Biomod/2012/TU Dresden/Nanosaurs/Project/Vesicles

From OpenWetWare

< Biomod | 2012 | TU Dresden/Nanosaurs | Project
Revision as of 13:39, 27 October 2012 by Gilbert S. Canon (Talk | contribs)
(diff) ←Older revision | Current revision (diff) | Newer revision→ (diff)
Jump to: navigation, search

Overview

In the very early stages of our project, we decided that our central “tethering species” would be Giant Unilamellar Vesicles (GUVs). We would then attach several DNA origami structures to GUVs via cholesterol modified DNA oligonucleotides. The structure encloses oligonucleotic “catcher strands” and is initially locked by means of an aptamer lock. When a ligand specific to the aptamer is introduced into the system the origami structure would open to reveal these catcher strands. Target species in the solution, which have “receiver strands” complementary to the catcher strands, can then get tethered to the GUVs when the catcher and receiver strands hybridize. We decided that our ideal system would contain Large Unilamellar Vesicles (LUVs) as the “tethered target species”.

Introduction

We decided to start with a simple system consisting of single stranded DNA oligonucleotides on both the “tethering” and “target” species until the DNA origami structures were fabricated .Based on previous research work (Beales P A, Vanderlick T K (2007), Specific Binding of Different Vesicle Populations by the Hybridization of Membrane-Anchored DNA. J Phys Chem A 111, 12372-12380), we assumed that the optimal number of anchored oligonucleotide strands per lipid molecule in the vesicles’ membrane was of, 5* 10-3 for GUVs and 4*10-4 for LUVs.

Materials

The giant unilamellar vesicles (GUVs) were prepared using electroformation (Hyperlink) and the large unilamellar vesicles (LUVs) were prepared using rehydration and extrusion (Hyperlink). The composition of both of the phospholipid vesicles was the same and consists of 1, 2-Dioleoyl-sn-glycero-3-phosphocholine (DOPC). The charged vesicles had in addition varying volume amounts (0%-10%) of 1, 2-dioleoyl-sn-glycero-3-phospho-L-serine (DOPS). For some experiments the lipids were labeled with fluorescent dyes, Fast-DiO with an emission at 488 nm and DiD with an emission at 647 nm. There were two different versions of the SLB buffer (Hyperlink) depending on the experiments. The single stranded oligonucleotides experiments were with SLB; meanwhile for the origami experiments the SLB buffer included magnesium.

Our cholesterol-modified oligonucleotides consisted were three: catcher A (tethering oligonucleotide used only for the experiments with single stranded oligonucleotides (Hyperlink)), anchor-complementary (part of the double-stranded tethering oligonucleotide for the origami structure (Hyperlink)), catcher-complementary (oligonucleotide for the LUVs as target species (Hyperlink)).The only non-cholesterol-modified oligonuclotide used was named as “receiver A” since it consisted in the complementary strand for catcher A. Receiver A was modified depending on the target molecule used for the experiments with single stranded oligonucleotides. The target species were Streptavidin-conjugated Quantum dots 625, Alexa 488, Alexa 488-conjugated Streptavidin and finally LUVs.

The experiments were carried out in multiwell plates, each well having a total volume of 40µl. Before using the imaging wells, we incubated for at least 30 minutes with a solution of bovine serum albumin (BSA) (Hyperlink) which was removed previously to setting the experiments. The imaging was done using Zeiss LSM 780 CC3 (Hyperlink) and the pictures were taken at the equatorial plane of the vesicles.

Experimental procedure

  1. GUVs electroformation
  2. Well pasivation (Hyperlink)
  3. GUVs observation under the light microscope (checking the stability)
  4. Target species preparation:
    1. LUVs formation and calibration (Hyperlink).
    2. Streptavidin-biotin interaction: the biotinylated receiver A oligonuclotide were incubated for 10 min with Quantum dot 625-Streptavidin or Alexa 488-Streptavidin (Hyperlink).
  5. Anchoring of cholesterol-modified DNA oligonucleotides:
    1. To GUVs and LUVs: the vesicles were incubated at room temperature for a period of two hours with the corresponding cholesterol oligos. At the end of this process, most strands were anchored to the lipid vesicles
    2. NOTE: for the experiments with origami previous to the anchoring, the cholesterol-modified anchor-complementary oligonucleotides were incubated for 30 min to hybridize with the corresponding origami structure.
  6. Hybridization: the target species were mixed with the GUVs and incubated overnight.
  7. The well was then imaged using Zeiss LSM 780 CC3 inverse confocal microscope.

Experiments with single stranded oligonucleotides

Our tethering species were always GUVs. Since the previous data on the optimal concentration of the oligos on the vesicles were available only for homogenous systems (consisting either GUVs or LUVs), we started with simple tethered target species (fluorophores) to find the optimal concentration of the components and moved on to our final target (LUVs). The tethering and target systems used in this first stage were the catcher A strands and receiver A strands respectively. The molar ratio between the tethering and target oligos was always kept as 1:1.

