we found that ERC1 proteins possess the ability to assemble into substantial droplets together with ATG8e proteins prior to ATG8 conjugation to the membrane. Through multiscale characterization, we demonstrated that the ERC1 membraneless droplet represents a distinct type of plant condensate. Additionally, ERC1 directly binds to NBR1 to promote NBR1 degradation. ERC1 dysfunction suppresses the turnover of ubiquitinated substrates and compromises plant tolerance to heat stress. Our study suggests a model for autophagic degradation in response to heat stress by the action of ERC1-mediated biomolecular condensation in Arabidopsis.

See https://www.pnas.org/doi/10.1073/pnas.2425689122

Figure 1:

Arabidopsis ERC protein family associates with ATG8e. (A) Volcano plot showing the ERC protein family in proximity to ATG8e in the atg2-1 mutant. Red dots indicate values with the statistical significance cut-off (S0 = 2, FDR = 0.01). Three biological replicates were included to generate the volcano plot. Domain architectures of the Arabidopsis ERC protein family, ERC1 and ERC2, are shown below. CC, coiled-coil domain; IDR, intrinsically disordered region. (B) Phylogenetic analysis of ERC orthologs (Left) and motif conservation (Right) among the representative species. Distinct motifs identified are illustrated in different colors. (C) Colocalization of GFP-tagged ERC1 or ERC2 with mCherry-ATG8e in transiently expressed Arabidopsis PSBD protoplasts. (Scale bar, 10 µm.) (D and E) Recruitment assay of GFP-tagged ERC1 or ERC2 with CNX-RFP-ATG8e in Arabidopsis PSBD protoplasts. CNX-RFP was used as a control. (Scale bar, 10 µm.) (F and G) Immunoprecipitation (IP) analysis of Flag-tagged ERC1 or ERC2 with YFP-ATG8e. Cell lysate from Arabidopsis PSBD protoplasts transiently expressing ERC1-5xFlag or ERC2-5xFlag with YFP-ATG8e or GFP was subjected to GFP-trap assay, followed by immunoblot using anti-GFP and anti-Flag antibodies, respectively. Similar results were obtained from three different independent experiments. (H) GST pull‐down analysis of purified GST or GST‐ATG8e with ERC1 proteins using Glutathione magnetic beads. 10% of input and pull-down fractions were subjected to SDS-PAGE and Coomassie Blue staining. Similar results were obtained from three different independent experiments. (I) RT-PCR analysis of ERC1 and ERC2 transcript levels in different Arabidopsis tissues. Actin was used as an internal control. Similar results were obtained from three different independent experiments.