henhaobio
Sign in

Nature CommunicationsMicrobiologyPeer-reviewed journal article

Cryo-electron tomography reveals coupled flavivirus replication, budding and maturation

Selma DahmaneiD, Erin E. SchexnaydreiD, Jianguo ZhangiD, Bina Kumari SinghiD, Ebba RosendaliD, Nunya ChotiwaniD, Kiran Bala SharmaiD, Emma NilssoniD, Marie Berit Akprioro PetersiD, Wai‐Lok YauiD, Sebastian RönfeldtiD, Richard LundmarkiD, Benjamin Asher BaradiD, Danielle Ann GrotjahniD, Susanne LieseiD, Andreas CarlsoniD, Anna K. ÖverbyiD, Lars A. CarlsoniD

PDFDOIPreprint
In-depth full-text digest available · Jump to digest ↓

Abstract

Flaviviruses replicate their genomes in replication organelles (ROs) formed as bud-like invaginations on the endoplasmic reticulum (ER) membrane, which also functions as the site for virion assembly. While this localization is well established, it is not known to what extent viral membrane remodeling, genome replication, virion assembly, and maturation are coordinated. Here, we imaged tick-borne flavivirus replication in human cells using cryo-electron tomography. We find that the RO membrane bud is shaped by a combination of a curvature-establishing coat and the pressure from intraluminal template RNA. A protein complex at the RO base extends to an adjacent membrane, where immature virions bud. Naturally occurring furin site variants determine whether virions mature in the immediate vicinity of ROs. We further visualize replication in mouse brain tissue by cryo-electron tomography. Taken together, these findings reveal a close spatial coupling of flavivirus genome replication, budding, and maturation.

In-depth summary

Claude Sonnet 4.6 · xhigh effort · 8 Nov 2026 · from the full text of version 2 (21 Oct 2024) · CC BY 4.0

Cryo-ET of Langat virus-infected cells shows replication organelle size set by a membrane coat plus RNA pressure, with virion budding and maturation occurring beside the organelles.

Flavivirus replication organelles were known mainly from resin-embedded electron microscopy, which destroys protein structure, leaving their protein and RNA organization and the coordination of replication, assembly and maturation unresolved. Here in situ cryo-ET addresses organelle shaping, a membrane-linking complex, and where virions mature relative to organelles.

Key results

  1. ShownReplication organelles (ROs) in Langat virus (LGTV)-infected A549 cells exist in two forms. Most contain luminal filamentous density (filled), and a minority lack it (empty) and are much smaller.

    83±14% of ROs were filled across 23 tomograms. Empty ROs averaged 46±8 nm in diameter (N=25) vs 85±5 nm for filled ROs (N=63); unpaired two-tailed t test, p<0.0001. · Cryo-electron tomography reveals two states of…; Figure legends

  2. SuggestedRO membranes are thicker than adjacent ER membrane, consistent with a curvature-setting coat. A sphere-based Helfrich energy model gives a spontaneous curvature from empty ROs and an RNA pressure for filled ROs.

    Thickness 4.0 ± 0.2 nm (RO, N=132) vs 3.4 ± 0.2 nm (ER, N=4) in four tomograms. Model: H0 = 0.04 nm^-1 from empty ROs (2R = 46 nm); P = 5 · 10^-4 kBT nm^-3 for filled ROs (2R = 85 nm). · A combination of membrane coat and RNA-induced…

  3. ShownImmature and mature virions sit in separate but intertwined membrane compartments near ROs. Mature virions are slightly farther from the nearest RO, which the authors read as maturation close to ROs.

    Center-to-center distance to nearest RO in 5 tomograms: immature 95±56 nm (N=37) vs mature 147±41 nm (N=24); p=0.0003, unpaired t test. Furin and dsRNA signals were also close by immunofluorescence. · Virions form and undergo maturation in the…

  4. ShownA single prM residue in the TBEV furin site (R86 vs Q86) changes furin cleavage and the fraction of mature virions seen near ROs. R86 gave faster peptide cleavage and more mature particles.

