Biomod/2012/TU Dresden/Nanosaurs/Project/DNA origami

From OpenWetWare

(Difference between revisions)
Jump to: navigation, search
Line 84: Line 84:
<div class="clear"></div>
<div class="clear"></div>
-
<div style="margin:auto; width:250px; box-shadow: 0 0 10px #888888; border: 5px solid white;"><a  rel="lightbox[origami]" title="3D image created with Maya" href="http://openwetware.org/images/4/43/BM12_nanosaurs_MayaOrigami.jpg"><img  src="http://openwetware.org/images/a/a6/BM12_nanosaurs_MayaOrigami_s.jpg.jpg"></a></div>
+
<div style="margin:auto; width:250px; box-shadow: 0 0 10px #888888; border: 5px solid white;"><a  rel="lightbox[origami]" title="3D image created with Maya" href="http://openwetware.org/images/4/43/BM12_nanosaurs_MayaOrigami.jpg"><img  src="http://openwetware.org/images/a/a6/BM12_nanosaurs_MayaOrigami_s.jpg"></a></div>
Line 90: Line 90:
<h2>Functional parts</h2>
<h2>Functional parts</h2>
-
<h3>Scaffold</h3>
+
<h4>Scaffold</h4>
-
<p>The scaffold is a 7560 bases long single stranded circular DNA derived from the E.coli virus M13p18. It provides the basis to which the staple oligos can hybridize to form the structure. It winds its way through the whole DNA origami and ends exactly at the point where it started.
+
<p>The scaffold is a 7560 bases long single stranded circular DNA derived from the E.coli virus M13p18. It provides the basis to which the staple oligos can hybridize to form the structure. It winds its way through the whole DNA origami and ends exactly at the point where it started.<br/>It also connects the upper and lower half of the shell via two hinges. These are simply about 12 bases of single stranded DNA, meaning there are no staple strands hybridizing in this region.
</p>
</p>
<div class="img_right img_link">
<div class="img_right img_link">
Line 99: Line 99:
<a  rel="lightbox[origami]" title="Scaffold - Top view" href="http://openwetware.org/images/7/70/BM12_nanosaurs_1_Scaffold_top.jpg"><img src="http://openwetware.org/images/a/a6/BM12_nanosaurs_1_Scaffold_top_s.jpg"></a>
<a  rel="lightbox[origami]" title="Scaffold - Top view" href="http://openwetware.org/images/7/70/BM12_nanosaurs_1_Scaffold_top.jpg"><img src="http://openwetware.org/images/a/a6/BM12_nanosaurs_1_Scaffold_top_s.jpg"></a>
</div>
</div>
-
<p>It also connects the upper and lower half of the shell via two hinges. These are simply about 12 bases of single stranded DNA, meaning there are no staple strands hybridizing in this region.
+
-
</p>
+
<div class="clear"></div>
<div class="clear"></div>

Revision as of 20:43, 27 October 2012

What is DNA origami?

DNA origami is a well-established technique in nanotechnology which involves the folding of DNA to create various 2D or 3D patterns and shapes at the nanoscale. More in detail, it makes use of a long single strand of DNA known as scaffold strand, which acts like a backbone or support for a particular structure to be made. In order to shape the construct in a certain way, several shorter strands of different lengths, called staple strands, are hybridized (bound) to specific parts of the scaffold profiting from the specificity of interactions between complementary base pairs in DNA. The way these binding sites are chosen, determines how the structure is going to fold and in which shape it ends up.

Our DNA origami structure

Requirements

In order to accomplish the purpose of our project, the DNA origami shell must fulfill the following requirements:

  • Comprise a tethering platform for the attachment to lipid bilayers.
  • Include a variable catching platform which allows the structure to bind specific target species.
  • Avoid unspecific binding of non-wanted targets.
  • Provide a trigger mechanism which enables binding upon signal.
  • Contain one or more parts with fluorescent labels for testing and imaging purposes.

Design

DNA origami is a well-established technique in nanotechnology which involves the folding of DNA to create various 2D or 3D patterns and shapes at the nanoscale. More in detail, it makes use of a long single strand of DNA known as scaffold strand, which acts like a backbone or support for a particular structure to be made. In order to shape the construct in a certain way, several shorter strands of different lengths, called staple strands, are hybridized (bound) to specific parts of the scaffold profiting from the specificity of interactions between complementary base pairs in DNA. The way these binding sites are chosen, determines how the structure is going to fold and in which shape it ends up.

