BGPT: Paper Review: US6 hijacks the peptide-loading complex by trapping transporter-chaperone dynamics

Explore by Goal

Quick Explanation Copied

Core finding

This cryo-EM paper reports an atomic mechanism for how HCMV US6 blocks the MHC-I peptide-loading complex (PLC) by (i) latching its C-terminal TM helix laterally onto TAP2 and (ii) inserting a disulfide-stabilized luminal plug that mimics a translocating peptide, trapping TAP in an outward-open, nucleotide-asymmetric state while preserving PLC architecture.

Evidence anchor: 2.59–2.88 Å cryo-EM maps of native US6-arrested PLC; US6–TAP buried interface areas; nucleotide occupancy; void-volume occlusion and MHC-I surface reduction assays.

Long Explanation

Paper review (rigorous & skeptical): US6 hijacks the peptide-loading complex by trapping transporter-chaperone dynamics

Preprint DOI: 10.1101/2025.06.25.661491

Tripartite message: (1) US6 arrests PLC by physically occluding TAP’s luminal gate from the ER-lumenal side; (2) US6 engagement locks TAP in an outward-open conformation with an asymmetric nucleotide occupancy consistent with a post-hydrolysis-like state; (3) TAP2’s N-terminal TMD0 appears to be the dominant scaffold for tapasin recruitment/orientation in this arrested state.

Primary structural evidence: 2.59–2.88 Å local refinements from a consensus map built from 109,445 particles (initial stack 109,445 from 726,723 picked; initial stack mentioned explicitly as polished 109,445 and 25,454 micrographs); with deposited cryo-EM maps and an atomic model (PDB 9RCV, EMDB: EMD-53326/30/31/32/34).

1) Mechanistic model (what the authors claim, mechanistically)

US6 latches to TAP2 and blocks the ER-lumenal exit. The C-terminal TM helix of US6 lies against the TAP2 side and the central disulfide-stabilized plug “seals” the luminal exit gate such that peptide passage is prevented (with only a narrow ∼5 Å aperture at the lumenal gate edge reported via void analysis/geometry).
US6 mimics peptide engagement to trap TAP in an outward-facing open conformation with asymmetric nucleotide binding. They report Mg2+-ADP at the canonical site and Mg2+-ATP at a non-canonical inactive site, arguing this asymmetry resembles a post-hydrolysis-like state and that US6 “mimics” peptide N/C termini within the TAP groove.
TAP2 TMD0–tapasin transmembrane helix interactions dominate PLC architecture in this arrested state. They emphasize strong hydrophobic interface (and salt bridges) between TAP2’s TMD0 and tapasin’s transmembrane helix, contrasting with a weaker TAP1–tapasin interface, supporting a division-of-labor model.

2) Visual evidence from the paper (quantitative anchors)

(A) Cryo-EM resolution after focused refinement and consensus scale

Resolution anchors come directly from the paper’s reported focused refinement range (2.59–2.88 Å).

(B) Structural occlusion geometry: TAP void volume with/without US6 (reported)

The paper states void volumes (without US6: 7495 Å³; with US6: 2786 Å³) and a remaining narrow ∼5 Å aperture at the lumenal gate edge.

Interface areas: US6–TAP total 3342 Å², with 2178 Å² attributed to US6–TAP2; TAP1–TAP2 one 6018 Å² used as a comparison stated in the paper.

3) Strengths (what is persuasive)

Atomic structural mechanism in native context. The paper emphasizes the PLC was isolated from engineered human cells expressing US6 SBP and then solved by cryo-EM with high local resolution (2.59–2.88 Å), with coordinates deposited.
Converging structural + functional readouts. The paper reports that wild-type US6 suppresses MHC-I surface expression in HeLa cells, while cooperative US6 interface mutants restore near-normal levels (single mutants often show little effect; triple mutants restore strongly).
Mechanistic specificity supported by multiple explicit molecular details. The paper links (i) plug insert geometry, (ii) salt-bridge/polar interaction sets, and (iii) nucleotide asymmetry to the proposed “trapping” mechanism rather than relying on a single observation.

4) Limitations & critical checkpoints (what could be wrong / incomplete)

Single-state snapshot risk. The structure is captured with US6 “arresting” TAP; the authors explicitly state they cannot resolve whether TAP is captured during an active translocation cycle or an assembly intermediate where spontaneous ATP hydrolysis may occur.
Detergent/lipid environment differences may affect subtle conformational energetics. The paper solubilizes PLC with GDN and uses cryo-EM in detergent micelles; this can shift equilibrium among PLC conformations.
Modeling flexible components remains limited. Calreticulin and ERp57 could not be modeled at atomic detail because of high structural flexibility; this limits direct atomic mechanistic claims about editing modules’ full behavior.
Engineered cell-line biology caveat. Functional readouts use HeLa transfection and PLC isolation uses engineered Raji lines expressing US6 SBP; cell context differences may matter.

5) Placement in the field (why this matters mechanistically)

US6 and ICP47 both target TAP but from opposite sides

The authors contrast US6 binding to the outward-facing open conformation from the ER-lumenal side with ICP47 binding to the inward-facing conformation from the cytosolic side, framing TAP as a vulnerable checkpoint.

They also situate their results within prior PLC structural work and TAP architecture/ABC transporter principles (e.g., known native PLC structure, and general ABC transporter mechanistic principles).

For ATPase/ABC transport cycles and conformational transitions, they reference broader mechanistic principles of ABC transporters.

6) Specific biological implications (confidence-tagged)

What seems well-supported

US6 requires physical mimicry/occlusion of TAP’s peptide-binding/gate region. The combined observations (plug insertion, peptide-pocket occupancy mimicry, void volume reduction, MHC-I suppression that depends on interface residues) make this plausible.
The nucleotide asymmetry is real in the solved density and is relevant to the trapping interpretation. While the cycle-stage uncertainty remains, the density-resolved bound nucleotides provide strong constraint for the inhibitory trapping model.

What is more tentative

ATP-cycle directionality and “post-hydrolysis” inference. They acknowledge inability to determine whether this is captured during turnover vs intermediate assembly state.
TAP1 vs TAP2 “division of labor” generality across dynamic PLC states. Their TAP2–tapasin dominance is demonstrated in the US6-arrested state, but dynamic remodeling during peptide loading/release may differ.

7) Suggested next falsification targets (what would change my mind)

Show that disrupting US6–TAP electrostatic complementarity removes occlusion without re-routing TAP into other permissive conformations. The paper already includes mutant functional effects, but a tighter mechanistic falsification would require ruling out alternative “trap” states not captured here.
Resolve additional arrested/transitional states (or use alternative inhibitors/conditions) to test whether TAP’s outward-open + asymmetric nucleotide occupancy is uniquely produced by US6 or also occurs in other PLC intermediates.

Author review links (bespoke)

Feedback:

Updated: April 23, 2026