Finally, MD simulations suggested that GT3, but not the shorter GD3 ganglioside, presents its glycan chain in a geometry compatible with spike binding at the cell membrane. They also indicate that the spike may need to approach at an angle and rearrange nearby N-glycans.

To connect this to airway biology, we used human nasal epithelial cultures. HKU1 infection and spike binding were restricted to ciliated cells and depended on 9-O-acetylated sialic acids. The spike signal overlapped with A2B5-reactive, GT3-like glycotopes, linking receptor display to cell tropism.

We then zoomed in on the binding site. The structure showed that HKU1-B engages both sialic acid residues, while mutagenesis confirmed that contacts with the second residue strengthen binding. Nearby N-glycans also tune both binding strength and linkage selectivity.

Next, we then determined HKU1-B spike structures with and without its preferred glycan ligand. In both cases, the spike remained fully closed, in contrast to Wang et al., who reported spontaneous, glycan-independent opening. We discuss several possible reasons for this difference in the paper.

First, we showed that both HKU1-A and HKU1-B spikes strongly prefer 9-O-acetylated α2,8-linked disialosides. HKU1-B bound somewhat more strongly and was also more selective, with monomeric S1A affinities of around 100 µM.

Why does human coronavirus HKU1 target ciliated airway cells, and do its two serotypes activate their spikes in the same way? Building on our 2023 work, we combined #cryoEM, BLI, molecular dynamics and human nasal epithelial cultures to compare serotypes A and B 🧵 www.biorxiv.org/content/10.6...

Really looking forward to visiting ISTA in just over two months for the 9th Austrian #CryoEM Symposium! Excited to give an EMBO YIP Lecture, hear from a great line-up of speakers, and catch up with the cryo-EM community. Hope to see you there! www.imp.ac.at/research/cry...

The first hetero refinement is usually quite rough, with only a relatively small fraction entering the target class, as in our recent NS3 paper. www.biorxiv.org/content/10.1...

Reprocessed an old Glacios dataset and pushed this Fab fragment to 3.8 Å resolution! Not a record-breaker by today’s standards, but still pretty satisfying for a C1 particle with only ~48 kDa of ordered mass. #cryoEM

Just back from #NIDO2026 in Hong Kong! 🇭🇰 It was a pleasure to present our latest work, catch up with colleagues, and see three Utrecht Virology Lab PhD candidates present their work. Thanks to the organisers for a fantastic meeting and EMBO YIP for the financial support. See you in Banff in 2029!

Honoured to receive the first Sir Dave Stuart Early Career Award today. Dave was my PhD external examiner, making this award, recognising his contributions to structural biology, especially meaningful to me. Thank you to Instruct-ERIC and Thermo Fisher Scientific for championing ECRs.

Excited to welcome @tejashyamk.bsky.social to the ERC starting grant team, as he starts his PhD! Teja joins us from @sc-lab.bsky.social, where he built a solid foundation in biochemistry & #cryoEM 🔬 He’ll be uncovering new insights into norovirus replication and assembly. Exciting science ahead!

Gaan met die banaan! 🍌🇳🇱

Back from the Physics of Viruses and Protein Cages GRC in the beautiful Tuscan hills. Great to see the latest developments in the field and to catch up with familiar faces. Thanks to the organisers for the invitation to present, and to everyone I spoke with for the excellent discussions.

My favourite experiment in the whole study is the negative-stain polyclonal epitope mapping we did using elephant serum, affectionately known in the lab as Ellie-EMPEM 😄

How can AI and #cryoEM help save some of the biggest babies on the planet? 🐘🔬 Our new preprint identifies Elephant Herpesvirus gH/gL/gO as a receptor-binding complex and a promising vaccine candidate, now being tested in young elephants across European zoos! 💉 www.biorxiv.org/content/10.6...

Honoured to be selected as one of this year’s EMBO Young Investigators! Grateful to my team and colleagues for all their support, and excited about the opportunities this will create for my group and for our work in structural virology. www.embo.org/press-releas...

Had a great time hosting @proteincapsid.bsky.social over the past few days. Thanks for the exciting lecture and for sharing your tomography expertise. Until next time!

Rounding off an incredibly busy week with the excellent third installment of the Dutch Structural Biology Meeting! Many thanks to Arjen for the kind invitation to present our lab’s work.

Back from the 9th International Calicivirus Conference in Banff! Robin Veenstra and I really enjoyed both the stunning mountain views and the exciting science being shared in the field. Thanks to the organisers for putting together this great meeting. Lots of exciting collaborations ahead!

Our model could help reconcile noroviruses, and perhaps enteroviruses, with other positive-strand RNA viruses that use large “crown” complexes to couple genome replication with RNA export. NS3 may represent a minimalist alternative to this strategy.

Putting all of this together, we propose that NS3 transports newly synthesised plus-strand RNA from within single-membrane vesicles. Once exported, the RNA can be translated or packaged in the cytoplasm. A streamlined mechanism for linking replication to the next steps.

Similar predictions of NS3 and its enterovirus homolog 2C, including a 50-mer polyA strand, suggest a plausible path for RNA translocation. This may point to a conserved mechanism among SF3 ATPases in positive-strand RNA viruses (but further validation is needed)!

Here is the NTD prediction in all its glory! A compact, high-confidence transmembrane channel with convincing architecture and hydrophobicity profile. The pore is positively charged, suggesting it could help guide negatively charged RNA through the membrane. Looks solid to us!

Amazingly, this gave us a high-confidence transmembrane channel for both mouse and human NS3, with oleic acids forming a bilayer-like pattern. Our group now call this the “fatty acid hack.” The #alphafold predictions with and without fatty acids shown below 👇

At this point, the missing piece of the puzzle was the N-terminal membrane-binding domain, which we had replaced with a soluble hexamerization domain to enable structural studies 🧩 To understand the full picture, we turned to #AlphaFold3.

Does our hex-NS3 construct bind RNA? 🧬 Yes! Single-molecule magnetic tweezers experiments with the @dulinlab.bsky.social show binding to ssRNA and dsRNA, with compaction of the latter. No ATP-driven helicase activity detected (thoughts on that in the paper).

The nucleotide-binding sites were particularly well resolved. Clear density for ATPγS, Mg²⁺, coordinating residues, and water molecules! 🤩 A great starting point for structure-based drug design! 💊

The NS3 hexamer adopts a split lock-washer conformation, characteristic of AAA+ ATPases that use a hand-over-hand translocation mechanism. Also, it kind of looks like a Rolo...

From ~1000 images collected on a Glacios (thank you @ievadr.bsky.social!), we obtained a 2.9 Å resolution #cryoEM map of NS3 in complex with ATPγS. This gave us our first high-res look at the norovirus NS3 hexamer in action! It is a fun (and small) data set. We have uploaded this to EMPIAR.