Our structure was inspired by the DNA origami Logic Nanorobot by Douglas et al. (Science 335, 831-8XX (2012)). We chose the model proposed in this paper because its shell-like shape provides binding specificity while its lock mechanism offered a triggering platform.
In our design, the height of the structure was lowered to 20nm while increasing its width to 45nm. This is intended to decrease the chances of molecules diffusing into the origami in its closed configuration while making it lay flatter on the vesicle surface to enhance binding. Most notably, the structure was adjusted to have anchor and catcher strands which allow binding to the carrier vesicle and to the target species respectively.
The functional principle of the DNA origami shell is that in its closed state (i.e its lock strands are hybridized to each other), the single stranded catcher oligonucleotides inside are shielded. This means they are not accessible for anything from the outside to hybridize to them. The locks can be triggered to open once a certain protein is around. When the lock strands are not hybridized anymore, the DNA origami shell opens up due to thermal fluctuations. The catcher strands inside are then freely accessible.

Functional parts

Scaffold

The scaffold is a 7560 bases long single stranded circular DNA derived from the E.coli virus M13p18. It provides the basis to which the staple oligos can hybridize to form the structure. It winds its way through the whole DNA origami and ends exactly at the point where it started.
It also connects the upper and lower half of the shell via two hinges. These are simply about 12 bases of single stranded DNA, meaning there are no staple strands hybridizing in this region.

Core

There are 171 core oligos which give the structure its stability, its basic shape and hold all the functional parts together. To avoid that several structures stack together, the turning points of the scaffold are left single stranded for about 10 to 30 nucleotides. This means that in those regions no staple strands hybridize thus comprising highly flexible single strand DNA.

Anchors

The anchor oligomers can hybridize to cholesterol labeled single strands that provide the attachment to the lipid membrane of the giant unilamellar vesicles and therefore connect the DNA origami structure permanently to them. There are 9 anchor strands to increase the probability of binding. They are 35nt long, containing a 5nt spacer at the DNA origami and a 30bp binding site for cholesterol oligomers on the vesicles.

Catchers

The catcher strands are complementary to the cholesterol labeled single strands that are integrated in the lipid membrane of the large unilamellar vesicles. Once the DNA origami shell is open and the catcher strands are freely accessible, the LUVs can bind to the DNA origami due to the hybridization of the single strands. There are 6 catcher strands to increase the probability of binding. They are 35nt long, containing a 5nt spacer at the DNA origami and a 30nt binding site for the single strands on the vesicles.

For our experiments the catcher strands were also labeled, for example with biotin to bind to streptavidin coated quantum dots.

Locks

The DNA origami has two locks, one on each side. Each lock consists out of two oligomers: an aptamer (blue) and its complementary oligo (green). The total length of each lock half extruding from the origami is 44 nt. 24nt of them are complementary to each other. The remaining 20nt that are not complementary, are between origami and the complementary region.

The aptamer is triggered by the protein PDGF. It preferably binds to a specific site on PDGF binding, which can thus open the lock by dissociating it from the complementary oligo (see (see Aptamer).

Locks are only present in closed structures. If a strictly open structure was needed for the experiments the lock staples without any overhang were applied (“locks_nohang”).

Edge staple

The edge staple is a single strand that is on the top side of the structure. It can be labeled with fluorophores such as Alexa 647 to add a fluorescent signal to the structure.

Guide staples

There are two guide staples that should help to close the structure during the assembly. It increases the amount of structures that are closed after the assembly. The structure cannot open while the guide strands are there. Therefore they have an 8 base toehold and if fully complementary oligos are applied later, they would hybridize and the structure would still be closed, but now has the chance to be opened (if the locks are opened).

cadnano

In order to turn the idea of our sketch into an actual 3D DNA nanostructure, we used a software called caDNAno(learn more at http://cadnano.org/); a computational tool for DNA origami design. Using this tool one define the shape of the desired structure within a graphic environment by providing the hybridization sites for the staple strands and the scaffold length as input parameters. The program will then display the necessary sequences of the staple strands which can be ordered from a suitable company.

