Folding of an origami is a critical step. Besides the question whether a designed structure folds at all, the yield and the fraction of misfolded structures are also important. The yield determines which concentrations could be applied for further experiments. If significant amounts of misfoldings occur, thorough purification is needed, which in turn corresponds to reduced yield due to purification losses. For these reasons, it was not only aimed to design a structure which can be expected to fold easily, but some effort was also dedicated to testing different thermal folding ramps and check the efficiency of folding.
For more detailed information, please consult the instruction in the labbook.
Several hundreds of staples have to be mixed for a folding batch. Those staples, which are frequently used together for different variants of the structure, were mixed in prestocks. Later on, for small modifications of the structure, only the appropriate prestock needs to be remixed, while all other parts could be used again. Prestocks may vary in the number of included staples, but in one prestock, all staples are equally concentrated (usually 10µl 100µM staple are mixed).
For every structure variant, one working stock is mixed. Here, all needed prestocks are mixed in such a way, that all staples finally have the same concentration, 500nM.
The folding batch contains FOB20, 20nM scaffold and 200nM of each staple. Dye-labeled oligonucleotides are added extra at this step.
Three different ramps were tested, which proceed at different velocities during the critical steps of folding.
This shortest and most simple folding ramp starts at 65°C. Every 15 minutes the temperature is decreased by 1°C, to a final temperature of 25°C. After finishing this ramp, temperature is set to 4°C.
Here again, the starting temperature is 65°C. Every 15 minutes the temperature is decreased by 1°C, to a final temperature of 57°C. The next part of the ramp, down to 26°C, proceeds with a cooling velocitiy of 1°C per 1:30 hours. After finishing this ramp, temperature is set to 4°C.
For this longest ramp, again the first part is cooling down from 65°C to 57°C by -1°C per 15 minutes. Then, the batch is cooled down by 1°C per 3 hours. Finally, temperature is set to 4°C.
Validation of Folding Efficiency
Success of folding was evaluated with agarose gel electrophoresis. Thereby, it could be investigated, whether additional bands besides the main structure do occur. Subsequently, these bands could be analyzed at the TEM.
The usual procedure for purification of folded origamis uses agarose gel electrophoresis. Thereby the main product is separated from single staples and some misfolded structures, e.g. stably linked dimers. While electrophoresis is a helpful tool for analysis of the folding products, for preparative means it has the disadvantages of low yields (estimated to be roughly 10% of the loaded structure) and, more importantly, the samples are contaminated with ethidium bromide. We investigated our origami with gel electrophoresis and found only one weak band in addition to the correctly folded structure. With such few misfolding events, the aim of purification was only to remove unused staples. This could be reached by using size exclusion filters with 100kDa molecular weight cutoff. This new procedure saves time (ca. 1/2 hour as compared to several hours), is less expensive than resolubilization out of agarose gels, loss of product is significantly reduced, and the origamis are not contaminated with intercalators.
For detailed information on the procedure, please read the instruction in our labbook. Results of the analysis of filtered structures with agarose gel electrophoresis can be found here.
For fluorecence analysis and FRET measurements, the dyes need to be attached to the origamis in a defined manner. Therefore, dyes were bound to the 3' ends of selected staples. Dye-labeled oligonucleotides are commercially available, but quite expensive. On the one hand, we need only small amounts of labeled staples, on the other hand a variety of ten different oligonucleotides was used for different experiments. Under these circumstances, it was preferable to buy the both dyes each bound to one ddNTP and then couple them to the oligonucleotides of interest. This could be achieved using terminal transferase, which is also commercially available. This enzyme binds (modified) dNTPs or ddNTPS to free 3' ends.
After labeling staples, the yield was determined by measuring absorption at 260nm. In this procedure, at first an absorption coefficient for the unlabeled oligonucleotide is determined, since these coefficients may display huge deviations from those obtained from online tools or suppliers. The coefficient of the one ddNTP with attached dye was determined separately. By adding these values, ε260(labeled oligonucleotide) could be estimated and used for concentration measurement of the labeled staples.
Fig. 1: The BioNano TEM facility.
Transmission electron microscopy (TEM) is an imaging technique to observe thin specimen. An electron beam interacts with the specimen while passing and is detected by a sensor, in this case a CCD camera.
DNA consists of atoms with small atomic numbers. This leads to only weak contrast. For fixation on the samples to enhance the contrast, uranyl formiate is applied as negative stain.
As DNA is light material, the samples were stained with uranyl acetate to enhance contrast. Electrons can pass through the origami-structure and are absorbed or scattered by the stain. Read more about preparing samples for transmission electron microscope.
The images taken for this project were made at the BioNano TEM facility (CM 100, Philips, Amsterdam, NL) at a high voltage of 100.0 kV and a nominal magnification of 28500. Images were 4x averaged and integrated to reduce noise.