Alexa labeled DNA oligonucleoides as target species

In the first set of experiments, receiver A strands labeled with Alexa 488 were used as the target species. The protocol for the labeling procedure can be found here (Hyperlink). In the control experiments, no catcher A strands were used. On imaging along the focal plane, clear distinguishable fluorescent rings were observed around the GUVs. Such rings were not observed in the control wells. This clearly indicated that the receiver A strands were hybridizing with the catcher A strands, resulting in such rings.

Control set up
GUV with Alexa 488 labeled receiver A
Transmitted light image of GUV
GUV w/o catcher A
Transmited light image of GUV

Alexa 488 labeled Streptavidin molecules as target species

Subsequently, Streptavidin labeled with Alexa-488 was used as the target species. In the control experiments, no catcher A strands were added. Fluorescent rings were also present around the GUVs.. The control wells didn’t present such rings. This confirmed that it was possible to hybridize more than just oligonucleotides.

Control set up
GUV with Streptavidin labeled with Alexa-488 receiver A
Transmitted light image of GUV
GUV w/o catcher A
Transmited light image of GUV

Quantum dots 625-Streptavidin as target species

Then we decided to use Quantum dots (QD) since they not only have a high quantum yield but also do not bleach and could provide better quality of images. The QD-625 has an emission maximum at 625nm. It has a size of around 25nm. Biotinylated receiver A was first hybridized with QD-Strep, QD-Strep-receiver A, and then later this complex was added to the GUVs bearing chol-catcher A oligos. The controls did not contain the chol-catcher A oligos on the GUVs. The molar ratio of chol-catcher A : Biotinylated-receiver A : QD-Strep = 1:1:0.5 was used. As opposed to the previous results, a fluoresdent ring around the vesicles was not observed. Even more puzzling was the fact that we could not see the QD in solution.

Control set up
GUVs with QD-Strep-receiver A
Transmitted light image of GUV

Gel Analysis

After an unsuccessful approach of visualizing the quantum dots, we decided to check the efficiency of oligo-oligo hybridization and binding to streptavidin coated quantum dots on PAGE. The entire set of the experiments below was performed in 12% PA gel (protocol 1-hyperlink).

Biotinylation and Streptavidin binding

First we checked an efficiency of biotinylation of oligos complementary to cholesterol-coupled oligos. For the gel experiments, oligos without cholesterol were used. There was a shift in the bands between the control oligos and the biotinylated oligo. Also the efficient binding of biotinylated oligo to streptavidin (in lane 4) was observed.

Optimal QD-Oligo ratio

In order to optimize the ratio of quantum dots to the oligonucleotides, different ratios of quantum dots were applied to constant amount of oligonucleotides (70ng). The optimal molar ratio of oligonucleotides to quantum dots was found to be 2:1 since the amount of oligonucleotides not bound to quantum dots was less and also to have a high probability that a single quantum dot is bound by just two oligonucleotides. The other lanes have more free oligonucleotides and therefore would lead to more unspecific binding in solution.

Oligo and Quantum dot hybridization

Finally, the complete system used for the experiments (in 3.1.) (hyperlink) was checked on the gel. The results show that catcher A hybridizes with receiver A with high efficiency. However when QD was added, efficiency drops down significantly.

Spectral analysis of Quantum dots

We were facing some difficulties in observing the quantum dots both in solution and on the lipid membranes. A poor signal was observed even at high laser powers (70%) in confocal microscopy (?). The PA gels proved that there was no problem with the hybridization of quantum dots to biotinylated oligos. Therefore, pure quantum dot samples of different concentrations was prepared and directly observed on cover slips and bright quantum dots at relatively low laser power could be observed. We then obtained the spectra of these quantum dots by doing a fluorescence emission scan (excited at 458nm). A peak signal was observed at 615-625nm which is consistent to the quantum dot manufacturer specifications that we used. When the same analysis was done with our vesicles containing samples mentioned before, we found that the spectra was not the same as of the quantum dots due to the background fluorescence of the contaminated lipids at high laser powers. This prompted us to increase the concentration of the quantum dots from 0.1 nM to 10 nM. Subsequently, the concentration of the oligos was also increased 50-fold.

Caption
False colour
Caption
Bright field
Caption
Maximum emission at 625nm of 10nM Quantum dot samples on the cover slip
Caption
Background signal of the GUVs
Caption
Bright field

Quantum dot 625-Streptavidin as target species (Higher concentration)

With the concentration mentioned before, bright fluorescent rings were observed around the GUVs and none in the controls. Thus, we were able to target a large species like quantum dots on the vesicles.

Caption
Control set up
Caption
Fluorescent of QD with 50x concentration
Caption
Transmited light image of GUV
Caption
GUV w/o catcher A
Caption
Transmited light image of GUV