Assembly

To assemble the desired structure the following things need to be pipetted together:

  • Scaffold
  • Set of staple oligos defining the features the assembled structure should have
  • Folding buffer
  • Water

The correct ratios and the recipe of the folding buffer can be found in the recipe section.
Following a detailed protocol the mixture is heated up to 85°C and then cooled down very slowly using a given temperature ramp. Especially in the area of 55°C the cooling process is extremely slow since most of the assembly process happens in that temperature region. The whole cooling process takes about 15 hours.
After the assembly the structures remain stable at room temperature.

Purification

To have a greater yield of assembled structures, the ratio of staple strands to scaffold strands is 7.5 to 1. To get rid of the leftover single strands after assembly, the samples are typically dialyzed for 1 to 2 hours using a 0.025µm filter.

Results

In order to examine the shape of the structure, the samples were imaged using transmission electron (TEM) and atomic force microscopy (AFM).

TEM

To image the structures via transmission electron microscopy the samples were stained with uranyl acetate (see protocols).

The TEM images demonstrate a successful assembly of both types of structures. In particular they show a significant difference in shape between the open and the closed structures. Open structures were typically twice as long as closed structures (see below). For these images the closed structures were assembled including the guide strands, but also the not guided constructs showed a conformational change with a high percentage of the structures being closed.
Producing negatively stained samples (using short staining times) it was possible to image the closed structures standing upright. The pictures show that the shape of the cross section is rather variable. However, most of the structures show a high degree of integrity, i.e. a closed circumference supporting that the structures are really closed.
Evaluating several individual structures the following average lateral dimensions of both types of DNA origami were obtained:

[nm] closed open
width length full length
# of measurements 26 29 40
result (95% STD) 48,9 ± 5,9 39,6 ± 3,4 71,4 ± 3,8
relative error [%] 12,0 8,6 5,3
Expected 45 40-44 80-88
Possible reason
for deviation
Hinges and edges
floppy single strands

For the closed structure the length, as well as the width, match nicely the expected values. The slightly higher width can be explained by assuming that the structures laying down flat which increases the lateral dimension due to the bending down of the side walls.
The open structure however appeared to be shorter than one would expect if one doubles the length of a closed structure. This can be explained due to the fact that the turning points, as well as the hinges, were left as single strands making them more flexible. Therefore they do not necessarily have to be stretched to their full lengths. In general the open structure shows increased flexibility and degrees of freedom compared to the closed constructs.

AFM

To further proof of the correct assembly, the open and closed structures were sent to the Spanish National Center for Biotechnology in Madrid. There Dr. Fernando Moreno-Herrero and Maria Eugenia Fuentes obtained a series of magnificent AFM images.The following pictures show the different samples in an overview (left) as well as an enlarged view of a single structure (right).

The open structures appear very homogeneous in shape, whereas a rather large degree of heterogeneity was found in the AFM images for the closed structures. A possible explanation for the less defined shape of the closed structures could be that those samples have been purified via Freeze ‘N Squeeze DNA Gel Extraction whereas the open samples have just been dialyzed. However, the Freeze’N squeeze purification gives less background, which means the sample is purer. Since the structures appear to be very fragile, the dialysis is a more suitable purification method to leave the structures intact.
Further measurements on seemingly intact closed structures provided three major classes of different shapes. These can be interpreted by the following model developed by Dr. Fernando Moreno-Herrero and Maria Eugenia Fuentes:

The evaluations of the lateral dimensions of the origami structures in the AFM images are depicted in the table below:

[nm] width length height
peaks
height
valleys
open one half 54,7 ± 3,9 41,5 ± 3,2 3,8 ± 0,4 1,2 ± 0,3
closed 2 blobs 79,2 ± 3,4 57,3 ± 5,3 8,9 ± 1,7 ---
3 blobs 85,8 ± 6,7 69,2 ± 5,6 8,5 ± 1,8 ---
4 blobs 90,5 ± 1,2 59,9 ± 6,0 7,3 ± 1,6 ---
Expected 45 40-44 10 / 20

The length of the open structure matches very well with the expectations. The width is slightly too large and the height is too low. This can be explained by the fact that the fragile structure preferably lays down flat on the surface and also gets pushed down by the AFM tip.
The increase of width and length for the closed structure can be explained by an increased tip convolution due to the increased height of the structure. However the height matches very well, since it is twice the height of the open structure indicating that the desired conformational change has been successfully achieved.
The various heights of the close structure also go along very well with the model of the different positions of the hexagonal DNA multifilaments.
In total the AFM and the TEM images confirm a successful DNA origami assembly and the expected change in conformation between open and closed structures for the majority of the objects. Also the dimensions are well in agreement with the expectations taking into account some explainable deviations due to the flexibility of the structure and the limitations of the method that was applied.

