Proceedings · Session S-169 · filed September 28, 2026

Lab Technology & MethodsSession paper

High-Speed AFM Catches Nucleases Cutting DNA in Real Time

Kanazawa University researchers used high-speed AFM to film DNase I cutting DNA in real time, finding shape-dependent attack patterns and protamine shielding that held for over six minutes.

By Sophie Lindqvist4 min read743 words

Summary

  • Kanazawa University's Nano Life Science Institute visualized individual DNase I molecules cutting DNA in real time using high-speed atomic force microscopy; results published in Nature Communications.
  • The team's STORM framework (Scan–Target–Occupy–Rupture–Mobilize) describes recurring interaction patterns; similar dynamics appeared with micrococcal nuclease, suggesting general physical principles.
  • Protamine-condensed DNA toroids resisted DNase I degradation for more than six minutes of continuous observation, with protection attributed to physical packing rather than electrostatics alone.
Scientists Observe Enzymes Breaking Down DNA in Real Time
FigureScientists Observe Enzymes Breaking Down DNA in Real Time — AI-generated

Researchers at Kanazawa University's Nano Life Science Institute have directly visualized individual nuclease enzymes locating and cutting DNA molecules in real time, using high-speed atomic force microscopy (HS-AFM). The work, published in Nature Communications under the title "Shield strike shatter in DNA topology and nuclease interactions," shows that DNase I does not attack DNA uniformly: the enzyme repeatedly returns to exposed ends and curved, locally bent regions before visible fragmentation occurs.

The observation matters for anyone working on DNA stability, extracellular DNA clearance, or gene delivery, because it reframes degradation as a geometry-dependent process. DNA is not a passive target, the researchers argue. "Its local shape and higher-order organization strongly influence where enzymes interact and whether degradation can proceed," says Richard Wong, PhD, professor and team leader at the institute.

The method and what it can and cannot see

HS-AFM records nanoscale changes in biological molecules in liquid without fixing, staining, or crystallizing samples. That capability let the Kanazawa team follow individual DNase I molecules as they sampled DNA, lingered at particular regions, and eventually fragmented the molecule — events that bulk biochemical assays cannot resolve.

The researchers are explicit about the technique's limit. HS-AFM cannot directly show the chemical reaction at the enzyme's active site. The relationship between DNA shape, repeated enzyme engagement, and subsequent cleavage is, in their words, a spatial and temporal correlation rather than direct observation of catalysis. R&D managers evaluating single-molecule imaging claims should hold that distinction firmly in view: the movies show where enzymes go and when DNA breaks, not the bond-making and bond-breaking itself.

The team observed that DNase I frequently accumulated near exposed DNA ends and curved regions. Longer-lasting interactions occurred disproportionately around those curves, which were also the regions most likely to be cut. Individual enzyme molecules revisited restricted zones multiple times before fragmentation became visible.

From these patterns, the researchers built a conceptual framework they call STORM: Scan–Target–Occupy–Rupture–Mobilize. An enzyme samples the DNA, localizes at a region, stays associated with it, fragmentation occurs, and the resulting pieces redistribute. They stress that STORM describes observed interaction patterns, not a mandatory sequence every enzyme molecule must pass through.

The framework held for a second enzyme. Micrococcal nuclease (MNase), which differs chemically from DNase I, showed similar behavior: sampling, localized interaction, repeated association, disruption, and fragment redistribution. That convergence suggests STORM-like dynamics may reflect general physical principles governing nuclease–DNA interactions rather than properties of one enzyme.

Protamine packaging as a physical shield

The experiments also quantified how DNA's higher-order structure protects it. The team condensed DNA with protamine, the small arginine-rich protein that packages the paternal genome in sperm cells. Under their experimental conditions, protamine produced two dominant compact forms: elongated rods and ring-shaped toroids.

DNase I molecules gathered around these condensed structures without destroying them. The toroids proved especially resistant: the enzyme rarely penetrated their central regions, and the structures remained intact under continuous observation for more than six minutes with enzyme present. When the condensation was partially loosened, exposed regions again became susceptible to cutting.

The interpretation is significant for formulation work: protection did not arise simply from electrostatic interaction between protamine and DNA. Tight three-dimensional packing itself created a physical barrier limiting enzyme access.

Implications, measured against the evidence

The biological stakes are concrete. DNase I clears DNA released from damaged or dying cells, and failures in that clearance process have been associated with inflammatory and autoimmune diseases. Extracellular DNA that escapes removal can stimulate immune responses. The Kanazawa results offer a mechanistic picture of why some DNA is accessible to nucleases while other DNA stays protected — including the extreme compaction found in sperm.

The therapeutic extrapolation is that packaging genetic material into structures restricting nuclease access could increase resistance to degradation, a consideration directly relevant to DNA-based therapeutics and gene-delivery vector design. That remains a projection; the published data cover DNase I and MNase acting on bare and protamine-condensed DNA under HS-AFM imaging conditions, not delivery vehicles in physiological environments.

"Our high-speed AFM imaging allows us to follow the interaction between individual nuclease enzymes and DNA as it happens," says Jingge Yang, a student on the team. The group's next step, implied by the correlation-versus-catalysis caveat, will be connecting these visualized dynamics to the reaction chemistry at the active site.

via nature.com (Original)

Filed under

  • high-speed-afm
  • nucleases
  • dna-degradation
  • single-molecule-imaging
  • gene-delivery
Share this article:

More from Sophie Lindqvist

Sophie Lindqvist

Show full bio

Correspondent covering business strategy at Hypothesis Wire.

86 articles

References

  1. Six of 79 Collisions: Graz Team Maps the 'Cone of Reaction'
  2. Anthropic's Claimed AI Biology 'Discovery' Draws Sharp Pushback
  3. Researchers Find Hundreds of Altered Western Blots in Thermo Fisher Catalogue
  4. Talus Bio Releases Structure-Free AI Model for Disordered Proteome
  5. Baylor Team Couples Proteomics and AI to Degrade VAV1

« Previous articleNext article »