Besides confirming the proper folding of our structure, the main purpose of the TEM measurements was to provide data on twists and lengths of the structures, dependent on concentrations of the DNA binders. These angle and length measurements were accomplished with the the free tool imagej.
Principle of FRET
Förster resonance energy transfer (FRET) describes the energy transfer through nonradiative dipole–dipole interaction between two chromophores. If the emission spectrum of one chromophore overlaps with the excitation spectrum of the other chromophore and both chromophores are in proximity (aprx. 10nm) to each other, an excited donor chromophore can transfer energy to an acceptor chromophore resulting in fluorescence with the emission wavelength of the acceptor chromophore.
The FRET efficiency (E) is the fraction of energy transfer occurring per donor excitation event:
where kET is the rate of energy transfer, kf the radiative decay rate and the ki are the rate constants of any other de-excitation pathway.
The FRET efficiency depends on many parameters that can be grouped as follows:
- The distance and orientation between the donor and the acceptor
- The spectral overlap of the donor emission spectrum and the acceptor absorption spectrum.
- The relative orientation of the donor emission dipole moment and the acceptor absorption dipole moment.
E depends on the donor-to-acceptor separation distance r with an inverse 6th power law due to the dipole-dipole coupling mechanism:
with R0 being the Förster distance of this pair of donor and acceptor, i.e. the distance at which the energy transfer efficiency is 50%.
The FRET efficiency relates to the fluorescence lifetime of the donor molecule as follows:
where τ'D and τD are the donor fluorescence lifetimes in the presence and absence of an acceptor, respectively, or as
where F 'D and FD are the donor fluorescence intensities with and without an acceptor, respectively.
Some test bulk measurements were performed both with a photospectrometer and with a RT-PCR. Therefore we used the MH_255/256 18bp double helix with FRET labels at both ends. Thus the distance of the labels should match the Förster distance. Excitation and emission spectra at the photospectrometer were used to determine optimal wavelengths for FRET measurements. At the more sensitive RT-PCR, Atto 550 was excited at 545 nm and fluorescence detected at 568 nm. Atto 647N was excited at 635 nm and detected 665 nm, thus matching the conditions of the RT PCR's optical filters.
Single Molecule Measurements
For single molecule fluorescence microscopy, the origamis need to be immobilized. We need similar orientations on the surface. The structures should stand on the base of the "U", since flat lying origamis might display limited flexibility. For this aim, seven additional adapter staples were added at the bottom of the structures. 21 bp of these staples reach beyond the base. When preparing samples for the fluorescence microscope, the slide is incubated with neutravidin and then with biotinylated staples. These are complementary to the free base pairs of the adapters, so hybridization can take place and the origamis are orientated in the wanted direction.
Fluorescence Microscope Setup
We used a homemade fluorescence microscope, that could be used both in an epifluorescence mode and a total internal reflection fluorescence (TIRF) mode. For the attempted bulk measurements, epifluorescence was used, for all single molecule measurements we used the TIRF mode. All FRET experiments were performed using alternate laser excitation (ALEX), i.e. the green laser and the red laser excite alternating. When the red laser is active, only the acceptors can be seen. So it can be checked, when the acceptors bleach. The green laser excites the donor, if an acceptor is in proximity, it can also be seen due to FRET.
- green laser: excitation at 532 nm
- red laser: excitation at 640 nm
- green camera filter: 590 +/- 35 nm
- red camera filter: 685 +/- 20 nm
Framerates and exposure times were changed sometimes, so the details here are recorded in the labbook.
Fig. 2: used homemade fluorescence microscope at Dietz Lab.
Evaluation of Data
In order to calculate the extent of structure deformation represented by the chance in distance of the two flouropores we evaluated the acquired data in two different ways. On the one side we determined the FRET efficiency by analyzing a recorded video on the other we simply measured the distance of two spots by comparing the two pictured we received by alternating excitation.
For analyzing the acquired FRET movies and extract accurate FRET efficiencies, a homemade MATLAB software was used.
The software automatically picks spots, based on user chosen thresholds and tracks them throughout the whole movie. It integrates a circular region of 5 pixel per spot in each frame and generates intensity traces. A mapping between the donor and acceptor detection channel was done to colocalize corresponding tracks.
Traces showing a donor as well as an acceptor signal were automatically extracted and afterwards manually selected for typical bleaching characteristics; I.e. bleaching of the acceptor is anticorrelated to increase in donor intensity and bleaching of the donor at last. This is necessary to determine background intensities, calculate the gamma factor and therefore obtain accurate FRET efficiencies. Hereby, intensity levels were extracted by fitting (multi-)step funtions.
For measuring the distance the MATLAB software was modified. After picking the spots in each picture they were aligned and for each spot the intensity was fitted with gaussian curve. Afterwards the distance was calculated between the peaks of two matching spots.
Fig. 3: evaluation of data received from ALEX for distance calculation.