"

From 2012.igem.org

RNA folding and 3D Structures

 

Introduction

 

For most small RNA molecules, secondary structure is enough for predicting their possible functions since their size limits their ability to fold into complex structures. However, our riboscaffold is a large one of 100~160 bp and has several arms. As in addition, our design of kissing loop is on the level of tertiary structure. So it is helpful if we can simulate RNA folding in silico.

 

Methods

 

NAST/C2A [10] is a set of Python tools that can generate full-atomic 3D RNA structures from secondary structure information. NAST generates coarse grained 3D structures from secondary structure information, and C2A adds the full-atomic details to these coarse-grained models. PyOpenMM [11] was required, VMD [12] and PyMOL [13] was used to view the results and trajectories.

 

Beginning with sequences and secondary structures, NAST generates 3D structures in two different ways. The first is to start an MD simulation from an unfolded circle state. The second is a more sophisticate approach that starts at a random position in the sequence, and adds residues to each end at a random plausible position. Details about the sequences and design strategy can be found here.

 

Results

 

The following two movies about the folding process of D0 explain the two different ways of 3D structure prediction provided by NAST.

 

Movie1. Start an MD simulation from an unfolded circle state.
Movie2. Starts at a random position in the sequence, and adds residues to each end at a random plausible position.

 

According to the document of NAST, the second method (random unfold starting structure) is more sophisticate, so in the following movies, we use this method to predict the folding of the riboscaffolds we designed.

 

Figure 1. Clover1 3D structure
Movie3. Clover1 Folding.

 

Figure 2. Colver2-1 structure
Movie4. One possible folding process of Clover2.

 

In this conformation, the interaction between theophylline aptamer and MS2 aptamer is not obvious, probably because they extend from the same loop, which is too small to be flexible.

 

Figure 3. Colver2-2 structure.
Movie5. Another possible folding process of Clover2.

 

In this conformation, the interaction between theophylline aptamer can be seen during the folding process, but as the MS2 aptamer extends, the internal rigidity pretends the torsion in the structure.

 

Figure 4. Colver3 structure.
Movie6. Clover3 folding.

 

In this movie, we can see the interaction between theophylline aptamer and MS2 aptamer is more obvious, which begins from the formation of MS2 and remains in the “mature” Clover3.

 

Figure 5. Clover3 mutant.
Movie7. Clover3 mutant folding

 

To demonstrate that the interaction in Clover3 is caused by our design, we simulate the 3D structure of a mutant of Clover 3, in which the theophylline aptamer remains unmutant so it cannot base pair with the sequence of MS2.