    Mature virions: R86 46±46% (N=7 tomograms) vs Q86 2.5±5.9% (N=7), p=0.02; wildtype LGTV 14±25%. R86 peptide cleaved faster by furin. In immunocompromised Ips1-/- mice, 5 of 5 died with R86 and 7 of 10 with Q86. · The TBEV furin site variants R86 and Q86 differ…

  5. SuggestedA protein complex at the RO neck links the RO membrane to an apposed ER membrane. Similar complexes appear between an RO and a nearby immature virion budding site.

    Complex mass estimated from tomogram volumes as 500±151 kDa (n=7). Seen at necks of both filled and empty ROs; too few occurrences for subtomogram averaging. · A protein complex connects the replication…; Figure legends

Methods, data and reproducibility

Cryo-FIB milling and cryo-ET of LGTV-infected A549 cells and high-pressure-frozen mouse choroid plexus, with membrane thickness morphometrics, subtomogram averaging of virions, and an adapted Helfrich membrane energy model of RO shape.

  • Cryo-ET of FIB-milled LGTV-infected A549 cells at 24 h post infection (RO state, size and virion analyses) · human (A549 cells) · n = 23 tomograms (filled-RO fraction)
  • Cryo-ET of cells infected with chimeric LGTV carrying TBEV prM/E, furin-site variants R86 and Q86 · human (A549 cells) · n = 7 tomograms each (text); legend: R86 n=6, Q86 n=4
  • Subtomogram averages of immature and mature virions from cellular tomograms · Langat virus (in human A549 cells) · n = 84 immature (11 tomograms), 51 mature (3 tomograms) · EMD-51640 (immature), EMD-51642 (mature)
  • Cryo-ET of high-pressure-frozen, FIB-milled choroid plexus from LGTV-infected Ifnar-/- mice · mouse
  • Survival of Ips1-/- mice infected intraperitoneally with chimeric R86 or Q86 · mouse · n = R86 n=5, Q86 n=10
  • Fluorogenic peptide cleavage assays with furin and PC1/3 on R86 and Q86 prM peptides · n = four independent experiments in duplicate
Methods & tools
SerialEM, MotionCor2, IMOD, CTFFIND4, WARP, Dynamo, cryoCARE, IsoNet, Amira, MemBrain V2, surface morphometrics toolbox, GraphPad Prism
Reproducibility gaps
No code repository or URL is given, including for the extended surface morphometrics toolbox. Only the two virion subtomogram averages are deposited at EMDB; deposition of tomograms is not stated.
Limitations (6)
  • Neck complexes were too few for structural analysis, so their identity and composition remain unresolved; larger tomographic datasets are needed for a decisive subtomogram average. · stated by the authors
  • No distinct repeating densities were visible on RO membranes, so the coat is inferred from membrane thickness. The authors only suggest NS1 as a candidate curvature generator. · stated by the authors
  • Immature and mature virions were never seen in the same compartment, but compartments may connect beyond the limited lamella thickness. · stated by the authors
  • Whether empty ROs are assembly intermediates, disassembly intermediates, or failed assembly events is unresolved. · stated by the authors
  • The maturation comparison rests on few tomograms with small virion counts and very large spread (R86 46±46%). Tomogram numbers for R86 and Q86 differ between text (N=7 each) and legend (n=6, n=4). · noted in this summary
  • The mouse brain data are qualitative single examples with no quantification. The RO model assumes spherical shapes and borrows membrane parameters from earlier work, so H0 and P are model-dependent estimates. · noted in this summary

Read this if you study flavivirus replication organelles or want an in situ cryo-ET workflow spanning FIB-milled cells and high-pressure-frozen brain tissue.

Citations
15
FWCI
40.53
Field percentile
top 1%

Reproducibility ledger

Nobody has reported trying to reproduce this paper yet. Did you replicate an experiment or run the code? Your report saves the next lab a week.

Sign in to file a reproducibility report.

Related papers

Similarity Map
SelectedSame fieldCross-field
THISOwen '25Renner '20Bignon '22Yau '24Klein '20Lin '23
74% semantic matchMicrobiology
Jonathan Einterz Owen, Cheyanne L Bemis, Qingyi Wang+10 more·bioRxiv·Nov 26, 2025· 1 citations