Gel shift assays

In order to test the specific binding of cargo to our structures and calibrate the sample conditions, several gel shift assays were performed. The most relevant ones are highlighted in this section.
For internal controls two different schemes for cargo attachment were followed: Loading the cargo based on streptavidin-biotin interaction and employing DNA strand hybridization. In these experiments we used streptavidin coated quantum dots which can be attached to the origami in two ways:

  • Directly binding to internal 5’ biotinylated strands.
  • Binding of the quantum dots to 3’ biotinylated oligonucleotides which can then hybridize to the internal catcher strands of the origami.

To make the gels easy to understand, we use the following conventions for defining which components were loaded in each lane:

Buffer calibration

In order to enhance the quality of the assemblies, the effect of the folding buffer on the yield and structural integrity of the origami was examined. Four different buffers with various MgCl2 concentrations (8mM, 10mM, 12mM, 14mM) were used for assembling open and closed structures, as can be seen in Fig.1. From the pictures obtained, one can see that by increasing the MgCl2 concentration, the band for the closed structure blurs and shifts up. This indicates that the structure becomes less homogeneous and possibly the structures are also more prone to dimerization. Based on this, we took 8mM as our standard buffer for further experiments.

Structure overview

At first, the quality of the basic open and closed assemblies was tested. As shown in lanes 2 and 3, both assembled structures have a different structure and therefore run differently on the gel compared to the scaffold. Moreover, it can be seen that in lane number 2 there is a second band above the expected band for the structure. This likely shows that the open structures tend to aggregate more than the closed structures, which can be attributed to two main factors; MgCl2 induced stacking interactions and hybridization between the free locks of adjacent structures.

Quantum dot binding

After confirming the assembly quality of our structures, cargo attachment tests were performed. In particular, we employed attachment through hybridization (Fig.3). Quantum dot cargos that carried oligomers complementary to the catcher strands of the origamis were added to the open and closed structures. Subsequently binding preferences were determined.

From the results obtained (Fig.3) one can identify a clear gel shift due to quantum dots binding in lanes 2 and 5. However, there’s not a noticeable difference between the open and closed configurations as the ratio of bound vs. unbound structures cannot be determined straight forward. In order to have a better idea about binding preference and to discard problems with the structure, a further experiment involving the catchers of the system was proposed.

Catcher influence on binding

The previous results showed that there was still considerable binding to the closed structures. This might be due to the catcher strands sticking out on the wrong side of the structure. Therefore, in addition to the construct with all 6 catchers to two other versions containing only one catcher were tested for cargo binding. One of them contained a single 5’ biotinylated oligo and the other contained only a single catcher for hybridization mediated binding.

The results shown in fig.4 suggest a preference for the binding to the open structures compared to the closed structures when only a single catcher strand was present. If six catchers are used this difference was greatly reduced.

To further support this, we quantified the binding preference of the structures from the gel image based on the relative intensities of the bands which showed a shift due to quantum dot binding and of the bands that contained the origami only. The obtained results are shown in the table below.

Lane Construct Shifted
band
Construct
band
Ratio
shifted/construct
QD affinity ratio
open/closed
4 Open 1C 1942 5580 0.35 1.61
11 Closed 1C 672 3111 0.22
6 Open 1C5' 5216 4025 1.30 1.70
13 Closed 1C5' 2617 3437 0.76
8 Open 6C 2937 2944 1.00 1.24
15 Closed 6C 4106 5123 0.80

From these data, it can be seen that:

  • The quantum dots have a binding preference for the open structures over the closed ones.
  • This preference decreases if the number of catchers is increased.
  • ? The attachment performance through hybridization or biotin-streptavidin interaction is comparable.

However, in all cases closed structures still bind the cargos to a significant extend. The reason for this unexpected behavior still need to be explored. It may be that still to many misfolded closed origami structures are formed during assembly. This could be improved by a more careful adjustment of the annealing conditions.