High avidity phase-separated RNA-protein sialogranules sense lectins and inhibit influenza infection

This study demonstrates that programmable, phase-separated RNA-protein sialogranules can be engineered to display high-avidity sialic acid ligands, enabling them to sense lectins and function as effective antiviral decoys that inhibit influenza virus entry.

The Gist

Imagine the influenza virus as a tiny, mischievous burglar trying to break into a house (your cells). To get in, the burglar doesn't have a master key; instead, he has a sticky, weak hand (a protein called Hemagglutinin) that tries to grab onto a specific doorknob on the house (a sugar molecule called sialic acid).

Normally, this grip is weak. The burglar might slip off. But because the virus has hundreds of these sticky hands, they all grab on at once, creating a super-strong "Velcro" hold that lets the virus break in and infect you.

The Problem:
Scientists want to stop this burglar. One idea is to build "decoy doorknobs" that float around in your body. If the virus grabs onto a decoy instead of your real cells, it gets stuck and can't infect anyone. The challenge is making these decoys sticky enough to actually catch the virus.

The Solution: The "Smart Cloud" (Sialogranules)
The researchers in this paper invented a new kind of decoy. Instead of building a solid plastic ball, they built a smart, squishy cloud made of two ingredients:

1. Synthetic RNA: A long, stringy scaffold (like the frame of a tent).
2. Proteins: Special proteins that act as the "sticky doorknobs" (sialic acids).

When they mix these two together, they don't just sit there; they spontaneously snap together to form a phase-separated granule. Think of it like oil and vinegar separating, or how a drop of water beads up on a leaf. These "sialogranules" are dense, gel-like blobs packed with thousands of sticky doorknobs.

The Magic Trick: The Shape-Shifting Test
How do the scientists know if their decoy clouds are sticky enough? They use a special "test agent" called SNA (a lectin that loves sialic acid).

• The Weak Cloud: If the cloud has few sticky doorknobs, the SNA agent just sits on the surface. The cloud stays small and round.
• The Strong Cloud: If the cloud is packed with sticky doorknobs, the SNA agent grabs onto many of them at once. This acts like a giant magnet pulling the cloud apart. The cloud suddenly expands, turning into a big, fluffy, multi-layered "bio-condensate" (a multiphasic agglutinated biocondensate, or MAB).

The Analogy:
Imagine a group of people holding hands in a tight circle (the granule).

• If you throw a ball (SNA) at them and they only have a few hands free, the ball bounces off, and the circle stays tight.
• If everyone in the circle has a free hand and grabs the ball, the circle stretches out, expands, and turns into a big, loose web.
• The scientists realized: The bigger the web gets, the stickier the cloud is. They can measure exactly how "sticky" (high-avidity) their decoy is just by looking at how much it expands.

The Results:

1. The Best Decoy: They tested many different types of proteins. They found that one specific protein, LAMP1, made the stickiest, most expandable cloud.
2. Stopping the Burglar: When they put these LAMP1 clouds in a petri dish with flu viruses, the viruses got stuck to the decoys. The viruses couldn't reach the real cells.
3. The Outcome: The LAMP1 clouds reduced the number of infected cells by about 50%. It's like putting a giant, sticky net in front of the front door; half the burglars get caught in the net and can't get inside.

Why This Matters:

• A New Tool: This isn't just a medicine; it's a new way to measure things. Because the cloud changes shape based on stickiness, scientists can use it to test how well proteins are decorated with sugars, which is hard to do with current tools.
• A New Medicine: This platform is "programmable." If a new virus comes out that grabs a different sugar, scientists can just swap the protein in the cloud to match that new sugar. It's like a universal remote control for viruses.

In a Nutshell:
The scientists built squishy, programmable clouds that act as super-sticky traps for the flu virus. They discovered that when these clouds are "sticky enough," they change shape dramatically, which lets them measure their own effectiveness. The best cloud they made stopped about half the flu viruses from infecting cells, offering a promising new way to fight viral infections.

Technical Summary

1. Problem Statement

• Viral Entry Mechanism: Influenza viruses (IFV) utilize hemagglutinin (HA) to bind weakly (low affinity, $K_d \sim$ mM) to $\alpha2,6$-linked sialic acid (SA) on host cell surface glycoproteins. Productive infection relies on multivalency: hundreds of HA proteins on a single virion create a high-avidity interaction that overcomes low individual binding affinities.
• Limitations of Current Glyco-Nanoparticles (GlycoNPs): Existing synthetic glycoNPs (gold, iron oxide, polymeric) face challenges in maintaining uniform glycan density, orientation, and accessibility. Furthermore, they often lack the structural flexibility and colloidal stability required for complex biological environments.
• Need for Sensing and Therapeutics: There is a need for programmable platforms that can both sense specific glycan modifications (like sialylation) with high sensitivity and act as decoys to inhibit viral entry by mimicking high-avidity host receptors.

2. Methodology

The authors adapted their previously developed synthetic RNA-protein (sRNP) granule platform to create "sialogranules."

