Nano-dinosaurs and DNA robots

Welcome to Plenty of Room!

Today I am going all in on modular DNA origami nanorobots. Where do dinosaurs come into play? Well, continue read and you’ll find out!

Plenty of Room is your guide to the cutting-edge news related to molecular machines.

Already subscribed? Share with a friend that might find this interesting! It really helps.

New here? Just go ahead and subscribe!

Let’s get into it now.

Nano-dinosaurs and DNA robots

DNA origami is the go-to way for creating programmable DNA nanostructures. A long single-stranded DNA (ssDNA) scaffold is folded together using short ssDNA staples to form robust, customizable structures. these structures have been explored for various uses, from biomedical applications like drug delivery and vaccines to molecular pegboards for arranging molecules with nanoscale precision.

And while DNA origami is super cool (I spent my whole PhD using it!), it has limitations. Sometimes, you need bigger or more complex structures, and this is hard to do using a single scaffold. The natural solution is to assemble multiple origami structures together, but this is easier said than done. Hierarchical assembly faces several challenges:

• Low Yields: Getting multiple DNA pieces to self-assemble often results in inefficiencies.

• Static Designs: Many structures can’t transition between different shapes or configurations.

• Limited Functionality: Adding functional elements like proteins or nanoparticles without compromising stability remains challenging.

• High Costs: Unique DNA strands are expensive, and protocols often involve labor-intensive purification steps.

The team from today’s paper set out to solve at least some of these challenges, and they succeeded! I was very excited when I arrived to the last part of the paper and I saw the yield of their structures. But let’s go in order and start from the beginning.

This paper introduces a modular DNA voxel system designed for efficient, reconfigurable, and scalable assembly of nanoscale structures. The authors were apparently inspired by biological systems like proteins, which can self-assemble and dynamically adapt. Very cool! So, first of all: what is a voxel? As the name suggest, it’s simply a 3D counterpart to a pixel, in practice representing a unit of 3 dimensional space. In this case, each voxel consists of two DNA origami barrels connected by scaffold strands. The barrels form the structural core, while the strands allow the voxel to be rigid or flexible, depending on its intended use. Each voxel also has 12 external connection points for precise assembly, with lateral and coaxial connections enabling versatile designs. The authors analyzed and optimized the monomer structures using electron microscopy, focusing on making the actual particles as close as possible to the design, to ensure the best yield for the assembly steps.

Using these basic building blocks, the authors moved on to the hierarchical assembly. They started “simple”, creating a big variety of dimers structures (like, 15 or so, a bit crazy). But they didn’t stop there: they designed three voxel variants using nearly identical DNA strands, altering only the sequences that govern connections (which makes the structures more economical). Using this set of trimer structures, they created an astonishing 17 distinct designs! And there is more: they also used 8 monomer structures to create a dinosaur (which was also patterned using fluorescent markers), a rectangle and a representation of Australia, and then a whopping 12 monomers to create a dragon, a robot and a square. The team also used the voxels to create 3D structures, such as boxes and a rocket. Very impressive, considering that sometimes can be very challenging to get two or three subunits to come together: in their case, they report impressive yields of over 30%. It’s clear they had to do a lot of optimization to reach these results.

The team also showed that 2D and 3D voxel-based structures are still dynamic: for example, they can change the configuration of a flexible chain to a rod, a wave, a rectangle, and even to a recliner! And this is the coolest part of the paper. The researchers saw that their yield was decreasing when they were creating assemblies with higher numbers of monomers. They found that some designs had naturally high yields, and reconfiguration steps were highly efficient. The authors used this realization to optimize the assembly paths, using efficient transitions between high-yield precursors structures, mimicking folding mechanisms found in proteins (and other types of DNA nanostructures). My favorite example was a 9-voxel 3D circle, that had an assembly yield of 0.4% (practically nothing): using 2 intermediate reactions, the yield increase 40-fold, to 28%. Incredible!

This paper is exciting because it tackles an interesting and important problem for DNA origami. Still, this approach has some limitations:

• Sequence Complexity: it requires numerous unique DNA strands, increasing design complexity and cost.

• Concentration Constraints: assembly must occur at low DNA concentrations (I think the max they used was 1nM) to avoid nonspecific interactions.

• Physiological Compatibility: current designs are not yet optimized for use in biological environments.

But I think it’s a huge step in the right direction! By increasing yields and reducing costs, the voxel system opens the door to exciting applications:

• Nanorobots for targeted drug delivery

• Programmable molecular machines for synthetic biology

• Advanced materials for photonics and bioengineering

I couldn’t cover everything here, so go and read the paper here!

In other news:

• A glow up for RNA: If you like RNA more, this study introduces RNA lanterns, a new bioluminescent platform for RNA imaging. By using engineered RNA tags to recruit luciferase fragments, the system allows high-sensitivity detection with just a single tag. It enables real-time RNA tracking in cells and live animals, providing a versatile tool for studying RNA dynamics!

• Keep it positive with peptides: Peptides are quite interesting, since they are easier to synthetize and use than most other proteins. And they can also be used for nanotech: for example, this study explores how surface charge distribution affects the assembly of coiled-coil nanoparticles. These findings were optimized using machine learning and reveal the critical role of electrostatic patchiness in directing particle assembly.

• Automatic EM structure models: Modeling protein complex from EM data is very hard. To help, this paper introduces EModelX (very cool name). EModelX is an automated method for modeling protein structures from cryo-EM maps using sequence-guided modeling and deep learning. It aligns cryo-EM data with protein sequences, refines gaps with Cα threading, and integrates AlphaFold for enhanced accuracy. Pretty cool!

• Not yet a subscriber to Plenty of Room? Sign up today — it’s free!

• You think a friend or a colleague might enjoy reading this? Don’t hesitate to share it with them!

• Have a tip or story idea you want to share? Email me — I’d love to hear from you!

• You have something you would love me to cover? Just reach out here or on my social!