LC3C-ATG4D regulatory axis supports coronavirus replication organelle formation independently of canonical autophagy

Vlachova, S.; Iovine, L.; Marano, V.; Polishchuk, E.; Cillo, M.; Donnici, L.; Machado, P.; Swuec, P.; Settembre, C.; Grumati, P.; De Francesco, R.; Herhaus, L.; Cortese, M.

doi:10.64898/2026.08.10.743971

bioRxiv · 10.64898/2026.08.10.743971

LC3C-ATG4D regulatory axis supports coronavirus replication organelle formation independently of canonical autophagy

Vlachova, S.·Iovine, L.·Marano, V.·Polishchuk, E.·Cillo, M.·Donnici, L.·Machado, P.·Swuec, P.·Settembre, C.·Grumati, P.·De Francesco, R.·Herhaus, L.·Cortese, M.

View original source ↗Export metadata ↓

Abstract

Coronaviruses hijack host membranes to assemble ER-derived double-membrane vesicles (DMVs) that shield viral RNA replication from the cell intrinsic surveillance. Although DMVs morphologically resemble autophagosomes, whether and how autophagy factors actively support their biogenesis has remained elusive. Here, we identify a non-canonical requirement for the autophagy protein LC3C in {beta}-coronavirus replication. Loss of LC3s impaired viral RNA replication, whereas genetic ablation of ATG7 did not, indicating that canonical ATG7-dependent lipidation is dispensable in this context. Reconstitution experiments showed that only LC3C substantially restored replication in LC3-deficient cells and that LC3C phospho-mutants, differing in accessibility to ATG4-mediated processing, displayed distinct proviral activities. Additionally, ATG4D, the main protease responsible for maintaining the LC3 non-lipidated pool, is selectively required for viral replication. Both ATG4D and LC3s depletion triggers formation of aberrant DMV-like structures and potently suppresses SARS-CoV-2 replication. Ultrastructural analysis of nsp3-nsp4-induced membranes showed that depletion of LC3s or ATG4 proteases altered DMV abundance and morphology, supporting a role for the LC3C-ATG4D axis in replication organelle biogenesis. These data establish that {beta}-coronaviruses repurpose ATG4D-driven LC3C de-lipidation for non-canonical LC3 recruitment to replication organelles, identifying the lipidation state of LC3 as a molecular determinant of replication organelle biogenesis and efficient viral replication. HighlightsThe manuscript shows that {beta}-coronavirus replication depends on LC3 proteins and particularly on LC3C in reconstitution experiments, that this dependency is independent of ATG7-mediated lipidation, and that ATG4D promotes efficient replication and replication organelle morphology. Together, the data support a model in which a non-canonical LC3C-ATG4D pathway contributes to DMV biogenesis and viral RNA replication.

Explore related subjects

cell biology Preprint

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

Vlachova, S., Iovine, L., Marano, V., Polishchuk, E., Cillo, M., Donnici, L., Machado, P., Swuec, P., Settembre, C., Grumati, P., De Francesco, R., Herhaus, L., Cortese, M.. 2026-08-11. LC3C-ATG4D regulatory axis supports coronavirus replication organelle formation independently of canonical autophagy. https://doi.org/10.64898/2026.08.10.743971

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related preprints

bioRxiv2026

A dynamic phosphoproteome reveals stage-specific regulation of cytokinesis in Leishmania

Leishmania parasites undergo extensive morphological and metabolic remodelling during their transition from motile insect promastigotes to non-motile intracellular amastigotes, yet how dynamic protein kinase signalling drives this process remains poorly understood. Here, we present a high-resolution temporal phosphoproteome of L. mexicana promastigote to axenic amastigote differentiation, profiling 16,495 phosphosites on 3,101 proteins across seven developmental time points and identifying 5,795 dynamically regulated sites. This temporal map reveals stage-specific signalling programs, including regulated sites in MPK2 and non-regulated sites in MRK1, protein kinases known to be essential for amastigote differentiation. Using analogue-sensitive (AS) chemical genetics, we show that acute inhibition of either kinase in axenic amastigotes disrupts the spatial coordination of cleavage furrow ingression with nuclear and kinetoplast segregation, revealing essential cytokinesis roles obscured in constitutive gene deletions. Despite generating similar cytokinesis phenotypes, genetic over-expression and cellular localisation analyses demonstrate that MPK2 and MRK1 act through spatially segregated signalling pathways. Time-resolved phosphoproteomics upon MPK2 inhibition places its activity upstream of cytoskeletal remodelling networks. Importantly, we show that the stage-specific requirement for MPK2 is defined by the parasite's morphology: while motile promastigotes tolerate MPK2 inhibition, genetic ablation of flagellar motility sensitises them to loss of MPK2 activity. Thus, flagellar motility acts as a mechanical buffer during promastigote division, explaining why non-motile amastigotes are vulnerable to MPK2 inhibition. Together, our findings provide a systems level framework for Leishmania signalling and demonstrate that cellular architecture and biomechanics dictate protein kinase dependency across developmental stages of the parasite.

cell biology↗

bioRxiv2026

Single-cell Transcriptome-Proteome-PTMome(x8)-Metabolome (scTPPM) co-profiling decodes multimodal regulation during mouse oocyte meiosis

Single-cell multi-omics technology is crucial for comprehensive characterization of cellular molecular features. Despite recent advances in single-cell multi-omics approaches, co-profiling of nucleic acids, proteins, metabolites, and post-translational modifications (PTMs) within the same cell individuals remains major challenges. Here, we developed a single-cell transcriptome, proteome, PTMome, and metabolome (scTPPM) profiling approach, enabling simultaneous analysis of transcriptomes, proteomes, eight types of PTMs (e.g., phosphorylation, methylation, ubiquitination, O_GlcNAc, acetylation, crotonylation, lactylation, and succinylation), and metabolomes in individual cells. We demonstrated the scTPPM's performance in tumor cell lines and oocyte samples, reaching total identification and quantification of 27,668 genes, 4,847 proteins, 1,508 PTM sites, and 680 metabolites in single mouse oocytes. We profiled previously-uncharacterized dynamic crosstalk among the entire metabolic pathways from metabolic enzyme expression and regulation to metabolite-mediated post-translational modifications during oocyte maturation. In summary, scTPPM provided an effective workflow for multi-omics co-profiling in single cells, enabling multi-layered insights into complex regulatory network.

cell biology↗

bioRxiv2026

Sequence context and partner availability specify divergent cell fates from a single transcription factor

Transcription factors drive the gene regulatory networks that control developmental transitions, yet which of their many targets a cell uses remains unresolved. Expressing Cebp, Cebp{beta} or Cebp{delta} identically, alone or with Oct4, Sox2, Klf4 and cMyc, gave four outcomes: adipocyte from fibroblasts and neural stem cells, trophectoderm from embryonic stem cells, and myeloid from B cells. Raising Cebp levels changed how many sites it reached, not which sequence it read: clustered motifs maintained somatic enhancers, while spillover onto low-copy sites created de novo enhancers carrying composites of Cebp and cell-type-restricted partners. In fibroblasts, Cebp also redistributed Oct4, Sox2 and Klf4 from early targets, blocking iPSC formation. Supplying the predicted partners rewrote the output: Spi1 redirected Cebp into parallel adipogenic and myeloid programs, via a transient PpargCD45 state using separate enhancers for each; TFAP2C abolished adipogenic conversion. Motif density and partner composites therefore assemble a factor's instructive output in the receiving cell.

cell biology↗