• Platform Design:
  • Scaffold: Synthetic long non-coding RNA (slncRNA) containing a cassette of connected hairpin structures (PP7 binding sites).
  • Binding Protein: A tandem-dimer PP7 coat protein fused to mCherry (tdPCP-mCherry).
  • Functionalization: The tdPCP-mCherry backbone was fused to various proteins/peptides containing N-linked or O-linked glycosylation motifs (N-X-S/T).
• Candidate Constructs:
  • Natural Proteins: LAMP1, GYPA, Fetuin, and ACE2 (soluble domains).
  • Synthetic Peptides: NXST1, NXST2, and tdNXST1 (tandem repeats of glycosylation sequons).
  • Controls: Bacterial versions of proteins (lacking mammalian glycosylation machinery) served as negative controls.
• Assay Mechanism (The "Sialo-Sensor"):
  • Granules were formed in vitro by mixing slncRNA and sialylated proteins.
  • Lectin Induction: FITC-labeled Sambucus nigra agglutinin (SNA), a lectin specific for $\alpha2,6$-SA, was added.
  • Readout: The transition from a compact "spot-like" granule to a Multiphasic Agglutinated Biocondensate (MAB) was monitored via confocal microscopy.
  • Quantification: The distance ($D_{RNA}$) between the center of the RNA cluster (AF405) and the protein cluster (mCherry) was measured. In MABs, RNA is excluded to the periphery, increasing $D_{RNA}$.
• Validation & Inhibition:
  • Enzymatic Perturbation: Treatment with Endo H (removes N-linked glycans) and Neuraminidase (NA, removes sialic acids) to verify glycan dependence.
  • Viral Entry Assay: A549 lung cells were infected with Influenza A (H1N1) in the presence of LAMP1 sialogranules. Infection was quantified via flow cytometry (HA staining).
  • Modeling: A 3D Brownian diffusion simulation modeled virions interacting with decoy obstacles to predict infection flux reduction.

3. Key Contributions

1. Novel Glyco-Nanoparticle Platform: Demonstrated that phase-separated sRNP granules can be engineered to display high densities of glycans, creating "Velcro-like" high-avidity interactions.
2. Avidity-Based Phase Transition Assay: Established a quantitative method where the morphological transition of the granule (from compact to multiphasic) serves as a direct readout for sialylation density and avidity.
3. Quantitative Correlation: Showed a strong correlation between the number of putative sialylation sites and the magnitude of the phase transition ($\Delta D_{RNA}$), allowing for the extraction of effective avidity constants ($K_d$) and Gibbs free energy changes ($\Delta\Delta G$).
4. Therapeutic Proof-of-Concept: Validated that LAMP1-based sialogranules act as effective decoys, inhibiting influenza entry by ~50% in cell culture.

4. Key Results

• Phase Transition Specificity:
  • Granules containing mammalian-expressed sialylated proteins (e.g., tdPCPm, LAMP1m) underwent an abrupt transition to MABs upon SNA addition.
  • Bacterial-expressed (unsialylated) controls remained as compact granules.
  • Structural Change: In MABs, the SNA-sialoprotein complex forms a dense core, while the slncRNA is segregated to the periphery, significantly increasing the $D_{RNA}$ metric.
• Quantitative Avidity:
  • $\Delta D_{RNA}$ (difference in distance between low/high SNA) correlated with the number of sialylation sites (Pearson $r = 0.6$).
  • Dose-response curves allowed the calculation of effective $K_d$ values. The data followed a logarithmic dependency on valency, consistent with thermodynamic models of multivalent binding.
• Enzymatic Validation:
  • Treatment with Endo H or Neuraminidase abolished the MAB transition, restoring the compact granule morphology.
  • Endo H treatment reduced the effective sialylation sites by ~81%, while Neuraminidase reduced it by ~49%, confirming the assay's sensitivity to specific glycan modifications.
• Antiviral Efficacy:
  • IC50 Shift: LAMP1 sialogranules shifted the viral dose-response curve, increasing the IC50 by ~2.25-fold compared to controls.
  • Inhibition: The granules reduced influenza infection by approximately 50% in A549 cells.
  • Mechanism: The diffusion model confirmed that high-avidity decoys act as static obstacles that trap virions, disrupting their diffusion to the cell surface and reducing the flux of infectious particles.

5. Significance

• Diagnostic Utility: The platform offers a new, label-free (or minimally labeled) method to quantify post-translational modifications (PTMs) like sialylation on proteins, distinguishing between genetically identical proteins produced in different cell types (e.g., mammalian vs. bacterial).
• Therapeutic Potential: It establishes a programmable "glyco-decoy" strategy. By fusing different soluble receptors or glycanated peptides to the sRNP scaffold, the platform can be adapted to target various viruses (e.g., SARS-CoV-2 via ACE2, Influenza via LAMP1) or bacteria.
• Biophysical Insight: The study provides a mechanistic understanding of how multivalent ligand binding can drive phase transitions in biomolecular condensates, suggesting a general regulatory mechanism for biocondensates by external ligands.
• Technology Readiness: Currently at TRL 4 (validated in relevant biological systems). The authors outline a clear path toward preclinical development, including testing in mucus/sera and animal models.

In summary, this work successfully bridges synthetic biology, phase separation physics, and virology to create a versatile, high-avidity glyco-nanoparticle platform capable of both sensing specific glycan patterns and acting as a potent antiviral decoy.