Taming polymorphism of tubule self-assembly using templated growth

Abstract

Self-closing assembly is prone to polymorphism due to thermally-excited bending fluctuations, which permit the formation of off-target assemblies at the point of self-closure. One way to overcome this source of polymorphism is to use templated growth, a process in which assembly initiates from a precisely-defined seed rather than by spontaneous nucleation. We explore this approach to quelling polymorphism in the self-closing assembly of cylindrical tubules assembled from DNA-origami subunits with user-specified inter-subunit binding angles and specific interactions. We develop two strategies to create seeds with precisely-defined diameters and helicity: 1) using multicomponent assembly; and 2) purifying a specific seed-type from a polymorphic mixture using gel electrophoresis and gel extraction. By tuning the seed and monomer concentrations, and adjusting the assembly temperature, we determine the conditions under which tubules grow from the seed while avoiding spontaneous nucleation. We observe that templated tubules tend to follow the guidance of the seed, thereby increasing the selectivity of the target geometry. Also, we find that by tuning the diameter of the seed, one can template the growth of monodisperse tubules over a range of target diameters, even while using a single monomer type with a single preferred local curvature. Our results demonstrate that employing precisely defined seeds to guide assembly can significantly decrease polymorphism in self-closing assembly in a controllable and economical way.

Figures (4)

• Figure 1: Assembling self-closing tubules via templated growth. (A) cryo-EM single-particle reconstruction and schematics of our DNA-origami triangular monomer. Six single-stranded DNA handles extend from specific locations of each face (red circles) and mediate specific interactions via base pairing. (B) Assembly can proceed along two pathways: (1) nucleation and growth; and (2) templated growth from a seed. (C) Specific inter-subunit interactions needed to assemble a tubule and a seed. Lines indicate favorable interactions. Solid lines represent stronger interactions and dashed lines indicated weaker interactions. (D) TEM image of an assembled seed (Fig. S19 shows the widefield of seed assembly). (E) TEM image of a templated tubule. The presence of DNA-conjugated gold nanoparticles (black dots) in the middle of the tubule indicates the location of the seed.
• Figure 2: Templated assembly state diagram. Each data point represents an assembly outcome from experiments. We define the templated tubule state as an assembly with an average tubule length greater than 700 nm in which more than 50% of the tubules grow from a seed (Fig. S6 shows the state diagram for different cutoff lengths). The size of the circle is proportional to the fraction of templated tubules, which is the number of templated tubules divided by the total number of tubules. The micrographs show representative structures for various kinetic regimes of the state diagram.
• Figure 3: Using multispecies tilings to assemble specific seeds. (A) Multispecies tilings can be used to define the seed length and restrict the allowed seed states. A seed can be conceptualized of as a segment of a planar tiling with periodic boundary conditions. We break up the seed into parallelograms with different widths (the purple and red shaded regions); these parallelograms are possible principle unit (PU) cells. Only when the width of the PU cell is a divisor of the circumference is the formation of a seed with that circumference allowed. Once a size of PU cell is chosen, we color in different triangles to identify the number of species required to assemble it hayakawa2024symmetry. (B) Tubule width probability distribution from spontaneously nucleated tubules in the absence of seeds. (C--F) Tubule width probability distributions for assemblies with multispecies seeds. The red points show the distribution of the assembled seeds, and the gray bars represent the probability of the assembled tubules. The inset figures show the associated PU cell: (C) (1,0) PU cell, (D) (3,0) PU cell, (E) (4,0) PU cell, and (F) (7,0) PU cell. All tubule assemblies were grown for two weeks. All templated assemblies have 0.33 nM of seed-monomers, which corresponds to estimated seed concentrations of 2.8 pM (C), 1.64 pM (D) , 0.6 pM (E), and 1.62 pM (F).
• Figure 4: Welding and purifying seeds from a polymorphic mixture (A) UV-weldable monomer handle and leg designs. Shaded blue regions show welding sites. (B) Width distribution of the UV-welded seed purified from the (7,0) band using gel electrophoresis. Insets are the PU cell and the most probability seed geometry. Due to the stronger interactions of the seed monomers, the seed has four particle species to avoid any homophilic interactions. (C) TEM image of a purified (7,0) seed. (D) A templated tubule grown from a purified (7,0) seed. (C) and (D) share the same scale bar. (E) Sketch of how a lattice mismatch between the tubule monomers and the seed monomers could cause polar assembly.