In order to achieve the project goal, we designed two constructs; the Receptor for the sensing system and the Motor-Monomer for the moving system. The Receptor and Motor models were based on the DNA origami structure from the earlier studies (see Early Trial section). We especially focused to improve the DNA origami structure to create motility upon a signal reception.
The Receptor is developed to be inserted in the liposome with two functions: recognition of an outside signal and activation of the Motor-Monomers inside to start polymerization.
For this, we designed the Receptor that consists of the Activator and the Wall surrounding the Activator. The Activator stimulates the Motor-Monomers to polymerize. In the absence of the outside signal, the Wall component, which is connected to the Activator with a linker, physically blocks the Activator to associate with Motor-Monomers. Upon outside signal recognition, the Wall separates from the Activator and the Activator is released into the liposome, resulting the activation of Motor-Monomers (the activation mechanism is described in the Motor-Monomer part).
To develop the Receptor, the following five mechanisms have to be implemented:
Structure (The Wall)
The Wall is constructed using DNA origami building block of 18 honeycomb structures. It is composed of the main body and the extension bar. The body part is U-shaped to surround the receptor, and the extension bar part is 16 nm long to penetrate into the liposomal membrane.
The upper ends of the honeycomb in the main body are smooth in order to attach to the liposomal membrane, while the other ends are rough to self- aggregation of the Wall.
The details of our strategy for the Receptor to fulfill the five aforementioned requirements are described below.
1. Embedding the Wall in the liposome
The whole Wall structure is so big that it cannot penetrate into the phospholipid bilayers of the liposome. We therefore insert only one honeycomb structure extended from the body part of the Wall into the membrane of liposome. Moreover, the upper side of the Wall is modified with cholesterol to decrease the polarity of the inserted part, supporting the embedding into the liposome.
2. Linking the Activator to the liposome
The Activator is a restriction enzyme HindⅢ. HindⅢ is an restriction endonuclease that recognizes and cleaves base sequence 5’-AAGCTT-3’. The Activator is associated with the liposome via linkage to the following Anchor embedded in the liposomal membrane.
The Anchor is composed of a double-stranded DNA chain and MISTIC protein (Membrane Integrating Sequence for Translation of Integral Membrane Protein Constructs). The reason we chose MISTIC was that it folds into the liposomal membrane easily compared to other membrane proteins1. The double-stranded DNA chain links the MISTIC and the Activator.
3. Connecting the Activator and the Wall
To prevent the Motor-Monomers from associating the Activator before the signal induction, the Activator has to be inactivated by the Wall.
For this, we bind biotin-modified staple strands to the Activator and the Wall. When divalent streptavidin (mutant of wild type tetravalent streptavidin) is added, the Activator and the Wall are combined by the biotin-streptavidin binding.
4. Separating the Wall from the Activator
To activate the polymerization of Motor-Monomers, the Wall has to separate from the Activator upon recognition of the outside signal. Our ultimate goal is to start PoLICe patrolling when it recognizes the cancer markers. However, it is complicated and difficult to use a native cancer marker as a signal. We therefore used a simplified model signal: the restriction enzyme.
When the restriction enzyme is added, it cuts the recognition site on the DNA staple strand that connects the Wall and the Anchor. The Wall consequently separates from the Anchor and the Motor-Monomers are able to associate with the Activator.
5. Releasing the Activator in the liposome
For efficient interactions between the Activator and the Motor-Monomers, the Activator has to be released inside the liposome because the Motor-Monomers are less likely to access the membrane anchored Activator because of steric hindrance.
To release the Activator, we put another restriction enzyme in the liposome. The restriction enzyme cuts the double-stranded DNA chain that links the Activator and the Anchor, and the Activator is therefore released.
The Motor-Monomers are developed to start polymerization to form the Motor-Polymer when the Receptor recognizes the outside signal, and consequently deform the liposome.
For this, we design a Motor-Monomer with both biotin and streptavidin; biotin at the top part and streptavidin at the middle part. The streptavidin is inactivated and unable to bind biotin before being activated by the Activator. The. The Activator of the Receptor restores the binding capacity and the Motor-Monomers therefore start polymerizing after activation by the Activator.
To develop the Motor-Monomer, the following four requirements have to be considered:
The Motor-Monomer is constructed using DNA origami building block of six honeycomb structures. It is shaped like fingers pointing, and two honeycomb structures are longer than the others for 65 nm. There is a hole in the middle part of the Motor-Monomer to embed streptavidin. The length of each part is shown in the picture. The ends of the honeycomb structures are rough to prevent aggregation.
The ends of the honeycomb structures are rough to prevent from cohesion.
The details of our strategy to fulfill the requirements are described below.
1. Encapsulating the Motor-Monomers into the liposome
If the Motor-Monomers and the Activator are put into the liposome at the same time, the Motor-Monomers are activated before the signal induction. We therefore planned to connect the Activator and the Wall in the liposome first, and then fuse that liposome to another liposome containing the Motor-Monomers.
2. Deactivating the polymerizing activity of the Motor-Monomer
To prevent the Motor-Monomers from polymerizing in inactivated state, despite allowing polymerization in activated state, we used the following approach.
We used the difference of the binding strength between the biotin and desthiobiotin to streptavidin. Desthiobiotin is a biotin analogue that binds less tightly to biotin-binding proteins and is easily displaced by biotin*.
For the inactivation process, two complementally oligonucleotides having biotin or desthiobiotin were hybridized and the resulting duplex were bound with streptavidin. Supported by biotin strand, the desthiobiotin are stably tethered to the streptavidin, giving resistance to biotin displacement.
Finally, this inactivated streptavidins were connected to the Motor-Monomers preventing the polymerization before activation. We used alkyne-azide huisgen cycloaddition (known as "click reaction") for connection.
3. Restoring the polymerizing activity of streptavidin with the Activator
As mentioned above, desthiobiotin-streptavidin binding is supported by stable biotin-avidin binding through the hybridization of DNA duplex. We supposed that by cutting the DNA duplex, desthiobiotin loses its supporter and is easily dissociated from the streptavidn, allowing the binding of biotin on the different Motor-Monomer.
HindⅢ, a restriction enzyme cleaves the specific DNA sequence, was chosen as the Activator to cut the DNA duplex. After cleavage by HindⅢ, the length of the remaining sequence is short, therefore dehybridization of the duplex are supposed to be occurred easily.
4. Forming the Motor-polymer enough to deform the liposome
As described in Structure, the Motor-Monomer has the pointing-finger shape and the hole to embed streptavidin. These features enable the Motor-Monomers to connect with others like Lego blocks, which makes the Motor-Polymer stiffer. There are two types of structures the Motor-Polymer would take as shown in fig.9; both are supposed to be stiff enough.
Moreover, two single-stranded DNA “adhesives” complementary to each other are bound to different parts on the Motor-Monomer (showed as yellow lines in fig.10). Those adhesives form duplexes with the adhesives on other Motor-Monomers, which strengthen the connection of the Monomers.
1.M. Douglas et al, “A logic-gated nanorobot for targeted transport of molecular payloads”, Science, 2012 Feb 17; 335(6070): 831-4.
2.Science. 2014;344(6179):65-9.Polyhedra self-assembled from DNA tripods and characterized with 3D DNA-PAINT.Iinuma R, Ke Y, Jungmann R, Schlichthaerle T, Woehrstein JB, Yin P.